You are on page 1of 336

Informing the Nation: Federal Information

Dissemination in an Electronic Age

October 1988

NTIS order #PB89-114243


GPO stock #052-003-01130-1
Recommended Citation:
U.S. Congress, Office of Technology Assessment, Informing the Nation: Federal In-
formation Dissemination in an Electronic Age, OTA-C IT-396 (Washington, DC: U.S.
Government Printing Office, October 1988).

Library of Congress Catalog Card Number 88-600567

For sale by the Superintendent of Documents


U.S. Government Printing Office, Washington, DC 20402-9325
(order form can be found in the back of this report)
Foreword
Federal information is essential to public understanding of many issues facing Congress
and the Nation, and is used by all sectors of society. Technological advances are opening up
many new and potentially cost-effective ways to collect, manage, and disseminate this information.
Although traditional ink-on-paper publications will continue to meet important needs for the
foreseeable future, many types of Federal information—such as statistical, reference, and scien-
tific and technical-are well suited to electronic storage and dissemination. For example, an entire
year’s worth of the Congressional Record or several Bureau of the Census statistical series can
be placed on one compact optical disk that can be easily read with a low-cost reader and basic
microcomputer. Press releases, weather and crop bulletins, and economic or trade indices can
be disseminated immediately via electronic bulletin boards or online information systems.
This report addresses the opportunities to improve the dissemination of Federal information.
It also highlights two major problems: maintaining equity in public access to Federal information
in electronic formats, and defining the respective roles of Federal agencies and the private sector
in the electronic dissemination process. The report focuses on current and future roles of the
U.S. Government Printing Office (GPO) and Superintendent of Documents, the Depository Li-
brary Program (administered by GPO), and the National Technical Information Service (NTIS).
In addition, this report examines electronic dissemination of congressional information, the
Freedom of Information Act in an electronic environment, and electronic dissemination of gov-
ernment information to the press.
In conducting this assessment, OTA drew on expertise and perspectives from numerous
sources in and outside of the government. OTA received special assistance from the General
Accounting Office (GAO) for the surveys of Federal information dissemination practices and
Federal information users, from GPO with respect to Federal printing and related dissemination
activities, and from NTIS with regard to dissemination of scientific and technical information.
OTA appreciates the participation of the advisory panelists, contractors, working group partici-
pants, Federal agency officials and Federal information users who responded to the GAO surveys,
and members of the library, academic, business, labor, consumer, and Federal agency communities,
among others, who helped bring this report to fruition.
The report responds to an initial request from the Joint Committee on Printing and subsequent
expressions of interest from the Subcommittee on Government Information, Justice, and Agri-
culture of the House Committee on Government Operations, the House Committee on Science,
Space, and Technology, the Committee on House Administration, and the Subcommittee on
Legislative of the House Committee on Appropriations.
The report is solely the responsibility of OTA, not of those who assisted us in the assess-
ment or of the congressional committees who requested or endorsed the undertaking of the study.

///.,,
Informing the Nation Advisory Panel
Marvin Sirbu, Chairman
Associate Professor, Carnegie-Mellon University
Ben Bagdikian Paul P. Massa
Dean and Chairman President & Chief Executive Officer
Graduate School of Journalism Congressional Information Service, Inc.
University of California at Berkeley James A. Nelson
Nolan Bowie State Librarian and Commissioner
Assistant Professor of Communications Kentucky Department for Libraries and
Temple University Archives
Tom Davies Ron Plesser, Esq.
General Manager Nash, Railsback, and Plesser
SCT Corp. Howard Resnikoff
Miriam Drake President
Director of Libraries Aware, Inc.
Georgia Institute of Technology Katherine D. Seelman
Lee Felsenstein Director of Communications
President Massachusetts Commission for the Deaf
Golemics, Inc. and Hard of Hearing
James K. Galbraith Fran Spigai
Associate Professor of Economics President
LBJ School of Public Affairs Database Services, Inc.
Mary Gardiner-Jones Susan Tolchin
President Professor of Public Administration
Consumer Interest Research Institute The George Washington University
Robert Gibson, Jr. Congressional Agency Participants
Head Librarian (retired) Robert L. Chartrand
General Motors Technical Center Senior Specialist in Information Policy and
John A, Jenkins Technology (retired)
General Manager Congressional Research Service
BNA On-Line Vincent DeSanti
Earl C. Joseph Group Director
President General Accounting Office
Anticipatory Sciences Inc. Harold C, Relyea
Myer Kutz Specialist in American National
Executive Publisher Government
John Wiley and Sons Congressional Research Service

NOTE: OTA gratefully acknowledges the members of this advisory panel for their valuable assistance and thoughtful
advice. The panel does not, however, necessarily approve, disapprove, or endorse this report. OTA assumes
full responsibility for the report and the accuracy of its contents.

iv
.——

OTA INFORMING THE NATION ASSESSMENT STAFF*


John Andelin, Assistant Director, OTA
Science, Information, and Natural Resources Division

Fred W. Weingarten, Program Manager


Commumication and Information Technologies

Project Staff**
Fred B. Wood, Project Director
Prudence S. Adler, Assistant Project Director
Jamie A. Grodsky, Analyst
Carol S. Nezzo, Analyst

Other OTA Staff Contributors


Jean Smith
Darlene Wong

Administrative Staff
Elizabeth Emanuel, Administrative Assistant
Rebecca Battle, Secretary
Karolyn Swauger, Secretary

Publishing Staff
Kathie S. Boss, Publishing Officer
Chip Moore, Publishing Assistant
Debra Datcher Cheryl Davis
Dorinda Edmondson Steve Kettler
Ted Williams Susan Zimmerman

*see appendix A for acknowledgments of GPO, GAO, and NTIS staff, agency officials, workshop p~ticipants. reviewers, and
others who participated in the study.
**See appendix B for OTA project staff responsibilities.

v
Contractors
Brenda Dervin Charles McClure
Ohio State University Syracuse University
Stephen Frantzich Judith E. Myers
Congressional Data Associates University of Houston Library
Henry Freedman Thomas P. Riley
Consultant Riley Information Services
Gregory Giebel Frank J. Romano
University of the District of Columbia Consultant
Mark P. Haselkorn, Philip L. Bereano, and Barry M. Schaeffer
Barbara Lewton Consultant
University of Washington Jacob W. Ulvila
Peter Hernon Decision Science Consortium, Inc.
Simmons College Carol Watts, Sarah Kadec, and Dorothy Weed
Earl Joseph Washington Information Network
Anticipatory Sciences, Inc.
Contents

Chapter Page
CHAPTER 1: SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Opportunities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Problems and Challenges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Government Printing Office . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
National Technical Information Service. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
National Technical Information Service/Superintendent of Documents. . . . . . . . 14
Depository Libraries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Technical/Management Improvements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Statutory/Oversight Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Legislative Branch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
About This Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
CHAPTER 2: OVERVIEW OF FEDERAL INFORMATION
DISSEMINATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Size and Scope of Federal Information Dissemination Enterprise . . . . . . . . . . . . 28
Technological Initiatives by Federal Agencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Institutional Infrastructure for Federal Information Dissemination . . . . . . . . . 36
CHAPTER 3: KEY TECHNOLOGY TRENDS RELEVANT TO FEDERAL
INFORMATION DISSEMINATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Introduction and Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Information Systems Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
The Microcomputer Revolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
The Continuing Role of Paper and Microform. . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Electronic Publishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Desktop Publishing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
High-End Electronic Publishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Electronic Forms Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Computer Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Scanners and Printers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Online Information Dissemination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Online Information Retrieval. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Telecommunications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Electronic Mail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Optical Disks.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Expert Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Technical Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
CHAPTER 4:ALTERNATIVE FUTURES FOR THE GOVERNMENT
PRINTING OFFICE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Traditional GPO-Centralized . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Demand for Traditional GPO Services . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Impacts of Medium-Term Reductions in
Traditional Demand. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

vii
Page
Traditional GPO–Legislative Branch
Only . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Financial Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Labor Force Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Other Vulnerabilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Electronic GPO–Decentralized . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Trends in Technology and Demand. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Opportunities and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
CHAPTER 5:AN ELECTRONIC NATIONAL TECHNICAL
INFORMATION SERVICE AND NTIS/SUPERINTENDENT OF
DOCUMENTS COOPERATION . . . . . . . . . . . . . . . . . . . . . . . . . ............107
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............................107
Role and Current Status of NTIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ....108
Opportunities and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............111
Trends in Demand and Revenues . . . . . . . . . . . . . . . . . . . . . ................111
Possible New Initiatives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ., , .116
NTIS/SupDocs Cooperation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , .. ....119
Differences and Similarities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..119
Disadvantages and Advantages . . . . . . . . . . . . . . . . . . . . . .................121
CHAPTER 6: INFORMATION TECHNOLOGIES, LIBRARIES, AND THE
FEDERAL DEPOSITORY LIBRARY PROGRAM . ....................127
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................127
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...128
Role of Information Technologies in Libraries . . . . . . . . . . . . . . . . . . . . . .......131
Use of Specific Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .132
Online Database Services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...133
Library Communication Networks . . . . . . . . . . . . . . . . . . . . . ...............134
Electronic Bulletin Boards. . . . . . . . . . . . . . . . . . . . . . . . ...................135
Optical Disks.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...................136
Facsimile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................137
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........................138
Federal Depository Library Program . . . . . . . . . . . . . . . . . . . . . . ..............138
Origins and Operations of the Depository Library Program . . . . . . . . . ......138
Format of Depository Library Materials: Paperv. Microfiche . ............140
Dissemination of Information in Electronic Format. . . . . . . . . . . . . . . . . . . . . .142
Online Catalogs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...........144
CHAPTER 7: ALTERNATIVE FUTURES FOR THE DEPOSITORY
LIBRARY PROGRAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .149
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................149
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........151
Alternative I: Status Quo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ...151
Alternative II: Electronic Depository Library Program . . . . . . . . . . . . . . . . . . ..153
A Subalternative for Distributing Electronic Formats. . ..................157
Alternative III: Reorganized Electronic Depository System . . . . . . . . . . ......158
Disseminating Electronic Information Products–Two Case Studies. . ........160
Congressional Record on CD-ROM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....161
Federal Register Online . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....164
Issues Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......171
Dissemination Formats in the Depository Program . . . . . . . . . . . . . . . . . . . . ..171
Changing Costs of the Depository Program . . . . . . . . . . . . . . . . . . . . . .......173

,.,
VII!
Page
Reorganized Depository Program . . . . . . . . . . . . . . . . . . . . . ................175
Changing Roles of Stakeholders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .176
CHAPTER 8: ELECTRONIC DISSEMINATION OF CONGRESSIONAL
INFORMATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ., .....183
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,183
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...184
Current Methods of congressional Information Dissemination . . . . . . . . .. ...185
Congressional Information Products Case Studies:
Congressional Record and Bill Status Information . . . . . . . . . . . . . . . . . . . . .187
Congressional Record . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ....187
Bill Status Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .190
Dissemination Practices of congressional Support Agencies . . . . . . . . . . . . . . . .192
Office of Technology Assessment . . . . . . . . . . . . . . . . . . . . . ................192
General Accounting Office . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...194
Congressional Research Service. . . . . . . . . . . . . . . . . . . . . . .................195
Discussion of crosscutting Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .196
Benefits of Electronic Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....196
Congressional Responsibility for Electronic Access . . . . . . . . . . . . . . . . . . . . ..199
Need for an Index to Congressional Information . . . . . . . . . . . . . . . . . . . . . . 201
Role of GPO.... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................201
Need for Congressional Coordination . . . . . . . . . . . . . . . . . . . . . .............202
CHAPTER9: THE FREEDOM OF INFORMATION ACT IN AN
ELECTRONIC AGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........,,,207
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..207
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............209
Applicability of FOIA to Electronic Media. . . . . . . . . . . . . . . . . . . . . . .........210
Computerized Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...210
Other Media.. . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . .............211
Defining the Limits of Searching Under FOIA . ..........................213
Traditional Interpretations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..213
In the Computer Context: The Distinction Between Searching and
Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................215
Determining the Format of Information Delivered . . . . . . . . . . . . . . . . . . . . ...221
Expanding the Legal Frontiers: Public Access to Software and
Online Databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .222
Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................222
Online Databases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .........223
Fee Assessment and Fee Waivers: Charged Issues in an Age of
Electronic Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ....225
New Technologies and the Need for Amending FOIL . . . . . . . . . . . ...........226
Electronic Information Technologies Are Obscuring the Boundary
Between Record and Nonrecord Material . . . . . . . . . . . . . . . . . . . . . ........228
Computers Are Facilitating Faster and More Complex Searches,
Thereby Encouraging a Broader Definition of a
“Reasonable’ ’Search. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..............229
Electronic FOIA Requests Can Be Incompatible With the Ways
Agencies Collect and Organize Information . . . . . . . . . . . . . . . . . . . . . . . . , .230
Computer Searching Raises New Staffing and Budgetary Problems,
as Well as Opportunities for Federal Agencies . . . . . . . . . . . . . . . . . . . . .232
Federal Agencies Are Using Information Products Whose Status
is Unclear Under FOIA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .233
Page
Paper Printouts of Electronic Information May Not Satisfy
Public Access Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............234
Computers Are Prompting New Discussion About the Basic
Purposes of FOIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............236
CHAPTER 1O:THE ELECTRONIC PRESS RELEASE AND
GOVERNMENT-PRESS RELATIONSHIPS. . ..................,......239
Slummary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................239
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............240
Functions and Problems of Agency Press Offices . . . . . . . . . . . . . . . . . . . . .. ...241
Status of Automation in Federal Agency Press Offices . . . . . . . . . . . . . . ......242
U.S. Department of Agriculture . . . . . . . . . . . . . . . . . . . . . .................242
U.S. Supreme Court . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .243
Environmental Protection Agency . . . . . . . . . . . . . . . . . . . . . ...............243
White House . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............,244
Bureau of Labor Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..244
The Federal Election Commission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...244
Status of Automation in Press Newsrooms . . . . . . . . . . . . . . . . . . . . . .........245
Challenges to Government/Press Automated Dissemination . ...............247
Need for Coordination. . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . .247
Need for Improved Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . ., .. ...247
Need for Completeness and Quality Control . . . . . . . . . . . . . . . . . . . . . .......247
Private Contracting and Price Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..247
Potential Unavailability of Paper Copy . . . . . . . . . . . . . . . . . . . . . ...........248
Technological and Strategic Choices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...248
Technological Choices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..., ...248
Strategic Choices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........250
CHAPTER 11: FEDERAL INFORMATION DISSEMINATION POLICY IN
AN ELECTRONIC AGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...255
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .255
Renewed Commitment to Public Access . . . . . . . . . . . . . . . . . . . . . ............257
Clarification of Governmentwide Information Dissemination Policy . .........261
Cost-Effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .........261
Electronic v. Paper Formats . . . . . . . . . . . . . . . . . . . . . . . . . .264
Possible Congressional Actions . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .267
Clarification of Institutional Roles and Responsibilities . ...................271
Improvements in Information Dissemination Management . ................276
Electronic Publishing/Dissemination Technical Standards. . ...............276
Governmentwide Information Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .277
Government Information Dissemination Innovation
Centers/Committees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......278
Revised Information Resources and Personnel Management . .............280
Improvements in Conventional Printing . . . . . . . . . . . . . . . . . . . . . . . . . . . . .283
Cost.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..283
Timeliness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............288
Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ....288
Cost Estimating and Billing Procedures . . . . . . . . . . . . . . . . . . . . . ..........289
General Themes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......................289
CHAPTER 12: SETTING FUTURE DIRECTIONS FOR THE
SUPERINTENDENT OF DOCUMENTS AND NATIONAL TECHNICAL
INFORMATION SERVICE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........295

x
Page
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................295
Introduction . . . . . . . . . . . . . . . . . . . . . .296
The Competitive Electronic Environment . . . . . . . . . . . . . . . . .296
Institutional Alternatives for SupDocs and NTIS Electronic Information
Dissemination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .300
Centralizing Government Electronic Information Dissemination . . . . . 300
Privatizing SupDocs and NTIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...3oI
Reorganizing SupDocs as Part of a Legislative Printing Office . ...........303
Consolidating NTIS with SupDocs and/or Reorganizing as a
“Government Information Office” or Government Corporation . .........305
Authorizing SupDocs or the Consolidated SupDocs/NTISto Produce
and Disseminate Electronic Formats . . . . . . . . . . . . . . . . . . . . . . . .307
Broader Implications of SupDocs/NTIS Electronic Information
Dissemination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .309
Government Productivity and Cost-Effectiveness. . . . . . . . . . . . . . . .309
Agency Missions and the Role of SupDocs and NTIS... . ................310
Private Sector Economy. . . . . . . . . . . . . . . . . . . . . . . . . .312
Other Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .316
APPENDIXA: ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . .. ...323
APPENDIX B: CONTRIBUTING AUTHORS . ..........................330
APPENDIX C:LIST OF CONTRACTOR REPORTS . ....................331
APPENDIXD: SOME KEY TERMS AND DEFINITIONS . ..............332
CONTENTS

Page
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Opportunities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. ... ... sO..... 5
Problems and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Government Printing Office . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
National Technical Information Service. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
National Technical Information Service/Superintendent of Documents . . . . 14
Depository Libraries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Technical/Management Improvements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Statutory/Oversight Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Legislative Branch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
About This Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Boxes
Box Page
A. Information, the Lifeblood of the Federal Government . . . . . . . . . . . . . . 4
B. General Accounting Office Surveysof Federal Agencies and
Federal Information Users . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
C. Some Opportunities for Productivity Improvement or Cost Avoidance
Through Electronic Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
D. National Technical Information Service and Superintendent
of Documents, How They Compare.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
E. The Importance of Text Markup and Page Description Standards for
Information Dissemination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Tables
Table Page
l-1. Civilian Departmental Agency Dissemination of Statistical
Information, by Format Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
I-Z. Civilian Departmental Agency Dissemination of Scientific and
Technical Information, by Format Used, Current and Projected . . . . 6
l-3. Depository Library Demand for Federal Information,
by Type and Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
l-4. Civilian Departmental Agency Use of Selected Electronic
Publishing-Related Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
l-5. GPO Workload Distribution, Fiscal Year 1987 . . . . . . . . . . . . . . . . . . . 12
l-6. Trends in Sales of Selected NTIS Products, Fiscal Years,
1980, 1987. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
l-7. Trends in New Titles Received byNTIS, Fiscal Years, 1983, 1987 . . 13
l-8. Depository Library Demand for Federal Information in
Electronic Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
l-9. Depository Library Access to Information Technology . . . . . . . . . . . . 17
l-10. Federal Agency Policies on Electronic Information Dissemination . . . 19
Chapter 1
Summary

INTRODUCTION
use of electronic technology offers many new
opportunities for cost-effective dissemination,
If a Nation expects to be ignorant and free
serious conflicts have arisen over how to main-
in a state of civilization, it expects what never
was and never will be . . . if we are to guard tain and strengthen public access to govern-
against ignorance and remain free, it is the ment information and balance the roles of in-
responsibility of every American to be informed. dividual Federal agencies, governmentwide
–Thomas Jefferson, Julv 6, 1816
dissemination mechanisms, and the private
sector.
OTA has concluded that congressional action
Federal information is used by all sectors of is urgently needed to resolve Federal informa-
society. For example, the business and finan- tion dissemination issues and to set the direc-
cial communities look to price levels and gov- tion of Federal activities for years to come. The
ernment indicators of economic activity as government is at a crucial point where opportu-
important inputs to business planning and in- nities presented by the information technologies,
vestment decisions. Similarly, the agricultural such as productivity and cost-effectiveness im-
community regularly uses government crop provements, are substantial. However, the
and weather bulletins, as well as forecasts, to stakes, including preservation andlor enhance-
aid in scheduling crop planting. Scientists and ment of public access to government informa-
engineers benefit from technical information tion plus maintenance of the fiscal and adminis-
generated by federally conducted or sponsored trative responsibilities of the agencies, are high
research in areas like superconductors, super- and need to be carefully balanced by Congress.
computers, and solar energy. Indeed, informa- Congress has enacted numerous laws that
tion generated by the Federal Government emphasize the importance of broad public ac-
spans the entire spectrum of issues and pro- cess to Federal information (such as the Print-
grams relevant to agency missions-from pub- ing Act of 1895, Depository Library Act of
lic health crises, such as AIDs; to environ- 1962, Freedom of Information Act of 1966, and
mental problems, such as hazardous waste Paperwork Reduction Act of 1980) and assign
disposal and water pollution; to demographic various information dissemination functions
and employment trends. And at the most basic to individual Federal agencies (see box A) and
level, information about governmental proc- governmentwide clearinghouses. The latter in-
esses—such as the Congressional Record for clude principally the Superintendent of Docu-
Congress and the Federal Re~”ster for the ex- ments (SupDocs) at the U.S. Government
ecutive branch agencies—is used by citizens Printing Office (GPO), Depository Library Pro-
and organizations that wish to monitor and gram (DLP) also at GPO, National Technical
participate in a wide range of government Information Service (NTIS), and Consumer In-
activities. formation Center (CIC). However, the exist-
For most of this Nation’s history, Federal ing statutory and institutional framework was
information has been disseminated predomi- established by Congress largely during the pre-
nantly in the form of paper documents and, electronic era. It is important, therefore, that
in recent decades, to a lesser extent in micro- Congress review this framework to determine
fiche. However, in the last few years, techno- what actions are needed to ensure that legis-
logical advances have resulted in a rapid in- lative intent is carried out in an electronic envi-
crease in the use of electronic formats for ronment and whether any adjustments in legis-
Federal information dissemination. While the lative objectives or legislation are needed.

3
4

Box A.—Information, the Lifeblood of the Federal Government

Information is truly the lifeblood of many Federal Government programs and activities
and is essential to the implementation of agency missions as well as to informed public debate
concerning such programs and activities. Congress has enacted hundreds of specific laws that
assign information dissemination and related functions to Federal agencies. Some illustrative
laws include:
● Public Law 96-374, Education Act Amendments of 1980, Department of Education to
establish an information clearinghouse for the handicapped;
● Public Law 96-399, Housing and Community Development Act of 1980, Department
of Housing and Urban Development to collect and report data on sales prices for new
homes;
● Public Law 96-482, Solid Waste Disposal Act Amendments of 1979, Environmental Pro-
tection Agency to collect, maintain, and disseminate information on energy and materi-
als conservation and recovery from solid waste;
● Public Law 97-98, Agriculture and Food Act, Department of Agriculture to develop an
agricultural land resources information system and to establish relations with foreign
agricultural information systems;
● Public Law 97-290, Export Trading Company Act of 1982, Department of Commerce
to disseminate information on export trading;
● Public Law 98-362, Small Business Computer Crime Prevention Act, Small Business
Administration to establish an information resource center on computer crime;
● Public Law 99-412, Conservation Service Reform Act of 1985, Department of Energy
to disseminate information annually to States and public utilities on residential energy
conservation; and
● Public Law 99-570, National Antidrug Reorganization and Coordination Act, Depart-
ment of Health and Human Services to establish a clearinghouse for alcohol and drug
abuse information.

SOURCE: Congressional Research Service and Office of Technology Assessment, 1988.

This assessment presents information and confidentiality considerations). The report fo-
analyses on a broad range of topics and issues. cuses on the process of information dissemi-
It is intended to: nation, including the Federal Government’s
technical and institutional infrastructure for
● help both Congress and the Nation bet-
dissemination, not on information collection
ter understand Federal information dis-
(although also important). The report consid-
semination in an electronic age; and
ers a wide range of information formats—from
● assist Congress in implementing improve-
paper and microfiche to computer tapes and
ments in Federal information dissemina-
diskettes, compact disks, and online databases.
tion activities.
And the report covers all major types of Fed-
The focus of this report is on public infor- eral information at a general level—including
mation, that is, Federal information that is or agency reports and pamphlets, rules and reg-
should be in the public domain and is not sub- ulations, periodicals and bibliographies, sta-
ject to exemption under the Freedom of Infor- tistical information, and scientific and techni-
mation Act (e.g., due to privacy, security, or cal information, among others.
5

OPPORTUNITIES
The Federal Government today stands at a Many individual Federal agencies already
major crossroads with respect to the future of are experimenting with and increasingly im-
Federal information dissemination. Techno- plementing information dissemination via elec-
logical advances have opened up many new and tronic bulletin boards, floppy disks, compact
potentially cost-effective ways to disseminate optical disks, desktop publishing, and elec-
Federal information, especially those types of tronic printing-on-demand. For example, sta-
information (such as bibliographic, reference, tistical data are highly suited to electronic for-
statistical, and scientific and technical) that are mats, and, based on the results of the General
particularly well suited to electronic formats. Accounting Office (GAO) survey of Federal
agencies (see box B), about one-third of the
OTA expects several key underlying tech- civilian departmental agencies use magnetic
nical trends to continue unabated for at least
tape or disks, one-fifth floppy disks and elec-
the next 3 to 5 years and 10 years or more in
tronic data transfer, and one-tenth electronic
many cases. These include:
mail for dissemination of statistical data (see
● continued, steady improvement in the Table l-l). By comparison, about three-fourths
price/performance of microcomputers, of the agencies use paper and roughly one tenth
nonimpact printers, scanners, and desk- use microfiche for disseminating statistical
top software; data. Overall, civilian agencies (departmental
● rapid proliferation of desktop publishing and independent) reported over 7,500 informa-
systems and continued improvement in tion products disseminated electronically, as
the ability of desktop systems to produce of fiscal year 1987. The number of civilian
higher quality, more complex documents; agency publications in paper format appears
● rapid growth in networking of desktop to be declining slowly, while the number of elec-
and high-end systems, nonimpact tronic products has more than tripled over the
printers, and phototypesetters used for past 4 years. The GAO survey results suggest
more complex, higher volume, and/or larg- that this trend will continue. For example, by
er institutional applications; 1990, agency use of electronic mail and bulle-
● continued increase in the number and use tin boards, floppy disks, and compact optical
of computerized online information serv- disks in disseminating scientific and techni-
ices and online information gateways (that cal information is expected to more than dou-
provide the channels for information ex- ble, on the average, as shown in Table 1-2.
change), and continued advances in the With respect to demand for Federal infor-
underlying computer and telecommunica- mation, OTA has concluded that, for the fore-
tion technologies; seeable future, paper will continue to be the
● rapid advances in optical disk technologies preferred format for many purposes, such as
and applications, including accelerating browsing government reports, and microfiche
penetration of CD-ROM (compact disk will continue to be used for document storage
read-only memory), maturation of WORM and archival purposes. However, OTA’S 3- to 5-
(write once read many times) and erasa- year outlook for the dissemination of Federal
ble optical disks, plus emergence of CD-I information indicates that overall demand for
(compact disk interactive, with audio, paper formats will decline modestly and the de-
video, graphics, textual, and software ca- mand for microfiche will drop rather markedly,
pabilities all on one disk); and while the demand for electronic formats will in-
● rapid advances in the development of ex- crease dramatically.
pert systems applicable to many aspects
of information dissemination—including There already is a significant demand for
technical writing, indexing, information Federal information in electronic formats among
retrieval, and printing management. user groups, and particularly within the library
6

Box B.—General Accounting Office Surveys of Federal Agencies and Federal Information Users
GAO, at the request of the Joint Committee on Printing, conducted several surveys that pro-
vided important input to the OTA report. Copies of the complete results are available from GAO.
Federal agency survey. In 1987, GAO surveyed all 13 cabinet-level departments and 48 major
independent agencies with respect to information dissemination practices, technologies, budgets,
plans, and policies. GAO asked department or agency senior Information Resources Management
officials to coordinate the response but to consult with agency printing officers, librarians, pub-
lishers, and public information officers, among others. GAO asked that the cabinet departments
provide a separate response for each major subdivision or component, such as bureaus or adminis-
trations. GAO received responses from 114 civilian departmental components, 11 Department of
Defense components, and 48 independent agencies. GAO edited responses for completeness and
internal consistency but did not independently verify their accuracy.
Overall, the survey results are very informative; however, the survey responses were unaudited
and undocumented. Also, it is unclear how the agency responses were developed, especially with
respect to evaluative questions. Nonetheless, the results present a useful overall picture of agency
information dissemination activities.
Federal information user surveys, In 1987-1988, GAO surveyed four user groups: (1) GPO deposi-
tory libraries; (2) other libraries; (3) scientific and technical associations; and (4) general associa-
tions. These groups were surveyed with respect to current and desired types and formats of Federal
information.
As with the Federal agency survey, the results of the user surveys were not verified, and the
exact process by which the responses were provided is not known. Also, the sampling error could
be high, but it does not affect the OTA analysis since OTA has emphasized only the major trends
and findings that emerged from these surveys.

Table 1-1 .—Civilian Departmental Agency especially among the more technically sophis-
Dissemination of Statistical Information, ticated user groups.
by Format Used

Percent of agencies
Format used responding
Table l-2.—Civilian Departmental Agency
Paper . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Dissemination of Scientific and Technical Information,
Magnetic tape/disk . . . . . . . . . . . . . . . . 32
19
by Format Used, Current and Projected
Floppy disk . . . . . . . . . . . . . . . . . . . . . . .
Electronic data transfer . . . . . . . . . . . . 18
Microfiche . . . . . . . . . . . . . . . . . . . . . . . . 12 Percent of agencies
8 responding
Electronic mail . . . . . . . . . . . . . . . . . . . .
Microfilm . . . . . . . . . . . . . . . . . . . . . . . . . 5 Use in
Electronic bulletin board . . . . . . . . . . . 4 Use next
Videotape . . . . . . . . . . . . . . . . . . . . . . . . 2 now 3 years a P e r c e n t
Film . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Format (1987) (by 1990) change
aTOtalS more than 100 percent since many agencies use more than one format Electronic mail . . . . . . . . . . . . . . 6.1 15.8 + 159
SOURCE’ General Accounting Office Survey of Federal Agencies, 1987 Electronic bulletin board . . . . . 6.1 10.5 +72
Electronic data transfer . . . . . . 14.9 18.4 +24
Magnetic tape/disk . . . . . . . . . . 14.0 16.7 +19
community, private industry, Federal agencies Floppy disk . . . . . . . . . . . . . . . . . 8.8 16.7 +90
Compact optical disk . . . . . . . . — 8.8 +
themselves, and various groups with special- %alculated by adding the percentage of agencies now (as of 1987) using the
ized needs (such as educators, researchers, and format indicated to the number who expect to use the format within the next
3 years (by 1990) Assumes that agencies currently using a format will continue
disabled persons). OTA projects that this de- to do SO,
mand will rise sharply over the next few years, SOURCE General Accounting Off Ice Survey of Federal Agencies, 1987
7

search, Federal, State, local, and public libraries)


indicated a strong preference for obtaining in-
creasing percentages of Federal information
in electronic form and declining percentages
in paper and microfiche. The survey results for
318 depository libraries out of a sample of 451
(34 of the 51 regional depositories and 284 of
the 400 selective depository libraries sampled)
are highlighted in Table 1-3. These results show
that, by and large, the depository library com-
munit y desires or anticipates decreases in use
of paper and microfiche formats and signifi-
cant increases in online databases and compact
optical disks. Trends for other surveyed seg-
GPO computer room ments of the Federal information user commu-
nity (e.g., nondepository libraries, scientific
The results of the GAO survey of Federal in- and technical associations) are not so dramatic,
formation users document this likely trend in but show a similar pattern.
demand. For example, the depository library
community (as intermediaries reflecting users Electronic publishing and related technol-
and user information needs in university, re- ogies, when coupled with essential technical

Table l-3.— Depository Library Demand for Federal Information, by Type and Format

Number of libraries responding


Demand
Demand in next Percent
Type of information Format now 3 years change
Congressional Recordlhearingsl
reports/ bi 11s paper 271 234 –14
microfiche 274 225 –18
online database 59 132 + 124
floppy disk o 27 +
compact optical disk 3 112 + 3600
Scientific and technical reports/
information paper 244 172 –17
microfiche 212 159 –22
online database 76 95 +25
floppy disk 1 27 + 2600
compact optical disk 9 78 + 770
Press releases/bulletins paper 246 183 –26
microfiche 39 35 –10
electronic mail or 9 51 +467
bulletin board
online database 24 50 + 108
compact optical disk 1 18 + 1700
Statistical data paper 309 270 –13
microfiche 241 134 –44
electronic mail or 12 27 + 125
bulletin board
online database 103 158 +53
magnetic tape/disk 11 25 + 127
floppy disk 12 65 +442
videodisk o 12 +
compact optical disk 15 140 + 833
SOURCE General Accounting Off Ice Survey of Federal Information Users, 1988.
8

standards, offer the near-term prospect for in- private business users. Private firms typically
tegrated information systems utilizing the “in- report 30 to 50 percent productivity improve-
formation life cycle” concept. Here, the collec- ment with a payback on investment in the 2-
tion, processing, storage, and dissemination to 3-year range. The Federal Government
(and ultimately retention or archiving) of in- spends, conservatively, $6 billion per year on
formation in multiple formats (paper, micro- information dissemination (not including the
form, and electronic) are viewed and imple- cost of collection, processing, or a prorated
mented as interrelated functions rather than share of agency automation). Thus, produc-
separate, unrelated activities. The life cycle tivity improvements on the order of hundreds
concept offers the prospect of improvements of millions of dollars per year appear to be read-
in Federal productivity or cost avoidance ily achievable. In addition, the substantial on-
through increased efficiencies in the publish- going investment by Federal mission agencies
ing of government reports, reduced paper and in agency automation, if planned and imple-
postage costs, and the like (see box C). mented properly, can incorporate multi-format
information dissemination at little additional
The Federal Government should be able to marginal cost, compared to the total cost of
realize at least a significant portion of the automation, and with the potential for net cost
productivity improvements demonstrated by savings in agency information functions.

PROBLEMS AND CHALLENGES


Technological advances are creating a num- eral agencies disseminate at least some of
ber of problems and challenges with respect their information in electronic formats,
to Federal information dissemination: the central governmentwide dissemina-
● At a fundamental level, electronic technol- tion mechanisms (SupDocs, DLP, NTIS,
ogy is changing or even eliminating many and CIC) are presently limited largely to
distinctions between reports, publications, paper or paper and microfiche formats
databases, records, and the like, in ways and thus disseminate a declining portion
of Federal information.
not anticipated by existing statutes and
policies. A rapidly growing percentage of Technology has outpaced the major govern-
Federal information exists at some point mentwide statutes that apply to Federal
in an electronic form on a computerized information dissemination. The Printing
system as part of “seamless web” of in- Act of 1895, Depository Library Act of
formation activities. 1962, and Freedom of Information Act of
● Electronic technology permits information 1966 predate the era of electronic dissem-
dissemination on a decentralized basis that ination, and have not been updated to ex-
is cost-effective at low levels of demand, plicitly reflect electronic as well as paper
but in ways that may challenge tradition- formats. The Paperwork Reduction Act
al roles, responsibilities, and policies. In of 1980 was amended in 1986 to include
contrast, conventional ink-on-paper print- information dissemination within its scope,
ing technology tends to be cost-effective but substantive statutory guidance on
with more centralized production and dis- electronic information dissemination per
tribution and higher levels of demand. se is minimal.
● Electronic technology is eroding the institu- The advent of electronic dissemination
tional roles of governmentwide information raises new equity concerns since, to the ex-
dissemination agencies. While many Fed- tent electronic formats have distinct ad-
9

vantages (e.g., in terms of timeliness,


Box C.–Some Opportunities for Productivity searchability), those without electronic ac-
Improvement or Cost Avoidance Through cess are disadvantaged. In general, the
Electronic Technology library, research, media, public interest,
● Electronic publishing consumer, and State/local government
—facilitates the document revision proc- communities, among others, argue that
ess by minimizing rekey boarding and the Federal Government has a responsi-
graphics redesign; bility to assure equity of access to Fed-
–produces documents that are generally eral information in electronic formats as
found to be more attractive and easier well as in paper. These groups contend
to read; that they are or will increasingly be dis-
—reduces the total publishing time typi- advantaged to the extent that Federal in-
cally by 25 to 50 percent; formation in electronic form is not available
—reduces the total number of document through normal channels.
pages typically by 35 to 50 percent,
since typeset pages contain more text
● Technological advances complicate the Fed-
than typewritten pages; eral Government’s relationships with the
—reduces the costs for paper and post- commercial information industry. While
age for hard copy print runs; and those companies that market repackaged
—can achieve rates of return on invest- or enhanced Federal information benefit
ment of up to 30 to 50 percent and pay- from access to electronic formats, some
back periods of 2 to 3 years or less. of these firms are concerned about possi-
● Compact disk-read only memory (CD- ble adverse effects of government compe-
ROM) tition. Efforts by the Office of Manage-
—can store and disseminate large amounts ment and Budget (OMB) to establish
of information at very low cost; policy in this area have proven to be con-
—is best suited for statistical, reference,
technical, and other information that troversial. Also, the privatization of ma-
does not require frequent updates; jor Federal information dissemination ac-
—can store up to the equivalent of about tivities (such as the NTIS clearinghouse)
250,000 pages of typewritten, double- has not yet been demonstrated to be ei-
spaced text on one disk, or the equiva- ther cost-effective or beneficial for impor-
lent of about 1,500 single-sided floppy tant governmental functions.
disks or about 10 of the 1,600 bits-per- ● OMB and industry representatives support
inch magnetic computer tapes; government dissemination of Federal infor-
—can reduce the cost of dissemination mation in raw electronic form without soft-
by an order of magnitude compared to ware enhancements or searching aids, but
magnetic tapes and up to two orders oppose government dissemination of en-
of magnitude compared to paper doc-
hanced or “value-added” information. This
uments (a typical estimate is that the
same amount of information that could conflicts with the long-established govern-
be disseminated for $50 per week on ment role in producing and disseminating
CD-ROM would cost $345 per week on value-added information products in pa-
magnetic tapes and $2,250 per week in per format and its logical extension to
paper); and electronic formats. Existing policy does
—permits searching, retrieval, and ma- not define “value-added’ or specify under
nipulation of the data in ways simply what conditions value-added electronic
not possible with paper (or microfiche) information products are inherently or
formats. appropriately governmental versus com-
SOURCE: Office of Technology Assessment, 1988. mercial in nature.
● In general information industry represent-
10

atives strongly favor open government and tion to existing agencies and institutions with
unimpeded and nondiscrimin atory access to respect to electronic information dissemination.
Federal information for philosophical and A central challenge is setting future directions
competitive fairness reasons (i.e., so that for the governmentwide information dissemi-
no single vendor has a captive or monop- nation institutions.
oly position over Federal information). In
these respects, the industry shares com- Any electronic future for GPO, NTIS, and
mon ground with the library, research, and DLP must consider the increasingly decen-
press communities, among others. tralized, competitive environment that char-
acterizes the electronic information market-
The absence of congressional action to ad- place. The Federal Government is moving in
dress these issues is likely to result in: the direction of implementing electronic infor-
mation systems at the heart of most agency
● continuing erosion in overall equity of pub- activities. In the long-term, the myriad of pos-
lic access to Federal information, sible information dissemination alternatives,
● continuing confusion over institutional made possible by technological advances, could
roles and responsibilities, serve as a catalyst for significant changes in
● a significant time and dollar cost to the the current institutional framework. Full un-
government and various stakeholders in derstanding of long-term alternatives will re-
seemingly endless debate over statutory quire several years of pilot tests, demonstra-
interpretation and legislative intent, tions, and experiments and related evaluation
● inefficiency and excessive duplication in studies. In the short-to medium-term (3 to 10
electronic information dissemination years), the basis for setting directions is bet-
research and pilot-testing, ter established.
● inability to capture learning from experi-
ence and economies of scale, and
● failure to realize the significant opportu-
nities for cost-effective improvements in . . . an intelligent, informed populace has been,
overall public access to Federal information. is, and will continue to be the fundamental ele-
ment in the strength of our Nation. Contrib-
OTA concluded that the government needs to uting greatly to that intellectual strength is
the so-called Government document, designed
set in motion a comprehensive planning process to disseminate to the American public impor-
for creatively exploring the long-term future (e.g., tant information relative to the activities and
10 to 20 years from now) when the information purposes of its Government.
infrastructure of the public and private sectors
–former U.S. Senator Frank J. Lausche, March 1962
could be quite different. At the same time, the
government needs to provide short-term direc-

GOVERNMENT PRINTING OFFICE


GPO has historically carried out most of the photocomposition, there is very little produc-
Federal Government ink-on-paper printing, tion or sales of products in electronic formats.
either directly or through private contractors, GPO does sell (through SupDocs) some agency
has marketed and sold selected government and congressional products in magnetic com-
documents (in paper and microfiche) to the pub- puter tape format. It also has ongoing pilot
lic (through the SupDocs), and has distributed projects involving both online and CD-ROM
government documents to the depository li- dissemination and both desktop and high-end
braries (through the DLP). While GPO already electronic publishing, pursuant to direction of
makes extensive use of electronic input and the Joint Committee on Printing (JCP).
11

Defining GPO’s future role in the dissemi-


nation of electronic formats presents a major
opportunity for Congress and GPO. One alter-
native, mandatory centralization of all elec-
tronic dissemination through SupDocs (or any
other central government office), would con-
flict with numerous existing agency activities,
would meet strong agency opposition, could
precipitate legal and political challenges, and
would not appear to be cost-effective. On the
other hand, excluding electronic formats from
the SupDocs sales program would erode the
viability and integrity of the program over
time, and compromise the ability of SupDocs
to facilitate broad public awareness and use f’hofo credft U S Government Pvnt(ng Off/cc
of Federal information. A middle ground alter- GPO operator using electronic photocomposition
native, with SupDocs including selected elec- equipment
tronic formats and products, would appear to
strengthen the SupDocs sales program, facili-
rently operating or pilot testing desktop pub-
tate public access, and preserve the prerogatives lishing, computer-aided page makeup, and
of the agencies to disseminate electronically
electronic composition technologies, and one-
themselves (and of private vendors to enhance
third are operating or testing full electronic
and resell electronic formats).
publishing systems, as shown in Table 1-4.
SupDocs sales of magnetic computer tapes, OTA estimates that, as of fiscal year 1987,
floppy disks, compact optical disks, and per- agencies had already spent at least $400 mil-
haps electronic printing-on-demand products lion on electronic publishing-related tech-
would appear to be straightforward, except for nologies.
a possible overlap with NTIS. Sales of online
GPO could have a key role in standards-
services could be more difficult due to staff-
setting, trainin g: and innovative activities rele~
ing, software development, and capital require-
ments, and to more intensive competition with vant to electronic publishing, but GPO will be
operating in a much more decentralized, com-
agencies and commercial vendors.
petitive environment than has traditionally
Another challenge is to define GPO’s role been the case with conventional ink-on-paper
relative to the growth in agency desktop and printing. The general demand for conventional
high-end electronic publishing systems. The printing is likely to continue for several years
GAO survey of 114 civilian agency compo- at a slow growth or steady-state level. How-
nents indicated that one-half or more are cur- ever, in the medium-term (3 to 10 years), a sig-

Table l-4.—Civilian Departmental Agency Use of Selected Electronic


Publishing-Related Technologies

Percent of agencies responding


Currently in Currently prototyping
Technology operational use or pilot testing Totals
Computer-aided page makeup ... .‘ . . . . . 50.0 8.8 58.8
Computer graphics . . . . . . . . . . . . . . . . . . . . . . 65.8 7< 9 73.7
Electronic photocomposition ... ... ... 43.9 7.9 51.8
Laser and other nonimpact printing ., ... . 64.0 1.8 65.8
Desktop publishing system . . . . . . . . . . 34.2 14.9 49.1
Electronic publishing system . . . . . . . . . . . 21.1 .—
10.5 31.6
SOURCE General Accounting Off Ice S“urvey of Federal Agencies, 1987
12

nificant portion of GPO inplant and procured procured printing), none of these alternatives
printing could be suitable for electronic dis- appears to be cost-effective. These alternatives
semination or vulnerable to competition from would largely eliminate concerns about sepa-
electronic formats. The plans and activities of ration of powers, since executive branch print-
defense agencies are particularly important, ing would no longer be done by or through a
since the Army, Navy, and Air Force together legislative branch agency. However, they could
account for roughly one-third of total GPO complicate the functioning of SupDocs and the
billings. Over the next few years, the defense DLP, and could have significant adverse ef-
agencies are hoping to place most manuals, fects on the GPO labor force.
directives, and technical documentation on
electronic media. GPO will have to be innova- OTA identified several opportunities for im-
tive in matching its expertise to agency needs, provement in GPO’s traditional printing serv-
which are likely to vary widely and change at ices, These include more competitive pricing and
an increasingly rapid pace. timely delivery of GPO main plant inhouse work
for executive agencies, itemized estimating and
With respect to GPO’s role in traditional ink- billing practices, regular surveys of customer
on-paper printing, the fiscal year 1987 GPO needs and problems, and revised and strength-
printing workload totaled $771 million, of
ened GPO advisory groups.
which about threequarters was procured from
commercial printing contractors and one-quar-
In principle, the GPO main plant is well posi-
ter carried out at the GPO main and regional
tioned to meet demands for conventional print-
printing plants. As shown in Table 1-5, about
ing, with one of the best equipped printing fa-
80 percent of legislative branch printing work
cilities in the United States and an experienced
is done inplant, while about 85 percent of ex-
work force. However, GPO inhouse printing
ecutive branch printing work is contracted out.
costs are high in part due to the need to main-
Overall, about 45 percent of inplant work is
tain operational capacity to handle a wide
legislative, while about 95 percent of con-
diversity of printing work, and to meet peak
tracted work is for the executive branch.
congressional and priority executive branch
OTA examined several alternatives, includ- workloads. A significant part of this workload
ing decentralizing GPO’s conventional print- is well suited for electronic formats (e.g., Con-
ing and procurement functions, transferring fessional Record, Federal Register). A grad-
GPO’s procurement program to the executive ual transition from paper to electronic formats
branch, and limiting GPO to legislative branch for these items could help reduce GPO costs,
work. Based on information available to OTA potentially increase access to this information,
(including comparative costs of GPO inhouse, and place the GPO main plant on a more com-
GPO procured, agency inhouse, and agency petitive footing for executive branch printing.

Table 1.5.—GPO Workload Distribution, Fiscal Year 1987


(in millions of dollars)

Procured Main plant Regional plant


printing printing printing Totals
Legislative branch . . . . . . . . . . . . . . . . . . . . . . . $ 23 $ 90 $113
Executive branch . . . . . . . . . . . . . . . . . . . . . . . . 552 90 :1!! 656
Judicial branch . . . . . . . . . . . . . . . . . . . . . . . . . 1
Totals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $576 $18; ;1’2 $77:
NA = not applicable.
SOURCE U S Government Prlntlng Office, 1987
. —

13

NATIONAL TECHNICAL INFORMATION SERVICE


NTIS has historically served as the Federal
Government’s archive and clearinghouse for
scientific and technical reports prepared by
Federal agencies or contractors, along with re-
lated indices and bibliographies. The bulk of
NTIS documents are provided in paper or mi-
crofiche format, although, in recent years,
NTIS also has served as a clearinghouse for
some electronic format products (e.g., software
and databases). Also, NTIS performs other re-
lated services such as patent licensing, Japa-
nese literature exchange, and FOIA request
and/or information sales processing for a few
Natlortal Technical In forrnatfon Serv/ce
agencies. Photo credit

NTIS staff pulls an archive document from


With respect to NTIS, the major opportu- the NTIS collection
nity is, quite simply, determining the future
of NTIS as a government entity. NTIS faces
strategic challenges on several fronts. First, Second, a significant percentage (estimated
the core NTIS business, as measured by sales at one-third to one-half, see Table 1-7) of Fed-
of paper and microfiche reports, has been eral scientific and technical reports are never
shrinking (by about 40 to 50 percent) over the provided to NTIS, since agency participation
past decade (see Table 1-6). In part as a result, is strictly voluntary. The NTIS collection is
NTIS prices for these reports have gone up con- thus becoming increasingly incomplete. Third,
siderably faster than the inflation rate in or-
der to help maintain break-even operations. Table 1.7.—Trend in New Titles Received by NTIS,
Over the last few years, NTIS has offset declin- Fiscal Years 1983, 1987
ing revenues from full-text reports and sub-
1983 1987 Net change
scription, bibliographic, and announcement
Number of titles received . ....79,471 62,856 –21 “/0
products with increasing revenues from serv- Estimated percentage of all
ices to other agencies (such as order billing and relevant titlesa. . . . . . . . . . . . 67°/0 530/0 – 140/0
processing), brokerage fees on sales of other aA~~umes the number of relevant agency titles remains constant al 119,000 Per
year
agency materials, and sales of computer-re- SOURCE National Technical Information Service and Office of Technology
lated products. Assessment, 1988.

Table 1-6.—Trends in Sales of Selected NTIS Products, Fiscal Years 1980, 1987

Net
1980 1987 —— change
in thousands of copies
Paper documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 752 393 –48 0/0
Microfiche documents . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 67 –57
in millions of copies
Selected Research in Microfiche (SRI M). . . . . . . . . . . . 2.72 1.33 –51
in thousands of subscriptions
Government Research Announcements and Index . . . 2.22 1.15 –48
Abstract Newsletters . . . . . . . . . . . . . . . . . . . . . .......16.0 6.8 –58
SOURCE National Technical Information Service, 1988
14

NTIS is being outdistanced by most of the Fed- firmed its intent that NTIS remain in the
eral science agencies with respect to use of elec- government, Congress now has the opportu-
tronic information technology. And fourth, nity to determine where NTIS should be lo-
NTIS has been caught in the middle of the on- cated and how it should relate to other Fed-
going debate over privatization of Federal in- eral agencies, including what agency materials
formation functions. Since Congress has af- should or must be submitted to NTIS.

NATIONAL TECHNICAL INFORMATION SERVICE/


SUPERINTENDENT OF DOCUMENTS
Proposals have been made to retain NTIS ideally suited for implementation of an electronic
in the Department of Commerce, as a govern- document system (using optical disk storage,
ment corporation or in essentially its present electronic printing, and multi-format output—
form; consolidate NTIS with SupDocs, either paper, microfiche, and electronic), perhaps using
within GPO or as part of a newly established the Defense Technical Information Center
Government Information Office; and consoli- (DTIC) system as a prototype, that could revital-
date NTIS with the Library of Congress. ize NTIS if coupled with improved agency par-
ticipation. Overall, an electronic NTIS should
Whatever the alternative chosen by Con- be able to greatly increase the diversity and
gress, strengthened NTIS-SupDocs coopera- timeliness of NTIS (and related private ven-
tion would likely lead to improvements in dor) offerings, increase the ability of NTIS (and
indexing, marketing, and international ex- private vendors) to match information prod-
change of Federal information. And strength- ucts with potential users, and reduce costs.
ened cooperation seems essential to the extent
both agencies pursue sales of electronic format
products and that SupDocs enters the low-
demand market. At present, demand for NTIS . . . the new [electronic] technology not only
documents averages about 10 copies per title, gives potential users quicker and more con-
compared to about 2,000 copies per title for venient access to wider bodies of information,
items in the SupDocs sales program (see box including instantly current information, than
D for a comparison of NTIS and SupDocs). can be provided by print alone; it also gives
the user a new kind of ability to search through
NTIS and SupDocs could cooperate on im- and manipulate the information, and in effect
plementing electronic technologies that would to create new information by the selection,
meet NTIS clearinghouse and archival needs, combination, and arrangement of data.
plus support a broadening of the SupDocs –Commission on Freedom and Equality of Access to Information,
American Z.ibrary .4ssociation, 1986.
product line to include selected low-demand
items. Wherever located, NTIS appears to be

DEPOSITORY LIBRARIES
The DLP is administered by GPO and serves charge to the general public. About 55 percent
as a mechanism for dissemination of Federal of the depository libraries are university
agency documents free of charge to the approx- libraries, 23 percent are public libraries, 11 per-
imately 1,400 participating libraries. The cent are law school libraries, 7 percent are Fed-
libraries, in return, provide housing for the doc- eral libraries, and 4 percent are special libraries
uments and access to this information free of and the like.
—.. ——

15

Box D.—National Technical Information Service and Superintendent of Documents,


How They Compare
NTIS SupDocs
Branch of government Executive Legislative
Location Department of Commerce GPO
Statutory authority 15 U.s.c. 1151-1157 44 U.S.C 1701-1722
Total annual revenues’ $22 million $100 million
(approximate)
Titles for sale 2 million 20,000
(approximate)
Total annual sales volumeh 6 million copies 27 million copies
Average sales per title 10 copies 2,000 copies
Primary document formats paper, microfiche paper, microfiche
Primary source of documents Federal agencies and con- Federal agencies, Congress
tractors
Electronic products’ 800 numerical or statistical few dozen magnetic
databases tape products
(approximate) 300 textual databases
300 computer software items
(incl. models)
Prepares bibliographies/ Yes Yes
catalogs
Conducts marketing activities Yes Yes
Carries out international Yes Yes
document exchange
Performs reimbursable services Yes—for agencies Yes—Consumer Information
Information Center, Deposi-
tory Library Program(’
a 1 ~clude~ fisca] ~.ear 19R7 revenues from reimbursable services and services funded through appropriations.
bFiscal “ear 198ti; SUpI)OCS data include Consumer Information Center SdeS.
cFiscal ~rear 198’7.
dRelmbursed through appropriations.
SOURCE; Nat]onal Technical Information Ser\’ice :ind US. Go\rernment Printing Office, 1988.

As with GPO and NTIS, there is a major ernment information in all formats, and other
opportunity to define the future role of the congressional committees concur in the deci-
DLP with respect to dissemination of Federal sion to disseminate certain electronic formats
informatione in electronic formats. As agencies to depositories. OTA concluded that, if it is
make increasing use of electronic formats, limit- to succeed, this emerging policy needs to be
ing the DLP to paper and microfiche products further developed and refined, and have the
would, over time, reduce the type and amount support of DLP participants (especially li-
of Federal information available to the public, braries, GPO, and the agencies that are the
and would erode the legislative intent of the DLP source of most DLP materials). A variety of
(e.g., as expressed in the legislative history of pilot projects, demonstrations, and tests in-
the Depository Library Act of 1962). The impe- volving various technologies, financial arrange-
tus for including electronic information in the ments, and delivery mechanisms (including
DLP is strong. The JCP has interpreted the possible involvement of the private sector) is
DLP statutory provisions as extending to gov- warranted. Ultimately, Congress may wish to
16

ways to make this information available to the


public.

Distribution of selected government infor-


mation products in CD-ROM format such as
the bound, cumulated Congressional Record
could improve access to such information and
could be a cost-effective dissemination mech-
anism for certain datafiles. There could be some
additional equipment and training costs asso-
ciated with this format for the depository li-
brary participants. Delivery of online datafiles
Photo credit Documents Center, Robert W. Woodruff Library, Emory Unlvers/ty (such as the Federal Register) to the public
Librarian assisting user at reference desk at the through depository institutions requires pilot-
Robert W. Woodruff Library testing to determine how best to provide ac-
cess to this information, and how to ensure that
consider a reorganization or restructuring of the additional costs associated with online for-
the current DLP in light of both electronic in- mats do not hinder public access or place un-
formation dissemination options now or likely realistic, unmanageable financial or adminis-
to become available and the evolving nature trative burdens on participating libraries.
of libraries and the telecommunication infras-
The results of the GAO survey of Federal
tructure.
information users indicate a substantial depos-
An important reason for electronic pilot itory library demand for electronic formats.
projects is to better understand the issue of The vast majority of libraries responding in-
costs to users, government, and depository in- dicated that the Record and Register, along
stitutions. If the basic underlying principle of with an index to Federal information and data-
the depository program is to retain free access base of key Federal statistical series, would
to government information for users, then Con- be moderately to greatly useful in both online
gress needs to be aware that there maybe addi- and CD-ROM formats, as shown in Table 1-8.
tional costs associated with the introduction of The GAO survey also found that many of the
certain electronic services, and assist depository depository libraries have access to key infor-
libraries and GPO in designing and financing mation technologies, as shown in Table 1-9.

Table 1-8.—Depository Library Demand for Federal Information in Electronic Formats

Percent of libraries responding


moderately to greatly usefula
Online Offline
immediate CD-ROM
Item access issued monthly
Congressional Record . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 74
Congressional Committee Calendar/Bill Status . . . . . . . . . . . . . . 70 60
Federal Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 80
Federal Agency Press Releases. . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 40
Agency Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 62
Comprehensive Index to Federal Information . . . . . . . . . . . . . . . . 94 90
Integrated Database of Key Federal Statistical Series. . . . . . . . . 90 88
aBaS8d on responses from 318 depository libraries out Of a sample Of 451
SOURCE General Accounting Office Survey of Federal Information Users, 1988
17

Table l-9.—Depository Library Access to Information Technology

Number of libraries
Information technology with access a
Microcomputer without modem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Microcomputer with modem for online access. . . . . . . . . . . . . . . . . . . . . . . 337
Microfiche reader without printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
Microfiche reader with printer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
CD-ROM reader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Videodisk player . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Mainframe computer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
a
Based on responses from 403 depository libraries out of a sample of 451 depository libraries
SOURCE General Accounting Office Survey of Federal Information Users, 1988

TECHNICAL/MANAGEMENT IMPROVEMENTS
OTA identified several important technical libraries and private vendors (perhaps in
management alternatives that could be imple- enhanced form).
mented under a wide range of institutional ● Innovation centers to exchange learning
scenarios and could be implemented by agency and experience about technological inno-
action using existing statutory authorities and vations and user needs relevant to infor-
with congressional concurrence. These alter- mation dissemination. Such centers could
natives include: be designated or established at, for exam-
Technical standards on text markup, ple, DTIC (for the defense sector), NBS
page/document description, optical disks, and NTIS (for the civilian executive branch),
and other areas important to information and GPO (for the legislative branch).
DTIC, NBS, and GPO, along with several
dissemination (see box E). The National
Bureau of Standards (NBS), DTIC (or mission agencies, already have a variety
another responsible Department of of laboratory and/or demonstration activ-
Defense component), and GPO could be ities under way. Agencies could be re-
assigned lead responsibility, presumably quired to conduct “Agency X-2000”
building on accepted or emerging private studies to creatively explore and develop
their own visions of future information dis-
sector industry standards to the extent
semination activities.
possible and working through the exist-
ing national and international standards ● Revised Information Resources Manage-
organizations. ment (IRM) program. A variety of train-
● Governmentwide information index to ma- ing, career development, budget report-
jor Federal information products, regard- ing, and management actions could be
less of format. GPO and/or NTIS could taken to give information dissemination
be assigned lead responsibility to consoli- (including printing, publishing, public af-
date and upgrade existing indices, direc- fairs, press, library, and related activities
tories, and inventories into one integrated and personnel) a stronger and better un-
index. The government could contract derstood role within the IRM concept.
with private firms or library and informa- ● Electronic press release service. Press re-
tion science professionals to carry out leases and other time-sensitive informa-
some of this work. The index could be tion (such as crop reports, weather bulle-
made available in multiple formats and tins, and economic and trade data) from
disseminated both directly from the gov- major Federal agencies could be electron-
ernment as well as via the depository ically provided directly to the press, via
18

Box E.—The Importance of Text Markup and Page Description Standards for
Information Dissemination

Text markup standards are particularly important to realize the full benefits of electronic
information dissemination. If government documents (whether reports, pamphlets, manuals,
other text, or text plus tabular and graphics material) are not prepared in a standardized elec-
tronic format using standardized codes and descriptors, substantial and costly recoding and
rekeyboarding may be necessary at later stages of the dissemination process. Text markup
standards are intended to establish a consistent set of codes for labeling key elements of a
document–such as chapter titles, paragraph indentations, tabular presentations, and the like.
If these electronic codes are widely agreed upon arid used (i.e., standardized), then the docu-
ments can be electronically transferred from one stage in the dissemination process to another
with little or no additional effort and cost, if the equipment is designed to be compatible with
the electronic codes, Three major approaches to text markup standards are:
. GPO logically structured full text database standard;
● Standard Generalized Markup Language (SGML), an international standard that has
been adopted by DoD and NBS; and
● Office Document Architecture (ODA), an international standard under consideration
by NBS.
Page description standards are also very important. If the language or code used by the
page composition equipment is not compatible with the code used by the output devices (e.g.,
printers), then additional work is required to convert the codes. Sometimes it is easier just
to rekeyboard and recode the entire document, at significant additional cost. Page description
languages are intended to establish a consistent set of codes compatible with both composition
and output equipment. One possible page description standard is PostScript, a defacto indus-
try standard under consideration by NBS and the national and international standards organi-
zations. Another possibility is the Standard Page Description Language (SPDL) now being
developed.

SOURCE: National Bureau of Standards, Defense Technical Information Center, and U.S. Government Printing Office, 1988.

private electronic news and wire services,


and to the DLP. A major issue concerns
equity of press access and the need to en-
sure that cost or technical requirements
do not discourage smaller, less affluent,
and/or out-of-town news organizations
from realizing the potential benefits. While
electronic press releases can be more timely
and cost-effective than messenger or mail
delivery of paper releases, dual format (pa-
per and electronic) would appear to be
necessary-at least for a lengthy transi-
tion period–for those news outlets with- Photo cred(t USA TODAY, Gannett, Co /nc all rights reserved
out, or lacking interest in, online electronic
Reporter sitting at video display terminal
capability.
19

STATUTORY/OVERSIGHT CHANGES
Congress could amend the Printing Act, De- jority of agencies do not have policies on elec-
pository Library Act, and Paperwork Reduc- tronic dissemination (see Table 1-10). As agen-
tion Act to provide statutory direction for spe- cies begin to develop such policies, the OMB
cific institutional and technical/management view is likely to have a dominant role, in the
alternatives, as well as to provide general phi- absence of clear and positive congressional
losophical guidance on electronic information guidance. Congress may wish to amend specific
dissemination. statutes or otherwise promulgate its own views
on the basic principles addressed and policies
At the most basic level, a fundamental cross-
enunciated in OMB Circular A-130 as it relates
cutting issue is public access to Federal infor- to Federal information dissemination. In par-
mation. Debate over the use of electronic for- ticular, Congress could provide more specific
mats, privatization, and the like is obscuring guidance on the role of the private sector and
the commitment of Congress, as expressed in contracting out of Federal information dissem-
numerous public laws, to the importance of ination, user charges, and provision of value-
Federal information and its dissemination in -added information products. Congress could
carrying out agency missions, and the princi- also make any necessary adjustments in over-
ples of democracy and open government. A re- sight mechanisms (such as establishing a Joint
newed congressional commitment to public ac- Congressional Committee on Government In-
cess in an electronic age may be needed. formation).
Congress may wish to legislate a govern- With respect to the Freedom of Information
mentwide electronic information dissemination Act (FOIA), this statute too was enacted in
policy. In so doing, Congress would need to an era when paper records were the dominant
consider several sometimes competing con- form of government information. The applica-
siderations, including: enhancing public access; tion of FOIA to electronic formats has created
minimizing unnecessary overlap and duplica- a number of problems. The courts have ex-
tion in Federal information activities; optimiz- pressed a need for Congress to clarify gray
ing the use of electronic versus paper formats; areas left open by the statute. For example:
and optimizing the role of the private sector.
OMB has promulgated its own view, albeit con- c The case law as applied to paper infor-
troversial, of appropriate public policy (in the mation establishes that FOIA does not re-
form of OMB Circular A-130). The vast ma- quire agencies to create new records in
Table I-l O.—Federal Agency Policies on Electronic Information Dissemination

Percent of agencies having


documented policies
Policies and procedures for Dept. a I nd;b
Public access to agency electronic databases?
yes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.6 10.4
no . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90.4 89.6
Electronic dissemination by agency contractors?
yes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.9 6.3
no . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43.0 41.7
do not use contractors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49.1 52.1
App/icabi/ity of FO/A to electronic formats?
yes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.4 25.0
no . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.6 75.0
ap~rC~nt of 114 depafirnental clvlllan agency Cofnponents responding
bPerCent of 4J3 independentcivilian agency Components responding.
SOURCE General Accounting Office Survey of Federal Agencies, 1987
20

fulfilling requests. When additional pro- ● Another issue is whether and under what
gramming is required to extract informa- conditions the advantages of electronic
tion from computer systems, agencies and formats are such that access to the format
courts have sometimes held that such pro- as well as the information itself should be
gramming would be analogous to record guaranteed. Although the case law and the
creation, and therefore would not be a re- FOIA fee guidelines have established that
quired part of the FOIA “search” proc- computer-stored information is subject to
ess. In the electronic age, however, some FOIA, requesters are not guaranteed ac-
degree of reprogramming or program cess to the information in formats other
modification may be essential to obtain than paper. If large quantities of data
access to electronic information. could be more effectively utilized with the
● Another gray area involves defining a flexibility offered by magnetic tapes,
‘‘reasonable effort on the part of the gov- disks, or online retrieval, access to these
ernment in searching for records respon- electronic media may be important.
sive to a FOIA request. In the computer
Congress could amend FOIA to bring elec-
context, the programming/no program-
tronic formats clearly within the statutory pur-
ming distinction has begun to separate de-
view, define the scope and limits of FOIA
cisions about ‘reasonableness from con-
searches in an electronic environment, and clar-
siderations of effort. This is incongruous
ify fees and procedures for FOIA requests for
with tradition, as significant expenditures
electronic information. For the 1990s and be-
of effort continue to be involved in man-
yond, Congress may need to decide whether
ual FOIA searches. Retrieval of paper doc-
the FOIA should continue to be viewed as an
uments may involve extensive tracking, “access to records” statute, or whether it
communication with various bureaus, con-
should be perceived more broadly as an “ac-
solidation of disparate files, and substan-
cess to information” statute. Due to the ex-
tial hand deletions of exempted materials.
plosive growth in electronic information storage,
As computer capabilities for searching,
processing, and transmission by the Federal
segregating, and consolidating of data be-
Government, traditional views about records
come increasingly efficient and cost-effec-
and searches may need to be modified to en-
tive, computer searches could be broadened sure even basic access to computerized public
and public access enhanced. Agencies may
information.
need to focus on designing new ways to
respond more readily to FOIA requests
for computer records.

LEGISLATIVE BRANCH
Congress itself is a major source of Federal ever, increasingly, electronic formats offer sig-
information. Congressional information ranges nificant advantages in terms of timeliness and
from the Congressional Record to congres- searchability, and are being utilized by private
sional calendars and schedules to the status vendors and congressional in-house support
of pending legislation to a wide range of com- offices (e.g., the House Information Systems
mittee reports, and to numerous documents Office and CRS) for a growing range of con-
produced by the analytical support agencies gressional information.
(Congressional Research Service [CRS], Gen-
eral Accounting Office [GAO], and Congres- To a large degree, OTA’S general findings
sional Budget Office ICBO], as well as OTA). about technological trends and opportunities
Most of this information has been and con- also apply to congressional information. Elec-
tinues to be available in paper formats. How- tronic options offer the potential to make con-
21

gressional information more quickly and widely and how that information should be made
available. This can be very important for citi- available (by GPO, other congressional offices,
zens and organizations-whether consumer, li- and private vendors). For example, because of
brary, research, labor, or business in nature— GPO’s growing role in providing electronic for-
that desire to closely follow congressional mats to Congress as part of the electronic pub-
activity and/or participate in the legislative lishing process, GPO is positioned to more ac-
process. As congressional offices automate, in- tively participate in disseminating electronic
creasing amounts of information are created, congressional information to the GPO deposi-
revised, and stored in electronic form. This cre- tory libraries and the public-at-large. At the
ates the potential to apply “information life same time, some commercial vendors would
cycle’ and ‘multi-format output concepts to like to contract directly with Congress, per-
the legislative branch as well as to the execu- haps on a bulk rate discount basis, for elec-
tive branch. Again, common technical stand- tronic dissemination of congressional informa-
ards will be important in realizing this po- tion to libraries, the public, and Congress itself.
tential. Finally, given the large number of House,
Congress has the opportunity to establish a Senate, and congressional support offices and
strategic direction for electronic dissemination units involved with the creation and dissemi-
of legislative branch information. The impor- nation of congressional information, Congress
tance of congressional information to an in- may wish to establish a formal coordinating
formed citizenry and the need to ensure equitable mechanism to maximize the exchange of learn-
channels of access for all interested citizens, in- ing and minimize the potential overlap, and
cluding access to electronic formats, are widely to take advantage of the opportunities for tech-
accepted in principle. The differences of opinion nologically enhanced access. In many respects,
focus on the means of implementation. congressional decisions on electronic dissemi-
nation of congressional information are just
In setting an overall direction, Congress will as important as prior decisions on radio and
need to determine its own level of responsibil- television coverage of congressional hearings
ity for ensuring that electronic congressional and floor sessions.
information is readily available to the public,

ABOUT THIS REPORT


The report is organized into 12 chapters. DLP. Chapter 4 examines three alternative fu-
Chapter 1 is the summary. Chapters 2 and 3 tures for GPO printing functions—continua-
together provide an overview of key technical tion of a traditional ink-on-paper printing role
and institutional trends and issues. Chapter only, for both the legislative and executive
2 presents a picture of current evolving Fed- branches; a GPO for the legislative branch
eral Government information dissemination only; and the so-called decentralized electronic
technologies and activities. The results of the GPO that would involve expanded electronic
GAO survey of Federal agencies are used ex- publishing activities and the inclusion of some
tensively. Chapter 3 discusses current techni- electronic formats in the SupDocs sales pro-
cal trends that are relevant to Federal infor- gram. The results of the GAO surveys of Fed-
mation dissemination and that are expected eral information users are used extensivel y in
to continue or intensify for 3 to 5 years into chapter 4. These three alternatives highlight
the future and in many cases longer. a range of considerations important to plan-
ning GPO’s future.
Chapters 4, 5, 6, and 7 provide substantial
analysis and discussion on the major govern- Chapter 5 examines the opportunities and
mentwide information dissemination institu- challenges facing NTIS. Some of the GAO user
tions–GPO (including SupDocs), NTIS, and survey results are included, and survey results
22

previously cited in chapters 2 and 4 are also Chapter 11 considers a wide range of policy
relevant. Improved cooperation between NTIS and institutional issues that are relevant to
and SupDocs is also examined. Much of the Federal information dissemination. Chapter 11
discussion is relevant to NTIS and SupDocs also highlights the debate over the applicabil-
regardless of the institutional structure as long ity and interpretation of key policy instru-
as NTIS remains in the government in some ments to electronic dissemination. Chapter 12
form. discusses possible future directions for and
broader implications of SupDocs and NTIS in-
Chapters 6 and 7 provide indepth analysis volvement in Federal electronic information
and discussion of the history and current sta- dissemination.
tus of GPO’s Depository Library Program
with respect to electronic dissemination. Chap- Several crosscutting themes are relevant to
ter 6 covers a range of electronic information many chapters. Three of the most important
technologies currently used or whose use is con- themes are:
templated by libraries, and introduces the 1. public access to Federal information,
DLP, current technology and several dissem- 2. user needs for Federal information, and
ination issues. Chapter 7 examines and evalu- 3. the private sector role in Federal informa-
ates in considerable depth a range of alterna- tion dissemination.
tive futures for DLP with specific illustrations.
Two case studies are presented on the Con- While there are not separate chapters devoted
gressional Record and the Federal Register. to these topics, they are discussed through-
Finally, chapter 7 provides an analysis of pend- out the report. Also, while there is a separate
ing DLP policy and institutional issues regard- chapter on technology trends, technology is
ing electronic dissemination. discussed to varying degrees in every chapter
of the report. Similarly, while there are sepa-
The next three chapters—8 through 10— rate chapters on GPO, NTIS, and DLP, there
examine other important dimensions of Fed- is at least something significant in every chap-
eral information dissemination. Chapter 8 dis- ter of the report that is relevant to planning
cusses congressional information dissemina- the future of these institutions.
tion with particular attention to two case
studies (on the Congressional Record and bill For discussion of related topics not covered
status information) and to the dissemination in this report, see the other OTA reports listed
practices of three congressional support agen- below. These reports cover such topics as: the
cies (OTA, GAO, and CRS). tension between public access to government
information and: protection of national secu-
Chapter 9 presents an indepth analysis of
rity interests; physical security and data in-
FOIA with respect to electronic formats. This
tegrity; privacy rights of individuals and orga-
chapter reviews statutory and judicial prece-
nizations; and intellectual property rights.
dents on the applicability of FOIA to electronic
media, and examines possible directions for Other reports cover the need to preserve gov-
ernment information for archival and histori-
amending FOIA in light of the implications
cal purposes, and the need to consider govern-
of technological change for basic FOIA con- ment information in the context of long-term
cepts. Chapter 10 focuses on the electronic
social, political, and economic changes relevant
press release and its implications for govern-
to the information and communication infra-
ment-press relationships. The chapter reviews
structure of the United States.
the status of automation in Federal agency
press offices and in the press newsroom, ex- ● Medlars and Health Information Policy—
amines the strengths and weaknesses of elec- A Technical Memorandum, OTA-TM-H-
tronic press releases, and discusses the tech- 11, September 1982. NTIS order #PB
nological and strategic choices. 83-168658.
23

Federal Government Information Tech- Commercial Newsgathering From Space,


nology: Electronic Surveillance and Civil OTA-TM-ISC-40, May 1987. GPO stock
Liberties, OTA-CIT-293, October 1985. #052 -O03-O1066-6; $3.00.
GPO stock #052-003-01015-l; $3.00. NTIS The Electronic Supervisor: New Technol-
order #PB 86-123 239/AS. ogy, New Tensions, OTA-CIT-333, Sep-
Federal Government information Tech- tember 1987. GPO stock ~052-003-01082-8;
nology: Management, Security, and Con- $6.50.
gressional Oversight, OTA-CIT-297, Feb- Defending Secretsf Sharing Data: New
ruary 1986. GPO stock #052 -O03-O1026-7; Locks and Keys for Electronic Informa-
$7.50. NTIS order #PB 86-205 499/AS. tion, OTA-CIT-31O, October 1987. GPO
The Regulatory Environment of Science, stock #052 -O03-O1083-6; $8.50.
OTA-TM-SET-34, February 1986. GPO Science, Technology, and the First Amend-
stock #052 -O03-O1024-l; $6.00. NTIS or- ment, OTA-CIT-369, January 1988. GPO
der #PB 86-182 003/AS. stock #052 -O03-O1090-9; $3.50.
Intellectual Property Rights in an Age of Book Preservation Technologies, OTA-O-
Electronics and Information, OTA-CIT- 376, May 1988. GPO stock #052-003-
302, April 1982. GPO stock #052-003- 01103-4; $5.00.
01036-4; $15.00 NTIS order #PB 87-100 Communication Systems for an Informa-
301/AS. tion Age, OTA-CIT, forthcoming, spring
Federal Government Information Tech- 1989.
nology: Electronic Record Systems and Scientific and Technical information Dis-
1ndividual Privacy, OTA-CIT-296, June semination: Opportunities and Problems,
1986. GPO stock #052 -O03-O1038-l; $7.50 forthcoming, spring 1989.
NTIS order #PB 87-100 335/AS.
Chapter 2

Overview of
Federal Information
Dissemination

Photo cradit: Chase Studios

National Institute of Health medical staff using the National Library of Medicine’s
Medline database
CONTENTS
Page
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Size and Scope of Federal Information Dissemination Enterprise . . . . . . . . 28
Technological Initiatives by Federal Agencies . . . . . . . . . . . . . . . . . . . . . . . . 32
Institutional Infrastructure for Federal Information Dissemination . . . . . . 36

Tables
Table Page
2-l. Federal Expenditures on Information Dissemination, Civilian and
Military . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2-2. Illustrative Agency Expenditures for Information Dissemination,
Fiscal Year 1987 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2-3. Selected Federal Agency Information Dissemination Activities,
Fiscal Years 1983 and 1987 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2-4. Types of Public Information Dissemination by Federal Agencies.. . . 31
2-5. Agency Dissemination of Scientific and Technical Information and
Statistical Data, by Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2-6. Agency Use of Nonpaper Formats for Information Dissemination
by Type of Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2-7. Agency Use of Information Technologies . . . . . . . . . . . . . . . . . . . . . . . . 34
2-8. Agency Prototyping or Testing of Advanced Technologies . . . . . . . . . 35
2-9. Federal Agency Use of Institutional Mechanisms for Information
Dissemination, by Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2-10. Approximate Distribution Volume, Fiscal Year 1987 . . . . . . . . . . . . . . 39
2-11. GPO/SupDocs and NTIS Sales Distribution, Fiscal Year 1987. . . . . . 39
2-12. Estimated Use of Depository Libraries, Fiscal Year 1985 . . . . . . . . . . 39
2-13. Federal Civilian Departmental Agency Evaluation of
Information Dissemination Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2-14. Federal Civilian Departmental Evaluations of GPO Services . . . . . . . 40
2-15. Federal Publishers Committee Survey of GPO Services,
Selected Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Chapter 2

Overview of Federal Information


Dissemination

SUMMARY
Information dissemination is a significant seminated electronically. Paper is still by far
function of the Federal Government, account- the dominant format (accounting for 80 to 90
ing for an estimated $6 billion per year in an- percent of total information products), but sig-
nual expenditures for relevant executive nificant agency use of some electronic formats
agency activities (including information clear- is already occurring for some purposes. For ex-
inghouse operations, printing and publishing, ample, statistical data are highly suited to elec-
library operations, and related research, devel- tronic formats, and, based on results of the
opment, and testing). This estimate does not General Accounting Office (GAO) survey,
include expenditures for the collection and de- about one-third of the civilian agencies use
velopment of the information disseminated, or magnetic tape or disks, one-fifth floppy disks
even a prorated portion of expenditures for and electronic data transfer, and one-tenth elec-
basic agency automation and information tech- tronic mail for dissemination of statistical
nology procurement. data. By comparison, about two-thirds of the
agencies use paper and roughly one-tenth use
The primary Federal mechanisms for infor- microfiche for disseminating statistical data.
mation dissemination are the Federal agencies
themselves; the U.S. Government Printing Of- Many Federal agencies have taken initia-
fice (GPO), which includes about 5 percent of tives with respect to the use of electronic
agency publications in the GPO Superinten- information technologies for information dis-
dent of Documents Sales Program and roughly semination. Electronic technologies have pen-
one-half of agency publications in the Deposi- etrated the majority of agencies in every aspect
tory Library Program (DLP); the National of the information process. The GAO survey
Technical Information Service (NTIS), which results suggest roughly one-half to two-thirds
sells scientific and technical documents pro- of the civilian agencies make at least some use
vided by the agencies; the Consumer Informa- of floppy disks, magnetic tapes or disks, elec-
tion Center (CIC), which distributes free or low- tronic data transfer, and electronic mail for
cost consumer pamphlets for the agencies; and information collection/filing and dissemina-
various private sector vendors operating un- tion. About one-third of the agencies have desk-
der government contract. Federal information top publishing systems, roughly one-half have
is also disseminated by numerous intermedi- electronic photocomposition capability, and
ary mechanisms, such as the press, libraries, roughly one-quarter have electronic publish-
and commercial vendors who, on their own ini- ing systems.
tiative, enhance and/or resell government in- A key characteristic of the current Federal in-
formation. formation infrastructure is that while Federal
The number of civilian agency publications agencies and private companies disseminate Fed-
in paper format appears to be declining slowly, eral information in paper and, increasingly, elec-
while the number of publications in electronic tronic formats, the central governmentwide dis-
format has more than tripled over the past 4 semination mechanisms (GPO/SupDocs, NTIS,
years. Civilian agencies reported, as of fiscal DLP, CIC) are presently limited largely to pa-
year 1987, over 7,500 information products dis- per (or paper and microfiche).

27
28

Evaluating agency satisfaction with the vari- higher in cost. Commercial vendors were rated
ous dissemination channels is difficult. Avail- about the same as the agency. With respect
able survey data for dissemination of paper to GPO, there appears to be overall agency
formats are subjective in nature. Not surpris- satisfaction with respect to traditional ink-on-
ingly, the civilian agencies rated their own dis- paper composition, printing, and binding,
semination services as generally of high qual- However, there is continuing dissatisfaction
ity, timely, and moderate to low in cost. among some agencies with respect to GPO
Agencies rated GPO slightly lower in timeli- cost, timeliness, estimating and billing proce-
ness and slightly higher in cost, and NTIS dures, and marketing/distribution of printed
somewhat lower in quality and timeliness and products.

INTRODUCTION
The Federal Government today stands at a Capturing the full benefits of these technol-
major crossroads with respect to numerous pol- ogies involves consideration of a wide range
icy, oversight, and operational aspects of Fed- of Federal policy, oversight, and operational
eral information dissemination. Advances in questions as they relate to information dissem-
information technology over the past decade, ination. In order to assess this broad topic, the
and especially in the past few years, have Office of Technology Assessment (OTA) com-
opened up many new opportunities for infor- missioned a series of staff and contractor re-
mation dissemination-for all segments of search papers, sought related studies and
American society. Each year the private com- information from various executive and legis-
mercial sector generates literally thousands of lative branch agencies, and drew on the results
new information technology-based products of an extensive GAO survey of Federal agency
and services (including hardware, software, practices and plans.
and application packages), many of which are
This chapter provides a technological and
currently or potentially applicable to Federal
institutional overview of Federal information
information dissemination.
dissemination. The chapter addresses the fol-
Over the past several years, technological lowing specific areas:
applications such as optical disks, electronic ● the size and scope of the current Federal
mail and bulletin boards, electronic and desk-
information dissemination enterprise;
top publishing, electronic printing on demand,
● the technological initiatives already un-
and the like have become technologically fea-
derway in Federal agencies; and
sible and economically viable for widespread
● the institutional bases for Federal infor-
application in the Federal Government as well
mation dissemination.
as the private sector. The vast majority of Fed-
eral agencies are experimenting with some of Each of these is discussed below. This over-
these technologies, and some agencies are al- all picture of the Federal information dissemi-
ready implementing major operational appli- nation enterprise provides an important part
cations. of the context for the rest of this report.

SIZE AND SCOPE OF FEDERAL INFORMATION


DISSEMINATION ENTERPRISE
For purposes of this study, OTA defined missions and considered “public” (legally avail-
“Federal information” as information collected able to the public and not subject to exemp-
and/or developed by the Federal Government tions under the Freedom of Information Act,
to carry out government functions and agency such as law enforcement, investigative, pro-
29

prietary, and classified information). Such pub- fer payments), or about twice the total figure
lic information runs the gamut from statisti- reported to GAO. Agencies vary widely in the
cal data and computer models, to reports, budget percentage reported to be allocated to
periodicals, and directories, to rules, regula- information dissemination, and many well ex-
tions, and circulars, to maps, charts, and pho- ceed the one percent level, as illustrated in Ta-
tographs. Also, OTA included most formats ble 2-2.
of Federal information in the scope of study Also, these estimates do not include the costs
—including paper, microforms, and electronic.
of dissemination of technical information for
Estimating the magnitude of Federal infor- weapon systems and other applications in De-
mation dissemination activities is difficult at partment of Defense (DoD), which are largely
best. There are no credible prior estimates and sensitive or classified in nature. Nor do these
only very rough estimates can be made, since estimates include expenditures for the collec-
there is no systematic reporting of budget and tion and development of the information dis-
activity data for Federal information dissem- seminated, or even a prorated portion of ex-
ination. penditures for basic agency automation and
information technology procurement. And these
Based on the GAO survey results, with 173 estimates do not include the cost of federally
agency components responding, the minimum funded research, development, or other activ-
dollar amounts spent by the Federal Govern- ities on which a significant portion of the in-
ment (civilian and military) in fiscal year 1983 formation collection, development, and/or dis-
and fiscal year 1987 for relevant activities are semination was based.
shown in Table 2-1.
The GAO results provide a rough profile of
The total of about $3.2 billion in reported the number of information dissemination activ-
fiscal year 1987 expenditures is undoubtedly ities. The data are presented in Table 2-3 for
conservative. Inspection of individual agency fiscal year 1983 and fiscal year 1987, with a
responses indicates that many agencies did not breakdown for DoD, civilian departments, and
provide complete responses because they did civilian independent agencies. Again, due to
not have and/or could not estimate relevant incomplete reporting from various agencies,
expenditures. Based on examination of se- these numbers must be considered as minimum
lected agency responses that appear to be espe- estimates of activity levels. For example, GPO
cially well done, it appears that about one per- reports that about 58,000 titles were distrib-
cent of agency budgets on the average are uted to depository libraries in fiscal year 1987,
devoted to information dissemination, which or about 40 percent more than reported by the
would translate into about $6 billion (1 percent agencies to GAO. However, assuming a ran-
of the roughly $600 billion Federal budget, ex- dom distribution of errors, the general trends
cluding interest on the national debt and trans- portrayed should be reasonably accurate.
The data suggest the following conclusions
Table 2-1 .—Federal Expenditures on Information
about the Federal information dissemination
Dissemination, Civilian and Military
(in billions of dollars) enterprise:

Fiscal year Fiscal year


DoD accounts for the largest share of to-
1983 1987

tal Federal Government publications, with
Agency information clearinghouse about 82 percent of the titles and 96 per-
operations . . . . . . . . . . . . $1.500 $1.70 cent of the pages (originals, not copies) as
Agency printing and publishing . . . 0.900 1,10
of fiscid year 1987.
Agency library operations . . 0.200 0.30 However, an insignificant percentage (less
Agency research, development and
testing on information
than 1 percent) of DoD publications are
dissemination ., ... . . . . . . 0.005 0.05 sold by GPO or included in the DLP. This
$2.605 $3.15 may be explained in part because many
SOURCE GAO Survey of Federal Agencies, 1987 of these documents are considered to be
30

Table 2-2.—illustrative Agency Expenditures for Information Dissemination, Fiscal Year 1987

Information dissemination budget (in millions of dollars)


Total Research, Printing Information
agency development and Library clearing- Totals
Agency budget and testing publishing operations house d o l l a r s a p e r c e n tb
Library of Congress . . . . . . . . . . 239.3 0,6 4.1 — — 4.7 2.0
US Navy . . . . . . . . . . . . . . . . . . . . 86,584.4 4.6 207.8 40.5 0.2 253.1 0.3
Economic Research
Services (USDA). . . . . . . . . . . . 44.0 1.1 0.1 0.05 1.25 2.8
Patent and Trademark
Office (DOC). . . . . . . . . . . . . . . 255.8 32.5 18.3 4.7 — 55.5 21.7
Natn’1 Bureau of
Standards (DOC) . . . . . . . . . . . 224.8 0.2 0.8 1.7 — 2.7 1.2
Natn’1 Oceanic and
Atmospheric Admin. (DOC) . . 1,113.1 56.8 1.8 — 58.6 5.3
Bureau of the Census
(DOC) . . . . . . . . . . . . . . . . . . . . 363.1 7.9 1.0 — 8.9 2.5
US Geological Survey
(DOI). . . . . . . . . . . . . . . . . . . . . . 632.4 0.5 14.6 3.2 1.2 19.5 3.1
Federal Elections
Commission . . . . . . . . . . . . . . . 12.9 0.4 0.2 0.4 1.0 7.8
Federal Energy Regulatory
Commission . . . . . . . . . . . . . . . 101.5 2.4 0.5 0.6 3.5 3.4
Federal Trade
Commission . . . . . . . . . . . . . . . . 65.0 0.6 1.2 — 1.8 2.8
Securities Exchange
Commission . . . . . . . . . . . . . . . 114.5 6.4 1.1 0.6 — 8.1 7.1
‘Total agency expenditure for information dissemination activities
b
Agency Information dissemination expenditures as a percentage of the total agency budget
SOURCE: GAO Survey of Federal Agencies, 1987

Table 2.3.—Selected Federal Agency Information Dissemination Activities, Fiscal Years 1983 and 1987

Fiscal year 1983 Fiscal year 1987


DOD a DEPb INDC DOD DEP IND
Publications printed
Number of titles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339Kd 60K 29K 334K 54K 20K
Number of pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93M e 4.2M 0.63M 93M 3.7M 0.55M
Printed publications accepted into GPO’s saies program
Number of titles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 3.6K 1.2K 295 2.8K 0.9K
Number of pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80K 435K 182K 72K 277K 105K
Printed publications inciuded in Federai Depository
Library Program
Number of titles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 762 38K 2.7K 776 36.5K 3.6K
Number of pages ., . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 K 7.1 M 0.27M 110K 7.7M 0.26M
information products disseminated eiectronicaiiy
Number of titles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 1,461 1,001 307 6,261 1,521
aDoD = Depaflment of Defense Agency components
b
DEP = Civilian departmental agency components
clND = civilian independent agency COrnPOnefItS
‘K = thousands
‘M = milllons
SOURCE GAO Survey of Federal Agenctes, 1987
—— ——.

31

sensitive and/or to have very narrow and departmental components and 48 civilian in-
limited demand. dependent agencies, the profile is shown in Ta-
● Of the civilian departmental and inde- ble 2-4.
pendent agency publications (totalling at
The formats currently used for Federal in-
least 74,000), about 5 percent are sold by
formation dissemination cover the entire spec-
GPO and about half (54 percent) are in-
trum. Paper is still by far the dominant for-
cluded in the DLP as of fiscal year 1987.
mat. However, significant use of some
While the number of DoD publications (ti-
electronic formats has already occurred. For
tles and pages) has remained roughly con-
the 114 civilian departmental agency compo-
stant over the past 4 years (fiscal years
nents and 48 civilian independent agencies
1983-1987), the number of civilian agency
reporting to GAO, nonpaper formats are used
publication titles has declined by about
most extensively for dissemination of scien-
17 percent and the number of pages by
tific and technical information and for statis-
about 12 percent. This appears to be
tical information, as indicated in Table 2-5.
paralleled by even a larger decline in the
number of titles accepted into the GPO The use of nonpaper formats is also occur-
sales program (down about 23 percent). ring, although on a more selective and limited
GPO reports that the total number of ti- basis, for certain other types of information.
tles in the sales program increased from The uses of nonpaper formats reported by more
17,513 in fiscal year 1983 to 26,123 in fis- than 5 percent of the civilian departmental
cal year 1987 (up 49 percent). But this in- agencies responding are shown in Table 2-6,
cludes periodicals, forms, carryover doc- by type of information.
uments, and the like in addition to current In sum, Federal information dissemination has
year publications, and is not necessarily already begun the transition to significant use
inconsistent.
The number of titles in the DLP appears
to have remained roughly constant over Table 2.4.—Types of Public Information Dissemination
the past 4 years, with the number of pages by Federal Agencies
showing a modest increase (about 8 per-
Percent of agencies responding
cent). The Depository Program includes,
Departmental Independent
as of fiscal year 1987, about one order of Types of public information agencies agencies
magnitude (10 times) greater number of Pamphlets/bulletins . . . . . . . . 82 94
titles than are available from the GPO Press releases. . . . . . . . . . . . . 79 94
sales program. Statistical data . . . . . . . . . . . . 75 75
Directories/catalogs/
s The number of information products dis- bibliographies . . . . . . . . . . . 69 83
seminated electronically appears to have Manuals . . . . . . . . . . . . . . . . . . 64 67
increased dramatically over the past 4 Scientific and technical
information . . . . . . . 63 65
years, by about 200 percent for DoD, 300 Contractual specs/
percent for the civilian departments, and documents. . . . . . . . . . . . . . 63 83
50 percent for the civilian independent Administrative reports ., ., 62 88
Rules, regulations,
agencies. The estimated total number of directives, circulars . . . . . 62 85
civilian agency electronic information Maps, charts, photos . . . . . 54 50
products for fiscal year 1987 was 7,782, Decisions/opinions. ., . . . 46 71
Professional journals/
up from 2,462 in fiscal year 1983. proceedings. . . . . . . . . . ., 45 54
Laws/statutes . . . . . . . . . . . . . 41 44
The scope of Federal information dissemi- Software products . . . . . . . 30 25
nation cuts across all types of public informat- Satellite imaaetvldata . . . . . . 6 6
ion. As reported to GAO by 114 civilian SOURCE GAO Survey of Federal Agencies, 1987
Table 2-5.-Agency Dissemination of Scientific and Table 2-6.—Agency Use of Nonpaper Formats for
Technical Information and Statistical Data, by Format Information Dissemination by Type of Information

Dissemination of scientific and technical information Percent of agencies


Type of information responding
Percent of agencies
responding Administrative reports
Electronic mail. . . . . . . . . . . . . . . . . . . 14
Departmental Independent Electronic data transfer . . . . . . . . . . . 12
Format agencies aaencies Floppy disk. . . . . . . . . . . . . . . . . . . . . . 8
Paper . . . . . . . . . . . . . . . . . . . . . 61 65 Magnetic tape/disk . . . . . . . . . . . . . . . 6
Microfiche . . . . . . . . . . . . . . . . 19 21 Microfiche. . . . . . . . . . . . . . . . . . . . . . . 5
Electronic data transfer . . . . . 15 10 Pamphlets/bulletins
Magnetic tape/ disk . . . . . . . . 14 13 Microfiche. . . . . . . . . . . . . . . . . . . . . . . 10
Videotape ., . . . . . . . . . . . . . . . 13 8 Electronic mail . . . . . . . . . . . . . . . . . . . 9
Floppy disk. . . . . . . . . . . . . . . . 9 10
Microfilm . . . . . . . . . . . . . . . . . 8 6 Press releases
Film . . . . . . . . . . . . . . . . . . . . . . 8 4 Electronic mail . . . . . . . . . . . . . . . . . . . 13
Electronic mail. . . . . . . . . . . . . 6 8 Electronic data transfer ., ... , , ... , 7
Electronic bulletin board . . . . 6 2 Videotape . . . . . . . . . . . . . . . . . . . . . . . 6
Videodisk . . . . . . . . . . . . . . . . . 2 — Electronic bulletin board . . . . . . . . . . 5
Directories/catalogs/bib/iographics
Dissemination of statistical data Microfiche. . . . . . . . . . . . . . . . . . . . . . . 11
Percent of agencies Electronic data transfer . . . . . . . . . . . 9
respond i ng Magnetic tape/disk . . . . . . . . . . . . . . . 9
Paper . . . . . . . . . . . . . . . . . . . . . 73 75 Floppy disk. . . . . . . . . . . . . . . . . . . . . . 5
Magnetic/tape/disk . . . . . . . . . 32 29 Manuals
Floppy disk. . . . . . . . . . . . . . . . 19 17 Floppy disk. . . . . . . . . . . . . . . . . . . . . . 5
Electronic data transfer . . . . . 18 10 Contractual specs/documents
Microfiche ., . . . . . . . . . . . . . . 12 13 Floppy disk. . . . . . . . . . . . . . . . . . . . . . 5
Electronic mail. ... , . . . . . . . . 8 8
Microfilm . . . . . . . . . . . . . . . . . 5 13 Rules, regulations, directives,
Electronic bulletin board ., , . 4 4 circulars
Videotape . . . . . . . . . . . . . . . . . 2 — Electronic mail . . . . . . . . . . . . . . . . . . . 9
Film . . . . . . . . . . . . . . . . . . . . . . 1 — Floppy disk. . . . . . . . . . . . . . . . . . . . . . 6
SOURCE GAO Survey of Federal Agencies, 1987. Maps, charts, photos
Film ... , . . . . . . . . . . . . . . . . . . . . . . . . 8
Software products
Magnetic tape/disk . . . . . . . . . . . . . . . 18
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 17
of electronic formats. While paper is still domi- Electronic data transfer ., . . . . . . . . . 6
SOURCE GAO Survey of Federal Agencies, 1987
nant, it appears that electronic formats are al-
ready used more frequently than microfilm or
microfiche for many types of information. While statistical data, directories, bibliographies) are
microform still has important archival benefits, also those in which new technologies, such as
the sectors in which microfiche is used relatively compact optical disks, offer the greatest po-
heavily (e.g., scientific and technical information, tential.

TECHNOLOGICAL INITIATIVES BY FEDERAL AGENCIES


Many Federal agencies have taken initia- tion dissemination. Collectively, agencies
tives with respect to the use of electronic in- reported to GAO that this expenditure in-
formation technologies for Federal information creased from $5 million to $50 million between
dissemination and related activities. The num- fiscal year 1983 and fiscal year 1987. This dol-
ber and scope of these initiatives have grown lar amount is undoubtedly low, since many
dramatically over the past 4 years. One indi- agencies did not report or reported incom-
cator is the amount of agency spending for re- pletely on this item. If DoD is included, the
search, development, and testing on informa- dollar amounts are low by at least an order of
33

magnitude, based on separate DoD estimates. absolute number of each technology in use. For
For example, the DoD Computer-Aided Acqui- example, 34 percent of civilian departmental
sition and Logistics Program (CALS) alone is agencies report use of desktop publishing, but
spending on the orderof$150 million per year. the survey instrument did not ask nor did the
The primary focus of CALS is on weapon sys- agencies provide, the absolute number of desk-
tem technical data (including technical docu- top publishing systems. Nonetheless, the qual-
ments such as engineering drawings and speci- itative penetration levels of these technologies
fications developed in support of weapon are, overall, far greater than indicated in any
systems acquisition), much of which is sensi- known prior survey.
tive or classified. However, the magnitude of It is also noteworthy that significant percent-
increase is probably accurate—a roughly 1,000
ages of civilian departmental agencies are cur-
percent cumulative increase over the past 4 rently prototyping or pilot testing advanced
years. There is, at present, no reporting sys-
technologies for information storage and dis-
tem in DoD or the civilian agencies that sys-
semination including those listed in Table 2-8.
tematically collects relevant expenditure or Also, it appears that about half of the civilian
activity data. departmental components will soon have desk-
The GAO survey results provide a remark- top publishing systems (34 percent already have
able picture of agency operational use of elec- operational capability, and another 15 percent
tronic information technologies for informa- are prototyping or pilot-testing), about one-half
tion dissemination. OTA has relied primarily will soon have electronic photocomposition ca-
on the GAO survey results for the civilian de- pability (44 percent now, plus 8 percent in pro-
partmental agencies as being the most repre- totyping or pilot-testing), and about one-third
sentative. As discussed in chapter 1, the de- will soon have electronic publishing systems
fense agencies did not circulate the GAO (21 percent now, plus 11 percent prototyping
survey instrument to many major subcabinet or pilot-testing). For the independent agencies,
agency components, contrary to GAO instruc- more than one half will have desktop publish-
tions; therefore, the defense agency responses ing (29 percent now, with another 31 percent
are likely to be biased by the aggregate re- prototyping or pilot-testing), onehalf already
sponses of the major military departments. On have electronic photocomposition (with another
the other hand, the independent agency 13 percent prototyping or pilot-testing), and
responses are dominated by a large number about two-fifths will have electronic publish-
of small agencies, with a similar result—the ing (31 percent now plus 13 percent in proto-
likelihood of bias in the overall aggregate re- type or pilot testing).
sults. However, both the departmental and in- Where available, quantitative estimates of
dependent agency results are reported where Federal Government use of key technologies
particularly appropriate. For 114 civilian de- are generally consistent with the results of the
partmental agency components and 48 inde- GAO survey. For example, based on all avail-
pendent agencies reporting, electronic technol- able data, OTA estimates that the Federal mi-
ogies have penetrated the majority of agencies crocomputer inventory has increased from a
in every aspect of the information process. The few thousand in 1980 to (conservatively) over
rank order of technologies in operational use 500,()()() today, with a million microcomputers
is listed in Table 2-7. likely within 2 years if current agency procure-
The survey results do not, of course, give the ment plans are fully implemented. The micro-
absolute magnitude of each of the above as a computer is a key component of agency elec-
percentage of total activity. They provide the tronic publishing and dissemination activities.
relative use, and thus may tend to overstate OTA estimates that the Federal agency inven-
actual use. In other words, the survey results tory of high-end electronic laser printers has
indicate the percentages of agencies respond- increased from a handful in 1980 to several
ing that use a specific technology, but not the hundred today, and low-end desktop laser
34

Table 2-7.—Agency Use of Information Technologies

Depart mental Independent


Technology agencies agencies
Information collection/filing
Floppy disk. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 67
Magnetic tape/disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 63
Electronic data transfer (computer to computer). . . . . . . . . 60 56
Electronic mail. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 48
Computerized telephone calls . . . . . . . . . . . . . . . . . . . . . 18 21
Nonpaper storage
Floppy disk. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 73
Magnetic tape/disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 73
Micrographics (microfilm/fiche) . . . . . . . . . . . . . . . . . . . . . . . 54 71
Videodisk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 6
CD-ROM (Compact Disk-Read Only Memory) . . . . . . . . . . . . 4 8
Optical disk (WORM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 4
CD-I (Compact Disk-Interactive) . . . . . . . . . . . . . . . . . . . . . . 0 2
Optical disk-erasable . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 2
Printing
Computer graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 58
Laser and non-impact printing . . . . . . . . . . . . . . . . . . . . . . . . 64 81
Photo-offset printing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 63
Computer-aided page makeup . . . . . . . . . . . . . . . . . . . . . . . . 50 52
Electronic photocomposition . . . . . . . . . . . . . . . . . . . . . . . . 44 50
Desktop publishing systems. . . . . . . . . . . . . . . . . . . . . . . . . . 34 29
Electronic publishing systems . . . . . . . . . . . . . . . . . . . . . . . . 21 31
Microform printing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 29
Electronic dissemination
Floppy disk. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 58
Magnetic tape/disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 60
Electronic data transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 52
Electronic mail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 40
Videotape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 52
Electronic bulletin board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 17
Teleconferencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 33
Film . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 31
Broadcast TV.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 19
Videodisk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 6
One-way cableTV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 10
Videoconferencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 8
Digital cartographic systems . . . . . . . . . . . . . . . . . . . . . . . . 7 2
CD-ROM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2
Selective dissemination of info. . . . . . . . . . . . . . . . . . . . . . 4 8
Expert systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2
Videotext/teletext . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 6
Interactive cable TV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2
CD-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 —
SOURCE GAO Surveyof Federal Agencies, 1987

printers and desk top publishing software have The GAO survey results are generally con-
increased from very few in 1980 to several tens sistent with the results of OTA’s own research
of thousands today. Since a microcomputer, and contractor case studies of selected agen-
laser printer, and software are the major com- cies. For example, all three military services
ponents of a desktop publishing system,OTA (Army, Navy, and Air Force) as well as the
conservatively estimates that there are 30,000 Office of the Secretary of Defense, have ma-
desktop publishing systems and 300 high-end jor electronic publishing and dissemination
electronic publishing systems in the Federal systems under development or in operation.
Government. In the civilian sector, the U.S. Geological Sur-
35

Table 2-8.—Agency Prototyping or Testing of ogy Center, in cooperation with OSD and
Advanced Technologies the Joint Chiefs of Staff. The Center in-
Percent of
cludes four laboratories:
agencies responding 1. Defense Gateway Laboratory, which
Departmental Independent will facilitate electronic access to over
Storage technology agencies agencies 800 diverse DoD, commercial, and Fed-
CD-ROM a . . . . . . . . . . . . . 15 1 0 eral databases via the Defense Gate-
CD-lb . . . . . . . . . . . . . . . . . . . . . 6 2 way Information System, and will uti-
C
WORM . . . . . . . ., . 10 10
lize user-friendly search software along
Dissemination technology
11 10
with an online database catalog;
CD-ROM . .
CD4 . . . . . . . . . . . . . . . . . . . 5 2 2. High-Density Information Systems
Expert systems. . . 7 8 Laboratory, which will develop high-
NOTES
a
Cornpa ct Disk Read Only Memory
density optical disk storage and re-
bCompact Disk Interactive
Cwr,te once Read Mawlmf=
trieval systems with electronic print-
SOURCE GAO Survey of Federal Agencies 1987
ing, publishing, and dissemination ca-
pabilities;
3. Artificial Intelligence/Decision Sup-
vey and Bureau of the Census (among others) port Laboratory, which will explore
are col.laboratingon information dissemination state-of-the-art software for diagnos-
via Compact Disk-Read Only Memory (CD- tics, monitoring, control, and informa-
ROM) and digital cartographic technologies. tion retrieval, and will research the ap-
A capsule description of selected highlights is plication of AI/expert system software
given below. and display techniques to defense in-
● DoD, Office of the Secretary of Defense formation needs, including online inter-
(OSD): DoD is implementing the Com- facing with the Defense Gateway In-
puter-Aided Acquisition and Logistics formation System; and
Support (CALS) program designed as an 4. Interactive Video Laser Disk Systems
integrated system for the creation, stor- Laboratory, which will explore ~nnova-
age, revision, and dissemination of tech- tive disk techniques for training pro-
nical information relevant to weapon sys- spective users of the various high-
tems. CALS is designed to use state-of- technology systems under devel-
the-art electronic publishing technology opment.
and incorporates an extensive set of tech- ● NOAA, National Geophysical Data Center

nical standards for electronic exchange of (NGDC): NGDC has prepared a prototype
information, page markup, graphics, and CD-ROM on selected geomagnetic and
the like. The objective is eventually to con- solar-terrestrial physics data, including
vert current paper flows of information to data on solar flares, sunspots, and wind.
digital electronic flows, so that engineer- NGDC makes this data available to users
ing drawings, technical manuals, logistics at reduced cost (e.g., the disks cost about
records, and life-cycle data are created and $50 each at a volume of 600 copies–
accessed in electronic formats. CALS par- including costs of data preparation, soft-
ticipants include OSD, Army, Navy, Air ware, premastering, mastering, and dupli-
Force, the Defense Logistics Agency, and cation —compared to a cost of about $5OO
the private defense contractors. The for the same data on magnetic tape). The
CALS consolidated budget for DoD is CD-ROM runs on any IBM-PC AT or XT
roughly $150 million per year. or compatible microcomputer with 512
● DoD, Defense Technical Information Cen- kilobyte random access memory, 10 mega-
ter (DTIC): DTIC, a component of the De- byte hard-disk drive, standard floppy-disk
fense Logistics Agency, is implementing drive, and CD-ROM reader and software
a Defense Applied Information Technol- using the High Sierra standard at a total
cost of under $4,000. By comparison, mag- ing data from such Census reports as the
netic tapes require a mainframe or mini- County and City Data Book and County
computer and peripheral equipment at a Business Patterns and, on request, data
total cost of several tens to hundreds of downloaded from magnetic tapes in the
thousands of dollars or more. Census inventory; and magnetic tapes
● DOI, U.S. Geological Survey (USGS): that contain large volumes of Census data,
USGS has prepared a prototype CD-ROM frequently in more detail than is available
on mapping data for the Gulf of Mexico, in the paper publications, and sell for $175
known as Project Gloria. The prototype per tape (6,250 bits per inch). In the fu-
was prepared with NOAA (which devel- ture, CD-ROMs will be used for dissemi-
oped the search software) and the Jet nation of statistical data to microcom-
Propulsion Laboratory (which developed puter users (Census has already prepared
an interactive image display program). prototype disks and envisions a signifi-
The combined software permits the user cant role for CD-ROM for distributing the
to search the database by geographical results of the 1990 census).
mapping areas, latitude, and longitude, and
to display the data in graphic and varia- In the legislative branch, GPO has initiated
ble image formats. USGS views microcom- technology innovation projects in several
puter-based CD-ROM applications as the areas, including dial-up desktop to mainframe
key to dramatically improving access to electronic printing capability, dial-up fiber op-
and reducing the cost of many earth tic links for remote photocomposition, and
science databases maintained by USGS, long-distance electronic data transfer. While
NOAA, NASA, and other Federal agen- GPO disseminates its information products
cies, and, accordingly, has already pur- primarily in paper format (and secondarily mic-
chased CD-ROM premastering equipment. rofiche), the majority of inputs to GPO is al-
● DOC, Bureau of the Census: The Census ready in electronic format. (GPO pilot projects
Bureau offers a full range of products in are discussed in ch. 4, 7, and 8. Other execu-
electronic format in addition to paper and tive agency electronic pilot projects are dis-
microfiche. Electronic formats include: cussed in ch. 3, 5, and 10.)
CENDATA, an online information serv-
ice including press releases, statistical In sum, the current initiatives of the Fed-
summaries, product announcements, and eral Government, taken as a whole, indicate
the like, and available via DIALOG In- a very significant use of advanced information
formation Services, a private vendor; elec- technology. While use varies widely by agency,
tronic bulletin boards that provide instan- and even within agencies, overall the govern-
taneous access to selected census data ment appears to be at or close to the thresh-
(including most CENDATA entries) to old where technology-based electronic informa-
participants in the State Data Center pro- tion dissemination can be a significant and
gram and the Federal-State Cooperative integral part of the Federal information infra-
Estimates Program; floppy disks contain- structure.

INSTITUTIONAL INFRASTRUCTURE FOR FEDERAL


INFORMATION DISSEMINATION
The primary institutional mechanisms used disseminated by numerous intermediary mech-
for Federal information dissemination are the anisms, such as the press, libraries, interest
Federal agencies themselves, GPO, NTIS, Con- groups, congressional offices, and the like.
sumer Information Center (CIC, located in
Pueblo, CO), DLP, and private sector ven- Almost all Federal agencies, and certainly
dors/contractors. Federal information is also all Cabinet departments, have some explicit
37
— —.

statutory authority for information dissemi- DLP is a cooperative program by which


nation and many have multiple statutory au- agency documents, whether or not they are
thorities. Because Federal agencies collect sold via GPO or NTIS, are provided to a net-
and/or develop the bulk of Federal information, work of about 1,400 libraries around the United
they are generally the most knowledgeable about States. Over 50 regional depository libraries
their own information products and services, and receive all documents distributed, while the
frequently are the best informed about the cur- other depository libraries select which types
rent and potential users of that information. of documents or titles they wish to receive.
Many agencies have formal and/or informal DLP was established by public law and is oper-
mechanisms to discuss information needs and ated by GPO. It serves as part of an “infor-
problems with users. According to the GAO mation safety net by which the government
survey results for 114 departmental agency funds the distribution of materials to desig-
components, many agencies directly dissemi- nated libraries. DLP includes all types of Fed-
nate a wide range of types and formats of Fed- eral information, but has been limited, until
eral information. now, to paper and microfiche formats. (See ch.
6 and 7 for further discussion of DLP
GPO, or more precisely the GPO Superin- activities. )
tendent of Documents (SupDocs), is statutorily
authorized to sell selected agency documents Finally, private sector contractors play a role
to the general public. The documents selected in disseminating information for many of the
for the GPO sales program represent only a agencies. Since, in general, government infor-
small fraction (a few percent) of all government mation cannot be copyrighted, numerous pri-
publications, and are ones judged by GPO mar- vate sector vendors repackage, add value, and
keting specialists to have significant demand sell or resell a wide variety of types of Federal
and/or those that by law must be sold to the information in a wide variety of formats. The
public. Documents sold by GPO cover a wide dissemination of agricultural information de-
range of types of Federal information, but the veloped by the U.S. Department of Agricul-
formats are limited primarily to paper and mi- ture (USDA) provides a good illustration of the
crofiche, with a few items available in magnetic private sector role. USDA information is dis-
tape format. (See ch. 4 and 5 for further dis- seminated directly by agency components,
cussion of SupDocs activities. ) through private contractors, via private sec-
NTIS, pursuant to public law, sells scientific tor online gateways, and by private sector
and technical information provided by the value-added providers. For example, the ED I
mission agencies. The types of information (Electronic Dissemination of Information)
products available from NTIS are much more service is provided online on a fee-for-service
limited than those available from the agencies basis by the U.S. Department of Agriculture
or GPO, and are provided to NTIS on a volun- (USDA) through a contract with Martin Mari-
tary basis. NTIS products have very limited etta Corporation. ED I provides time-sensitive
demand (about 10 copies per item) compared daily, weekly, and monthly reports and news
to GPO products. NTIS sells primarily micro- releases from USDA agencies. AGRICOLA, an
fiche and paper formats, with some sales of extensive USDA bibliographic reference data-
magnetic tape and floppy disk formats. (See base on all aspects of agriculture, is provided
ch. 5 for further discussion of NTIS activities. ) online via DIALOG Information Services, a
commercial database vendor. USDA Online,
CIC (operated on a reimbursable basis by a USDA current information service includ-
GPO for the General Services Administration), ing news releases and short reports, is provided
pursuant to public law, primarily facilitates via ITT Dialcom, a commercial online elec-
the distribution of consumer-oriented pamph- tronic mail gateway.
lets and bulletins from the agencies. These ma-
terials are usually short and are available free ED I, AGRICOLA, and USDA Online are in-
or at a small fee. CIC products are limited to formation products developed by USDA, but
paper formats. disseminated online via private vendors or
. . ,—, . ..— ., ., “
38

contractors. There are also many online infor- The four governmentwide dissemination
mation products that make use of USDA in- agencies collectively distribute about 107 mil-
formation, but are developed as well as dissem- lion copies of documents (in paper or micro-
inated by the private sector. For example, fiche format) per year, as summarized in Ta-
Deane Publishing sells AgLine, an online in- ble 2-10.
formation service that covers USDA daily
commodity reports and updates and also offers Of these dissemination agencies, only
electronic mail and software capabilities. Pi- GPO/SupDocs and NTIS maintain customer
oneer Hi-Bred International sells AGRIBUS- profiles. Percentage estimates are shown in Ta-
INESS U. S. A., a comprehensive online data- ble 2-11. The profiles for SupDocs and NTIS
base that indexes agricultural business, trade, are fairly similar, although the use of differ-
and government publications. This database ent categories makes comparisons somewhat
is available via DIALOG Information Serv- difficult. In any event, the largest customer
ices, a commercial vendor. As a final example, group is business. To keep this in perspective,
Vance publishing sells ProNet, an online news consumers are, by definition, the primary cus-
and information service on the produce indus- tomer group for CIC, and the libraries are the
try that incorporates a variety of price, mar- primary DLP customers. Of course, libraries
ket, weather, and related information from largely serve an intermediary role, and the ulti-
USDA and elsewhere. mate customers of DLP are the patrons of the
The GAO survey results indicate that the individual depository libraries. DLP does not
114 civilian departmental agency components at present maintain comprehensive user sta-
responding use several institutional mecha- tistics, although a survey is in progress. How-
nisms for information dissemination with re- ever, a 1985 estimate suggests that over 10
spect to the formats indicated, as shown in Ta- million persons use DLP each year, as detailed
ble 2-9. in Table 2-12, although these estimates have
not been validated.
This highlights one of the key characteris-
tics of the current Federal information infra- The GAO survey attempted to measure
structure: while individual Federal agencies agency satisfaction with the various dissemi-
and private companies disseminate Federal in- nation channels for typical agency reports (i.e.,
formation in paper and electronic formats, the 50-100 pages, paper format, typeset, some
central governmentwide dissemination mech- graphics, specified deadline). There are numer-
anisms are presently limited largely to paper ous problems in interpreting and using these
(or paper and microfiche). Both GPO/SupDocs data. Not surprisingly, the civilian depart-
and NTIS sell a small number of products in mental agencies rated their own dissemination
electronic format, but this represents an insig- services as generally of high quality, timely,
nificant percentage of total sales volume for and moderate to low in cost for paper prod-
either. ucts. This is, of course, a self-assessment, and

Table 2-9.—Federal Agency Use of Institutional Mechanisms for Information Dissemination, by Format

Percent of agencies responding
Electronic
Elect ronic data Magnetic Floppy
Institution Paper Microfiche mail transfer tape/disk disk
Own agency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 11 25 9 40 33
Gpo/Sup Docs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 9 1 1 3
NTIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 23 1 — : 7
CIC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 — — — — —
Depository Libraries . . . . . . . . . . . . . . . . . . . . 52 12 — — 2 1
Private sector vendors/contractors . . . . . . . . . . 48 7 9 3 15 11
SOURCE GAO Survey of Federal Agencies, 1987
39

Table 2“10.—Approximate Distribution Volume, DLP was evaluated as slightly less timely.
Fiscal Year 1987 Commercial vendors were rated about the same
Dissemination Distribution volume as the agency itself. The full comparative data
agency (millions of copies) are shown in Table 2-13 (normalized to 100 per-
GPO/SupDocs (free)a . . . . . . . . . . . . 30 cent) and expressed as a percentage of the
GPO/SupDocs (sales)b ., . . . . . . . . . . 27 agencies responding to each question, based
NTIS (sales ., ., ., . . . . . . . . . . . 6
CIC (free) ., . . . . . . . . . . . . . . 21 on 114 civilian departmental components re-
DLP (free) ., ., ., . . . . . . . . . . . . . . . . 23 sponding.
Total ., . . . . . . . . . . . . . . . . . . . . . . 107
Given the subjective and general nature of
aB law and re
q:hde~c,#:aMe the agency evaluations, these results should
SOURCE US Government Prlntlng Office and National Techn!cal Information
SewIce 1988
be interpreted cautiously. For example, the per-
ceived problems with the quality and timeli-
ness of NTIS documents could be due largely
Table 2”ll.— GPO/SupDocs and NTIS Sales
Distribution, Fiscal Year 1987
to poor quality and late delivery of copies pro-
vided to NTIS by source agencies. Also, the
Percent of total sales
— perception that GPO, NTIS, and DLP dissem-
GPOI ination is less timely than agency dissemina-
SupDocs NTIS tion could reflect the role of GPO and NTIS
Business . 59- 64(U.S. only) – as secondary rather than primary distributors
Private individuals . 27 4 (general
public) of agency documents and the delays inherent
Foreign . ..., . . . —a 20 (business and in a secondary role. And the perception that
government) NTIS documents are more costly than agency,
Federal, State, and
local government . 8 6 GPO, and commercial sources may reflect the
Universities and very low volume of sales per NTIS title (and
c o l l e g e s 6 6 (includes pub- resulting higher per unit costs). Finally, some
lic libraries)
a
GPO foreign customers Included In all other categories as appropriate except
of the agency responses appear to be question-
Government able. For example, the majority of agencies
SOURCE U S Government Prtntlng Off Ice and National Technical Information
Service 1988
rated the cost of DLP dissemination as mod-
erate (rather than low or very low), yet for many
agencies there is no cost for DLP dissemina-
Table 2-12.—Estimated Use of Depository tion. Agencies only pay printing and binding
Libraries, Fiscal Year 1985

costs when they provide copies directly to DLP
Actual a Projected b for documents not produced by or procured
No, Ilbrarles reporting . . . . . . 1,188 -
1,400 from GPO.
Avg. no. weekly users
per library . . . . . . . . . . . 141 141 The GAO survey requested additional evalu-
Total weekly users . . . . 167,508 197,400 ation detail for GPO. The results indicated that
Total annual users . ........8,710,416 10,264,800
a
the majority of the departmental agency com-
Based on I Ibrary estimates, numbers not valldated and may tnclude undercounts,
overcounts, or doublecounts (multlple users per person) ponents responding were satisfied or very
b
Projects the average use based on the 1,188 Ilbrarles reporting to all of the ap-
proximate 1,400 depository Ilbrarles
satisfied with publications layout, composi-
SOURCE U S Government Printing Off Ice and Office of Technology Assessment, tion, printing quality, printing timeliness, bind-
1988
ing, cataloging, marketing/sales, distribution,
and depository library services provided by
its validity cannot be objectively determined GPO. The one area where one-half were neu-
from the survey results. Agencies rated GPO tral (neither satisfied or dissatisfied) or dissat-
slightly lower in timeliness and slightly higher isfied was printing cost. Some dissatisfaction
in cost for paper products relative to agency was also indicated with respect to market-
views of themselves. NTIS was rated by agen- ing/sales, printing timeliness, and distribution,
cies as somewhat lower in quality and timeli- as indicated in Table 2-14 (in normalized per-
ness and higher in cost for paper products. centages), based on the civilian departmental
40

Table 2-13.—Federal Civilian Departmental Agency Evaluation


of Information Dissemination Channels

QUALITY
Percent of agencies responding
Dissemination channel Very high High Moderate Low Very
— low
Agency . . . . . . . . . . . . . . . . . . 29.8 45.2 25.0 - -

GPO . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.9 56.9 19.4 1,4 1-4
NTIS . . . . . . . . . . . . . . 12.5 30,0 30.0 25,0 2.5
Clc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45.0 50.0 5.0 — —
DLP . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19.3 44.2 34.6 — 1.9
Commercial . . . . . . . . . . . . . . . . . . . . . . 22.2 53.7 22,2 1.9 —

TIMELINESS
Percent of agencies responding
Moderate Some Little or no
Dissemination channel Very great Great extent extent extent
Agency . . . . . . . . . . . . . . . . . . . . . . . . . . 27,9 45.2- 23.1 2,9 1.0
GPO . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.6 31.5 45,2 11.0 2.7
NTIS . . . . . . . . . . . . . . . . . . . . . . . . . . — 22.5 57.5 10.0 10.0
Clc . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.1 36.8 31.6 10.5 —
DIP . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.8 25.9 50.0 3.7 5,6
Commercial . . . . . . . . . . . . . . . . . . . . . . 13.0 37.0 40.7 7,4 1.9

COST
Percent of agencies responding
Dissemination channel Very high High Moderate Lo~ Very low
Agency . . . . . . . . . . . . . . . . . . . . . . . . . . 1.9 15.3 63.5 13,5 ‘5.8 -

GPO . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.9 23.3 61.6 8.2 —


NTIS . . . . . . . . . . . . . . . . . . . . . . . . . 22.5 45.0 27.5 2.5 2.5
Clc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 15.0 65.0 5.0 15.0
DLP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.9 7,5 56.6 11.4 22.6
Commercial . . . . . . . . . . . . . . . . . . . . . . — 14.8 79.6 5.6 —
SOURCE GAO Survey of Federal Agencies, 1987

agency components responding. Note that of Table 2-14.—Federal Civilian Departmental


the l14 agency components that participated, Evaluations of GPO Services
the number that actually commented on spe- —
Percent of
cific GPO services ranged from 54 to 91, as agencies responding
indicated inTable 2-14. These 1987 GAO sur- GPOService N o ,a Satisfied Neutral Dissatisfied
vey results can be compared with the results Publications layout . . . 47 ‘78.0 12.0 10.0”/0
of a 1987 survey conducted by the Federal Pub- Composition ...,..,.. 66 81.5 10.8 7,7
lishers Committee(FPC~ an interagency group Printing quality . . . . . . 91 80.0 12.2 7.8
Printing timeliness . . . 91 66.7 17.8 15.6
of printing, publishing, and public information Printing cost . . . . . . . . 87 50.0 36.0 14.0
officials, and a 1983 survey conducted by GPO Binding . . . . . . . . . . . 82 74,1 19,7 6.2
itself. While the FPC survey included many Cataloging . . . . . . . . 39 73,0 24,3 2.7
Marketing/sales . . . . . 57 57.9 26.3 15.8
other topics, it did cover several cost, timeli- Distribution . . . . . . 67 67.2 21,9 10.9
ness, and marketing/distribution topics, with Depository library . . . . 54 78.9 19,3 1.8
a
the results indicated in Table 2-15. The FPC Numberof agency components commenting on each GPO service
SOURCE GAO Surveyof Federal Agencies, 1987
survey results must be interpreted cautiously
since the overall response rate was only about
10percent (48respondents out of the475per- ture, Commerce, Defense, Energy, Health and
sons who were sent the questionnaire). FPC Human Services, Housing and Urban Devel-
has noted thatthe48respondents included offi- opment, Interior, Justice, and Labor) and
cials from nine cabinet departments (Agricul- about a dozen independent agencies. FPC has
41

Table 2-15. —Federal Publishers Committee Survey 4, failure to provide accurate cost estimates
of GPO Services, Selected Results in a timely manner;
Number of respondents
5. improper or unclear levying of surcharges;
Needs
and
Area of concern Adequate I m provement 6. improper packaging, labeling, or deliver-
Bllllng delays and ing of jobs by GPO contractors.
discrepancies. ., ., . 8 29
Cost of GPO lnhouse work ., 9 7 This survey is, of course, dated, and GPO has
Delivery date reliability. 14 22 not conducted a similar followup survey.
Quallty, timeliness, and cost
controls of GPO contractors 11 25 In addition to being 5 years old, the 1983
Accuracy and adequacy of
S U PD O C S s a l e s i n f o r m a t i o n 1 3 11
GPO survey has been criticized because it was
SOURCE Fed2rd\ Publishers Committee Survey 1987 based on the opinions and perceptions of
GPO’s customer agencies. The GPO Inspec-
tor General did not attempt to validate the re-
submitted the complete survey results and re- sponses by checking records or seeking cor-
lated recommendations to GPO for comment roboration from multiple sources within a
and followup action where appropriate. ] given agency. However, the 1987 GAO and
In 1983, the GPO Inspector General con- FPC surveys are subject to this same criticism.
ducted an audit of customer satisfaction with Taking all information into account, there
GPO services, based on a questionnaire sent appears to be overall satisfaction with GPO
to agency customers. The response rate was services with respect to traditional ink-on-
over 90 percent, with 125 out of 136 agencies paper composition, printing, and binding, but
completing the questionnaire. Six areas ap- continuing dissatisfaction among some agen-
peared to be of greatest concern to customers, cies with respect to cost, timeliness, estimat-
with 38 to 70 percent of the respondents dis- ing and billing procedures, and, possibly, mar-
satisfied at least some of the time with regard keting/distribution of printed products. GPO
to: has instituted improvements in its customer
1. lack of advance notice to agencies when service operations in recent years. And FPC
due dates slip; has acknowledged that GPO “has greatly in-
2. failure to complete jobs on time; creased its responsiveness to agency needs;
3. failure to bill jobs in a timely manner; but FPC “is not satisfied that many of the long-
standing problems are being resolved. Op-
portunities for further improvement are con-
‘Nlemorandum from <John 11. hlounts, Chairman, Federal Pub-
lishers (’omrnittee, to Ralph E;. Kennickell, ,Jr., I’uhlic Print~r, sidered in chapters 4 and 11.
on ‘‘ Recommendations from Federal I’uhlishers C’ommit tee to
(;oiernment Printing Off ice,” hla~’ 13, 1 Wfi. ‘Ibid.
Chapter 3

Key Technology Trends


Relevant to Federal
Information Dissemination

i.
Clockwise from top left: National Library of Medicine CD-ROM Disk (phott )
credit: Doug Jones, National Library of Medicine); satellite (photo credit
USA Today, all rights reserved); and gallery of Bureau of the Census dati
products (photo credit: Neil Tillman, Bureau of the Census).
CONTENTS

summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Introduction and Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Information Systems Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
The Microcomputer Revolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
The Continuing Role of Paper and Microform . . . . . . . . . . . . . . . . . . . . . . . 49
Electronic Publishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Desktop Publishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
High-End Electronic Publishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Electronic Forms Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Computer Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Scanners and Printers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Online Information Dissemination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Online Information Retrieval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Telecommunications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Electronic Mail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Optical Disks.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Expert Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Technical Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Chapter 3

Key Technology Trends Relevant to


Federal Information Dissemination

SUMMARY
The results of the General Accounting Of- vide the technological underpinning for
fice (GAO) surveys of Federal information numerous information collection, process-
users (see chapter 4) and prior studies on the ing, and dissemination activities;
● continued, rapid proliferation of desktop
future of paper and paper-based media (such
as books) indicate that paper is expected to publishing systems, comprised of a
decline only marginally as a preferred format microcomputer, nonimpact printer, and
in the next few years, although this decline page composition software (and some-
could become significant for specific types of times a scanner for paper input) in the
information (e.g., bibliographic, reference, sta- most basic configuration;
● continued, rapid improvement in the
tistical, scientific, and technical) that are highly
suited to electronic access and manipulation. power of desktop publishing software to
The outlook for microform is less favorable. handle more complex documents, formats,
Microfilm is currently used very little for Fed- fonts, and the like;
eral information dissemination; microfiche, ● continued, steady improvement in the

while used extensively, is expected to decline price/performance of nonimpact printers,


significantly as a preferred format, except for with low-cost desktop printers offering
records storage and archival purposes. output quality acceptable for most docu-
ments, and high-end printers offering
In sharp contrast, the GAO surveys of Fed- quality comparable to some phototype-
eral agencies (see ch. 2) and Federal informa- setters;
tion users (ch. 4) indicate that plans and prefer- ● similar improvement in the price/perform-
ences for dissemination in electronic formats ance of scanners, with the capability of
(e.g., electronic mail and bulletin boards, opti- high-end scanners (to handle a wide range
cal disks) are projected to increase dramat- of type styles and sizes) migrating to desk-
ically. top scanners;
, as a combined result of the above trends,
This chapter surveys a number of major tech-
nologies and key technical trends relevant to overall continued improvement in the abil-
Federal information dissemination. Several ity of desktop systems to produce higher
key technical trends are expected to continue quality, more complex documents, thus
conservatively for 3 to 5 years and in many further reducing the gap between desktop
cases for at least 10 years, and are combining and high-end electronic publishing and
in such a way that most of these plans and phototypesetting systems;
c for complex, large-volume, and/or large in-
preferences are likely to become reality. These
trends include: stitutional applications, continued im-
provement in high-end electronic publish-
● continued steady improvement in the ing characterized by:
price/performance of microcomputers, –declining cost of software and work-
which already bring the power of main- stations;
frame computers to the desktop at the —increasingly heavy competition between
cost of a stereo set; microcomputers pro- desktop and high-end systems;

45
46

—rapidly growing networking of desktop remaining standards issues are re-


and high-end systems; solved;
—rapidly growing networking of worksta- —maturation of Write Once Read Many-
tions with high-end nonimpact printers times (WORM) and erasable optical
and phototypesetters; disks (compact and full size) as technol-
—increasingly heavy competition be- ogy stabilizes and standards are estab-
tween and among software, worksta- lished;
tion, phototypesetter, and computer —emergence of Compact Disk-Interactive
equipment vendors, as well as sys- (CD-I) and other optical disk appli-
tems integrators and service bureaus cations;
and; ● rapid advances in development of expert
—continued migration of electronic pub- systems applicable to many aspects of in-
lishing to other applications such as formation dissemination-including tech-
forms management and multi-format nical writing, indexing, information re-
output; trieval, and printing management; and
● continued, rapid increase in the number ● continued, steady progress in develop-
and use of computerized online informa- ment and acceptance of standards for tele-
tion services, especially for information communication, data transfer, optical
search and retrieval, electronic display, disks, and page description and text
and remote printing-on-demand when markup.
needed;
The net, cumulative effect of these techni-
● continued, steady increase in the number cal trends is to afford the Federal Government
of online information gateways that pro-
the opportunity to realize the kind of signifi-
vide the channels for electronic informa-
cant performance improvements and cost re-
tion exchange (such as electronic data
ductions that have been demonstrated in the
transfer, mail, facsimile, and bulletin
private sector. Also, the convergence of these
boards), but not the information itself;
technical trends, along with progress in
these gateways include common carriers standards-setting, makes information systems
(interexchange and bell operating compa-
integration a real possibility for the Federal
nies), value-added companies, and non-
Government and other users. Systems integra-
profit and governmental systems;
tion permits the coupling of input, storage,
● continued advances in the telecommuni-
processing, and output technologies in ways
cation technologies that underlie online in-
that permit multi-media (e.g., paper, micro-
formation services and gateways, includ- form, online electronic, and stored electronic)
ing packet switching, fiber optics, satellite
dissemination from the same electronic data-
networking, FM subcarrier transmission,
base. In essence, the key technologies and tech-
and integrated switched digital systems;
nical trends highlighted above are central to
● rapid advances in optical disk technologies
the emerging movement towards systems in-
and applications, especially for purposes
tegration.
of information storage and dissemination;
advances include:
—accelerating penetration of Compact
Disk-Read Only Memory (CD-ROM) as
47

INTRODUCTION AND OVERVIEW


A major objective of this study is to iden- vided, but for illustrative purposes only. Some
tify and discuss new or evolving ways in which cost and performance data also are included for
information technology can or might be applied illustrative purposes only. These data change
by the U.S. Government Printing Office (GPO), rapidly, and should be checked with vendors if
National Technical Information Service product or service-specific evaluation is contem-
(NTIS), and other Federal agencies to the dis- plated. Also, the presentation is geared to the
semination of Federal information. An impor- level of the informed lay person, not to the tech-
tant step in this process is the identification nical specialist. For discussion of specialized
of key technology trends that are relevant to information technologies not included here
information dissemination. (e.g., digital data tapes, digital cartographic
systems), the reader should consult a forthcom-
OTA has surveyed a representative cross-
ing OTA staff paper on Federal Scientific and
section of major technologies relevant to in- Technical Information Dissemination. Finally,
formation dissemination. The presentation in
for the discussion of telecommunication tech-
this chapter emphasizes electronic technol-
nologies not covered here (e.g., digital facsi-
ogies, although paper and microform are dis-
mile, videotext, cable television), the reader
cussed briefly. Conventional ink-on-paper
should consult OTA’S forthcoming report on
printing technologies, including prepress and
Commumnications Systems for An information
binding, are considered in chapter 4 in the con- Age (1989).
text of alternative futures for GPO. As dis-
cussed in chapter 4, GPO has already upgraded
its conventional printing technology to a level Information Systems Integration
comparable to most of the private printing in-
dustry, However, GPO has much less experi- One important finding is that a combination
ence with online information systems, expert of technological advances, cost reductions, and
systems, optical disks, and high-end electronic current or pending standards has opened up
for the first time a real possibility of informa-
publishing. NTIS is in a similar situation (see
ch. 5), as is the Depository Library Program tion systems integration in the Federal Gov-
(DLP). Libraries in general, especially the ernment. The importance of this development
larger research libraries, have more experience cannot be overstated. Until recently, the Fed-
with electronic systems (see ch. 6). Thus, this eral Government, along with other major in-
chapter is highly relevant to general consider- formation system users, had no choice but to
ation of future technological directions for obtain an essentially incompatible complement
GPO, NTIS, and DLP, as well as the Federal of information technology hardware and soft-
mission agencies. ware, such that system integration across the
government and major agencies was very dif-
This chapter emphasizes significant current ficult, if not impossible. Integration of specific
or emerging technical trends that are expected systems within agencies was possible. But
to persist for at least 3 to 5 years into the fu- even here, major integration efforts, for exam-
ture. In many instances, the key trends are ple in the Department of Defense, still led to
likely to continue even longer–perhaps for 10 numerous incompatible systems.
years or more. In the aggregate, the trends pro-
vide what OTA believes is a reliable overall The relevance to Federal information dissem-
technology planning framework for Federal in- ination is immediate and direct. It is now pos-
formation dissemination. However, the pres- sible to conceive of, plan for, design, and
entation in this chapter is not intended to be implement an integrated information dissem-
used in the evaluation and selection of specific ination system (or more likely a network or hi-
equipment and systems. Some examples of erarchy of systems) for the Federal Govern-
equipment, vendors, and/or applications are pro- ment. This is possible because of advances in
48

a whole range of relevant technologies-includ- dissemination environment that includes the


ing input, storage, processing, and output following illustrative elements:
technologies-that can deal with the entire
range of media, including paper, microform, document/data entry (e.g., scanning, word
magnetic disk, optical disk, and direct elec- processing, facsimile);
document revision/composition (e.g., elec-
tronic. And the cost/performance trends in
tronic publishing-desktop and high-end,
these technologies are likely to make a wide
range of applications cost-effective when com- computer graphics);
pared to conventional methods. document storage (e.g., electronic data-
base, optical disk);
Two other related trends are equally impor- document output (e.g., electronic publish-
tant. One is the trend toward standards for ing, laser printing, photo offset); and
systems interconnection at the hardware, soft- document distribution (e.g., optical disk,
ware, and applications levels. There is strong electronic mail, computer diskette, online
movement among the vendor and user com- electronic, paper copies, microform).
munities and in the various national and in-
ternational standards bodies towards a hier- Indeed, electronic publishing can be viewed
archy of standards that will make it possible as a key integrative technology because it can
for a wide range of information systems to talk serve to integrate the various formats (paper
with and exchange information with each and microform as well as electronic) of infor-
other. mation input, processing, storage, and output
within a common technical framework. Elec-
Another trend is the rapid penetration of tronic publishing can also serve to connect the
computerized information systems in all sec- various so-called islands of automation in an
tors of society, but especially in the business, organization-office automation, publishing
educational, and research communities. This systems, database systems, records manage-
means that many of those who provide infor- ment, document storage systems, and the like.
mation to the government and use information Standards on information exchange are criti-
from the government can now or soon be elec- cal, as is the need to find ways for the people
tronically connected, and can, where appropri- who work in various areas of automation to
ate, send and receive information in a variety work more effectively together.
of electronic formats. This in no way suggests
an end to paper-based information products—
but only that paper can be used where it is The Microcomputer Revolution
really needed and in a more efficient and cost- Most Federal Government information is ei-
effective manner. ther collected from the private sector, State/lo-
Realizing this potential for information sys- cal government, and the general public, or is
tems integration requires, of course, more than created by Federal employees and contractors
just the technology and standards. A variety as the result of studies, analyses, research, and
of institutional and policy changes maybe nec- the like. Even information collected from out-
essary, and various alternatives will be dis- side the government is frequently subject to
cussed in later chapters. analysis by Federal employees, and in that
sense has a creative or value-added aspect.
Nonetheless, it appears that the technology,
the industry, the standards, and the govern- The dominant technology relevant to the col-
ment are all moving towards systems integra- lection and creation of Federal information is
tion. It is now possible to envision, in the rela- the microcomputer. Over the last 5 years or
tively short term, a Federal information so, the United States has witnessed a revolu-
49
.-

tion in computer technology that has brought puters drop to the $300-500 price range in the
the power of the mainframe computer to the next 3-5 years. 4 This would place the
desk of millions of public and private sector microcomputer in the same price range as a
employees and citizens. And this revolution good quality 19-inch color television.
is expected to continue for at least another 5
The continuous improvement in price/per-
years.
formance of microcomputers is driven in part
The sheer magnitude of this microcomputer by advances in semiconductor chip technology,
revolution can be measured in many different which shows no signs of slowing down. The
ways. For example, the Federal Government 32-bit chip family (such as the Intel 80386 or
itself has gone from only a few thousand micros the Motorola 68020) made possible the latest
in 1980 to roughly 200,000 in 1986 to 500,000 personal computer systems that are more pow-
in 1988 to a projected 1 million by 1990. The erful, more user friendly, and more compati-
percentage of school districts with computers ble with each other. Price/performance is ex-
had already increased from about 18 percent pected to continue to improve as the 32-bit
in 1981 to over 90 percent by 1985, according chips are further assimilated in microcomputer
to the National Center for Educational Statis- product offerings and as next generation mi-
tics. ’ In the business community, microcom- crocomputers are developed and introduced.
puters drew even in total computing power
The important impact on Federal informa-
with mainframes and superminis as of 1985
tion collection and creation is that an already
and are projected to dominate by 1990, accord- large and increasing amount of information is
ing to Dataquest. z Indeed, various projec- generated in electronic form, that is, by cap-
tions show microcomputers growing at an
turing electronic keystrokes with a microcom-
average 10 to 15 percent through 1990, com-
puter or word processor. Today, much of this
pared to about 5 percent for mainframes and electronic information is submitted to or pro-
superminis. The logic of this trend is under- vided by the government in paper form. But
standable when one considers that the IBM the potential exists to substantially reduce the
personal computer systems, selling for less amount of rekeyboarding, and presumably the
than $10,000 are equivalent in computing cost of such information, by maintaining the
power (measured in millions of instructions per information in electronic form as long as
second) to the IBM 370-168 mainframe com- possible.
puters that sold for several million dollars when
introduced in the mid-1970s.
The Continuing Role of Paper and
Even the home market has had significant Microform
microcomputer penetration, with about 19 mil-
lion households buying a microcomputer since A note of caution with respect to the role
1981 (about 14 percent of all households).’ of paper is in order. Despite the dramatic in-
Link Resources projects an ultimate home crease in computer technology and electronic
penetration of about 35 percent, although this information, paper documents are expected to
may be conservative if full function microcom- have a continuing, major role for several
reasons. First and foremost, for documents of
significant length, research has found that
‘Cited in tJ. Bloomdecker, Computer (%rne, Computer Secu-
rity, Computer Ethics (Los Angeles, Calif. National Center for
reading from a computer screen is much more
Computer Crime Data. 1986). difficult than reading from paper, despite im-
‘Cited in G. Lewis, Zoom! Here Come the New Micros, ” provements in the design and resolution of
Business W’eek, Nov. 1, 1986 pp. 82-92.
‘F;. Roth, ‘*Power Surge in Personal Computers, ” Editoriaf
Research Reports, vol. 1, No. 1, ,Jan. 9, 1987, p. 4. ‘Ibid, p. 6.
50

screens and terminals. Even extensive prac- tinuing significant use of microforms for
tice at electronic reading does not appear to records storage and archival purposes for the
make a significant difference. Second, paper foreseeable future, or at least until electronic
continues to be a more convenient and porta- alternatives have been fully established and
ble medium for many purposes, and accommo- stood the test of time. Microform is well suited
dates a wide range of reading styles and loca- for archival purposes because it requires less
tions. Third, for many documents, paper is still storage space (compared to paper), has a longer
a bargain, although this is changing with the shelf life (compared to paper and electronic,
advent of optical disk storage technology. And although this may change), is a stable access
of course, electronic publishing can signifi- technology (compared to electronic), and is
cantly increase the efficiency of paper use, lower in cost (compared to paper and some elec-
even when the final product is still in paper tronic).’ Microform offers a lifetime of 100+
format. Fourth, the paper format (especially years, whereas the lifetime of acidic paper is
for lengthy reports and books) permits the perhaps several decades, and magnetic media
reader to browse through material and use a (tape and disks) a few years to a decade or two.
variety of conscious or subconscious search The main competitive threats to microform for
patterns that may be difficult if not impossi- archival purposes are from acid-free paper
ble to replicate even with today’s computer- (which can last 100 + years, but would still re-
based search and retrieval software. Reading quire more storage space and be more costly
paper formats can lead to greater compre- than microform) and optical disks. Optical
hension. disks do not as yet have proven archival capa-
bility (although manufacturers are claiming
Overall, most studies on the future of paper 40+ years), require less storage space, and can
and paper-based media (such as books) have be less expensive than microform. Microform
concluded that the paper format will play a ma- is likely to continue as a major archival medium
jor role as a medium of information storage,
at least until optical disks (or some related
exchange, and dissemination for the foresee- electronic-storage technology) are well estab-
able future.5 The results of the GAO surveys lished.
of Federal information users (summarized in
ch. 4) indicate that paper is expected to hold However, for many nonarchival purposes,
steady or decline only marginally as a preferred microform is not the preferred medium even
format in the next few years, although this de- today. For reading lengthy written materials,
cline could become significant for specific types users find microform to be inconvenient, un-
of information (e.g., bibliographic, reference, comfortable, and inefficient compared to pa-
and statistical) that are highly suited to elec- per. For information search and retrieval, users
tronic access and manipulation. At the same frequently prefer electronic formats, including
time, the preference for electronic formats (e.g., online database systems as well as, increas-
electronic mail and bulletin boards, floppy ingly, offline media such as CD-ROMs. The re-
disks, and compact optical disks) is expected sults of the GAO surveys of Federal informa-
to increase dramatically. tion users (summarized inch. 4) indicate that
microfilm is little used today for Federal in-
The outlook for microform (microfilm and mi-
crofiche) is not as favorable as for paper or elec-
tronic formats, but there is likely to be con- %ee, for example, Kenneth E. Dowlin, The Electronic Li-
brary: The Prorm”se and the Process, (New York, Neal-Schuman,
1984; F.W. Lancaster, Libraries and Libraries in An Age of Elec-
%ee, for example, Priscilla Oakeskott and Clive Bradley tronics (Arlington, VA: Information Resources Press, Arling-
(eds.), The Future of the Book: Part I -- The Impact of New Tech- ton, VA, 1982); Edward Gray, “The Rise and Fall of Techno-
nologies ( Paris Unesco, 1982); U.S. Congress, Joint Commit- logical Applications: Considerations on Microforms and Their
tee on the Library, Books in Our Future, A Report From the Possible Successor, ” International Journal of Micrographics
Librarian of Congress to the Congress, S. Print 98-231, U.S. and Video Technology, vol. 15, No. 1, 1986, pp. 31-38; National
Government Printing Office, Washington, DC, 1984; John Y. Research Council, Committee on Preservation of Historical
Cole, Books in Our Future: Perspectives and Proposals (Library Records, Preservation of Historical Records ( Washington, DC,
of Congress, Washington, D.C. 1987). National Academy Press, 1986).
51

formation dissemination, and that microfiche, depend largely on the type of information and
while used extensively, is expected to decline the type of information user. The kinds of cri-
significantly as the desired format for dissem- teria that will be relevant in selecting format
ination of many types of Federal information. include:
It should be noted, however, that the micro-
amount of information to be stored, ac-
graphics industry is itself using electronic tech-
cessed, and/or distributed;
nology to continuously upgrade microform ac-
amount of storage space available;
cess technologies, such as computer-assisted
frequency of information access or re-
microfiche retrieval systems and computer-
trieval;
output microfilm systems.7 AIso, the tecol-
length of time information is to be stored;
ogy for microfiche to paper conversion con-
desired speed of access or retrieval;
tinues to advance. For example, the Defense
costs of storage, access, and retrieval;
Technical Information Center recently funded
number of users; and
the development and installation of duplex
technical expertise of users.
(two-sided) microfiche copier machines.
The rest of this chapter focuses on several key
In sum, however, the current and future use
electronic technologies relevant to Federal in-
of paper, microform, and electronic formats will
formation dissemination. The price/perform-
ance characteristics of these technologies make
‘See, for example, Coopers and Lybrand, Information and them highly competitive with paper and micro-
Image Management: The Industry and the Technolo@”es, study
conducted for Association for Information and Image Manage- form for those types of information well suited
ment, Sil\’er Spring, MD, 1987. for electronic formats.

ELECTRONIC PUBLISHING
Desktop Publishing ● relatively simple and straightforward doc-
uments (although the desktop publishing
One of the microcomputer applications most software is much more sophisticated than
relevant to this study is desktop publishing. typical word-processing software).
Desktop publishing combines elements of ad-
vanced word-processing and computerized
page layout and composition systems. Desk- A typical desktop publishing configuration
top publishing can be defined as a set of hard- includes:
ware and software, including a multifunction ● a microcomputer with mouse or digitiz-
personal computer, which has the ability to ing tablet, keyboard, and screen (roughly
produce near-typeset quality output, and uti- $2,000 to $4,000 inclusive);
lizing multiple type fonts, sizes, and styles and ● page composition software (about $500 to
multiple page layouts. The characteristics of $1,000);
desktop publishing are: ● a low-end laser printer (about $1,500 to
low cost (about $10,000 for an entire $3,000); and
system),
● a low-end scanner for paper input (about
user-friendly software (frequently employ- $2,000 to $4,000).
ing icons and a mouse and a ‘what you see
is what you get or WYSIWYG screen dis- Desktop publishing is expected to become
play) that requires minimal training, a standard part of personal computing, and to
near-typeset quality output (but still con- grow significantly over the next several years.
siderably less than high-quality book and Growth in desktop publishing reflects the sub-
magazine printing, for example), and stantial potential savings for those types of
52

documents that do not require higher levels lishers frequently find that laser printer type-
of quality and complexity. For simple reports, setting quality is good enough.
newsletters, pamphlets, and the like, desktop
The potential implications of desktop pub-
publishing can cut composition costs from the
lishing for Federal information dissemination
$50 or more per page range (for commercial
seem just as significant. An increasing percent-
composition and typesetting) to the $1 to $5
age of Federal information collected, created,
per page range. Compared to straight word-
and disseminated would appear to be well
processing text, desktop publishing can reduce
suited for desktop publishing.
the page length by perhaps 40 percent on aver-
age, and this translates into substantial sav-
ings from reduced paper and mailing costs. High-End Electronic Publishing
There are also major savings from a stream- The distinction between desktop publishing
lined revision process, minimal rekeyboarding, and so-called “high-end” electronic publishing
and the ability to store text and graphics for is somewhat arbitrary, since microcomputer-
future use and revision. based desktop systems can be connected or
Two significant limitations of the low-end networked to high-end work stations, typeset-
desktop publishing are limited ability to deal ters, and printers. Electronic publishing is con-
with complex documents (e.g., complicated lay- sidered to be the electronic preparation of ma-
outs using text and graphics) and limited print terial at all pre-press stages of the publishing
quality (due to the typical 240 dots per inch process, including text and graphics prepara-
(dpi) or 300 dpi resolution of low~nd laser tion, page layout, and composition, with the
printer output). The first limitation is being actual printing in any of a variet y of formats—
mitigated rapidly by ever more powerful desk- paper, microform, magnetic tape or diskette,
top publishing software releases. Also, users optical disk, or direct electronic. In general,
can invest in more sophisticated software and, high-end electronic publishing is distinguished
if necessary, obtain software that supports by:
phototypesetters as well as laser printers. Fi- high volume (in number of pages and
nally, low-end laser printers are improving out- copies),
put resolution, thus reducing the print qual- ● high quality (of the final product),
ity differential between laser printers and ● high complexity (of the page layout and
phototypesetters. composition), and
Desktop publishing has made dramatic in- ● high cost (compared to desktop systems).

roads in the newspaper and newsletter indus- High-end systems typically cost $30,000 to
tries. An estimated 80 percent of newspapers $150,000 depending on the configuration, com-
with a circulation of over 100,000 use Mac- pared to $5,000 to $10,000 for desktop sys-
intosh-based desktop publishing, including the tems. For the software alone, high-end publish-
Wall Street Journal and USA Today, as do an ing systems typically cost $15,000 to $30,000
estimated 75 percent of newspapers with a cir- compared to $500 to $1,000 for desktop
culation over 50,000.H Knight Ridder and the software.
Gannett Corp. are using desktop publishing
systems to create and distribute graphic de- The demand for high-end electronic publish-
signs nationwide. While major newspapers ing (and to a lesser extent desktop publishing)
generally use photocomposition equipment for is driven by a powerful combination of advan-
typesetting in order to get higher print qual- tages that translate into significant cost sav-
ity, newsletter and some small newspaper pub- ings and productivity improvements. For ex-
ample, electronically published materials are
generally found to be:
“F. Seghers, “In News Graphics, Macintosh Makes the Front ● more attractive,
Page, ” Business Week, Jan. 19, 1987, p. 87. ● easier to read,
53

more timely (publishing time can be any- bit workstations (such as Sun, DEC, or Apollo),
where from 25 to 90 percent faster), and a 19-inch monochrome display with a high reso-
● much less expensive. lution screen, and a local area network. The
typical system can accept input from CAD
Cost savings can be realized in several ways. workstations, scanners, graphics (raster and
For example, electronic publishing generally vector, line art and halftones), spreadsheets,
reduces the total number of document pages and text (in standard formats compatible with
by 40 to 50 percent, but occasionally up to 80
almost any mainframe, mini, microcomputer,
percent, since typeset pages contain more text or word processor). The system provides out-
than typewritten pages. This can dramatically put to various laser printers (such as Xerox,
reduce paper costs for hard copy print runs. Kodak, Imagen, Apple, and Sun) and pho-
For documents with limited demand and low totypesetters (such as Linotronic and Compu-
volume, electronic publishing makes printing- graphic). Advanced software capabilities typi-
on-demand a realistic option. Electronic pub-
cally include:
lishing also facilitates the revision process by
minimizing rekey boarding and graphics re- ● integration of text and graphics in non-
design. structured pages;
● free-form drawing with a mouse;
Various market surveys project a strong de- ● tracing tablet to copy drawings:
mand for electronic publishing over the next ● editing of digitized line art;
5 years, based on a perceived need for electronic ● pixel-by-pixel editing of halftone photo-
publishing by major corporations and govern- graphs; and
ment agencies. ● simultaneous editing of different portions
Electronic publishing systems have made of the same document.
rapid technical advances in just a few years.
This trend is expected to continue due, in part, Overall trends in electronic publishing in-
to heavy competition among graphics work- clude the following:
stations, publishing software, traditional movement from a fragmented market to
photocomposition services, and computer an integrated market;
equipment companies, as well as systems in- aggressive competition from electronic
tegrators that combine hardware and software publishing systems offered by traditional
from numerous vendors. At the heart of elec- phototypesetters and by electronic pub-
tronic publishing systems is the 32-bit work- lishing service bureaus;
station that permits complex manipulation of standardization of information exchange
text, graphics, and, increasingly, halftones. among different types of hardware and
These are the same types of workstations used soft ware;
for computer-assisted design (CAD) and so- declining price/performance ratios; nar-
phisticated graphics applications. This work- rowing of the technical differences be-
station is now an established technology, with tween desktop and high-end publishing
a substantial track record. According to Data- systems;
quest, 32-bit workstation sales ($15,000 to increasing integration of direct-to-plate
$50,000 per workstation price range) grew from printing technologies; and
about 100,000 units in 1983 to 1 million units increasing speed and quality of perform-
in 1985, an estimated 2 million in 1987, and ance (including higher resolution, color,
a projected 4 million in 1989. 9 and multiple languages).
The technical power, sophistication, and flex- In the corporate community, investment in
ibility of electronic publishing systems are il- electronic publishing is generally claimed to
lustrated by a typical system which uses 32- have a rate of return of 50 to 60 percent and
a payback period of 2 years or less. Also, com-
“Cited in G, Lewis, “ Nrew Nlicros, ’ op. cit. panies typically claim to have cut overall pub-
54

lications turnaround time by 50 to 75 percent. direct savings on the order of 40 percent over
While similar data are not yet available from preprinted forms.
government users, Interleaf Corp. indicates For larger print runs, even greater savings
that the following Federal agencies are using may be possible where offset printing can be
Interleaf electronic publishing systems: De-
used to reduce the per-page printing cost of
fense Advanced Research Projects Agency (for
about two cents assumed for laser printers.
technical reports); Office of Naval Research (for Further savings seem likely since the forms
research studies); intelligence agencies (vari-
can be stored and edited electronically, mini-
ous applications); U.S. Coast Guard (for tech- mizing rekey boarding and redesign. As
nical manuals); U.S. Army (for technical
another example, the combination of micro-
manuals); Department of State (for regula-
computers (or mainframe terminals) and laser
tions); Department of Agriculture (for statis-
printers can be used to permit direct electronic
tical documents); Bureau of the Census (for sta- input of data collected by agencies into stand-
tistical reports); and Federal Reserve Board
ard reporting forms stored on the laser printer.
(for financial analyses). Xyvision reports sales The completed forms can be transmitted elec-
of electronic publishing systems to, among
tronically to a regional office or to Washing-
others: the National Center for Health Statis-
ton, DC, eliminating both cost and potential
tics, National Center for Disease Control, U.S.
errors associated with rekey boarding and the
Geological Survey, Bureau of the Census, and
time delays associated with mail delivery. Pa-
Central Intelligence Agency.
per copies can be printed out for archival
purposes.
Electronic Forms Management
Computer Graphics
Another growing application of systems re-
lated to electronic publishing is electronic Advances in computer graphics are central
forms management. Several companies special- to the recent breakthroughs in desktop and
ize in this applications area. The typical stand- high-end electronic publishing. Indeed, com-
alone workstation, including a processor and puter graphics capabilities are key aspects of
hard disk along with software and a high reso- most electronic publishing systems. And pub-
lution display, costs in the range of $25,000 lishing applications have themselves become
to $60,000, depending on memory size. The one of the driving forces for further advances
typical system has many of the capabilities of in and broader use of computer graphics. Other
electronic publishing systems discussed ear- driving forces include:
lier, and can be used for designing newsletters,
manuals, and technical documents as well as graphics needs of the scientific com-
forms. munity;
military applications of computer graphics,
However, it is not necessary to have full- most recently stimulated by the Strate-
capability electronic publishing systems for gic Defense Initiative’s requirements for
many forms-management applications. For ex- very sophisticated, three-dimensional, dy-
ample, among Federal agencies, the Air Force, namic computer graphics and modeling;
Army, Navy, Internal Revenue Service, Social continued movement toward graphics
Security Administration, Federal Reserve standards; and
Board, Federal Deposit Insurance Corporation, continued breakthroughs in price/perform-
and National Aeronautics and Space Admin- ance ratios.
istration are all using personal computers and Major technical trends in computer graphics
laser printers to manage forms. Microcom-
include:
puters are used to enter the data, and laser
printers are used to merge the data with stand- ● the continuing transition from film-based
ard forms for printing. The agencies indicate techniques to digital processing;

55

development of relatively low-cost (under cost-effective than rekey boarding or redraw-


$15,000) desktop color scanners and ing those materials not in electronic format.
printers;
The cost of scanners has dropped to the point
further improvements in high resolution
where low-end scanners are available in the
graphics (up to 2,000 x 2,000 pixels);
$2,000-4,000 price range with a speed of up to
further development of full color, interac- several pages per minute and a scanning reso-
tive, three-dimensional graphics worksta- lution of 200-300 dots per inch. While satis-
tions at relatively modest prices (e.g., factory for many desktop applications, higher
$30,000); speed and resolution are generally needed for
continued migration of high-end worksta- high-end publishing purposes. High-end scan-
tion capabilities to low-end workstations;
ners are available in the $15,000 to $40,000
and progress in developing standards for price range with speeds of 1 or 2 pages per sec-
exchanging graphics data between work- ond and resolution levels up to 400-dp~. Thus,
stations, such as the Digital Data Ex- the high-end scanners achieve speeds and reso-
change Standard. lutions similar to the high-end laser printers
Further technical progress in computer discussed later.
graphics seems assured as various companies
continually develop new products for top secret A major advantage of high-end scanners is
military applications. Advanced digitized map- the capability to approximate graphics-quality
ping techniques are used by the Defense Map- halftone pictures. This is accomplished by
ping Agency and by various Federal civilian scanning the image at up to about 120 scan
agencies such as the Fish and Wildlife Serv- lines per inch and recording multiple bits for
ice, National Oceanic and Atmospheric Admin- each pixel, rather than the one bit commonly
istration, and Bureau of the Census. Computer- used for scanning text and line art. Instead
ized graphics products can interpret infrared of recording black or white with one bit per
aerial imagery and produce maps. In general, pixel, multiple bits permit the recording of the
computerized mapping offers advantages sim- degree of blackness for each pixel, known as
ilar to computerized printing in that: gray-scale scarming. Also, many high-end scan-
ners can scan a wide range of type styles and
● The original map preparation is much sizes, and some scanners can be programmed
faster. to learn new (to the scanner) type styles. These
● Maps can be stored electronically to fa- capabilities are expected to migrate to the
cilitate relatively easy updating. desktop scanners.
● The original map and an-y revisions can
be displayed on a video screen. The technical status of printers is more com-
● Hard-copy output can be obtained rela- plicated because printers are now used for func-
tively quickly with a plotter or laser tions other than printing, such as typesetting,
printer. graphics input, and forms management.

Scanners and Printers For printing of straight textual material,


electromechanical line printers (known as im-
Almost all desktop and high-end electronic pact printers) are efficient for low-copy runs
publishing systems are configured to include (i.e., one or a few copies per original). Medium
one or more printers, and may include one or performance impact printers can print at about
more scanners. The price/performance of scan- 1,000 lines per minute (20 pages per minute
ners and printers has dropped dramatically in at 50 lines per page) and cost $10,000 to
the last few years. Scanners are used primar- $15,000. High-end impact printers can reach
ily to digitize text and images that are initially output speeds of about 3,600 lines (72 pages)
in paper formats. Scanners are not as efficient per minute. Low-end desktop line printers print
as direct electronic input, but are much more at a few pages per minute.
56

The role of impact printers is expected to con- laser printers are now on the market. High
tinue to decline, because of the need for print- resolution non-impact printers are adequate to
ing graphics and complex page layouts, the use meet many electronic publishing needs, either
of variable type styles and fonts, and the in- for demand printing or as camera-ready copy
tegration of forms and data at the point of to be used in subsequent plate-making and
printing. photo-offset printing. Continued technical ad-
vances and market forces are likely to push
Non-impact printers (using laser, light emit- the typical output resolution of laser and other
ting diode array, ion deposition, and other tech- non-impact printers into the 600 to 800 dpi
nical processes) provide better quality, greater range over the next few years, thus further
flexibility and diversity, and faster speed (at closing the quality differential.
the high end). While on a per-page cost basis,
non-impact printers may be more expensive With respect to print-volume requirements,
than impact printers, this is not an appropri- it is still far cheaper to use conventional photo-
ate comparison in most cases. When serving offset printers for high-volume print runs than
as a typesetter or graphics printer or proof laser and other non-impact printers. One can
printer, the non-impact printer can be an or- debate the various break-even points as a func-
der of magnitude cheaper than conventional tion of the length, format, number of copies,
methods. For example, for low-volume appli- and desired turnaround time for specific doc-
cations where 300 dpi output resolution is uments. In offset printing, plate-making (prep-
satisfactory, a desktop laser printer at $1,500 aration of masters or negatives of the original
to $3,000 may be perfectly adequate for pro- images by which ink is transferred onto paper
ducing camera-ready copy, compared to a pho- to make copies of the original) costs anywhere
totypesetter at $35,000 to $70,000. from a few dollars to $50 and up per page. This
cost indicates that non-impact printing is fre-
The high-end, non-impact printers are still quently less expensive for short print runs of
quite expensive, typically in the $100,000 to under a few hundred copies per original. For
$200,000 range although these prices are ex- larger print runs, the printing cost is likely to
pected to come down. A typical high-end, non- be cheaper with photo offset rather than laser
impact laser printer prints at 90 to 120 pages printing.
per minute. By comparison, a desktop laser
printer prints a few pages (e.g., 3 to 6) per min- It appears that the cost of non-impact print-
ute and costs as low as $1,500. At the next ing (including xerographic) is rarely below 2
level up, a typical mid-range laser printer might cents per page. Thus, assuming $2 per page
print at 12 to 20 pages per minute and cost for plate-making (this is for desktop relatively
$10,000 to $15,000. Again, price/performance low volume applications), and assuming all
ratios continue to fall. other costs are equal (purchase or lease, main-
tenance, supplies, and labor), the break-even
Non-impact printers are not well suited for point would be about 100 copies. In this
jobs requiring high output quality and/or print hypothetical and oversimplified case, print
volume. With respect to quality, most non- runs under 100 copies per original would use
impact printers can achieve an output resolu- a non-impact printer and print runs over 100
tion of 300 dpi (assuming that the input reso- per original would use a photo offset printer.
lution is at least at that level). This output qual- Other elements besides cost may enter into the
ity is adequate for a wide range of purposes, printing decision, such as quality, speed, turn-
but not for high-quality publications. By com- around time, and control. In the future, the
parison, photocomposition equipment can pro- break-even point between non-impact and
duce typeset output at resolutions of 1,200 or photo-offset printing will depend in part on
more dpi. Technical advances are reducing this their relative technical advances and cost re-
quality differential. Indeed, 400 to 1,000 dpi ductions.
57
— —

ONLINE INFORMATION DISSEMINATION

Online Information Retrieval end, access to the desired database is fre-


quently via an online database services com-
Previous discussions have focused on a num- pany (e.g., Lockheed DIALOG, Pergamon In-
ber of electronic information technologies– foline) or a database gateway company (such
microcomputers, page composition and pub- as is available from Western Union Easy Net).
lishing software, computer graphics, scanners, Gateway companies serve as intermediaries be-
printers—with the information maintained in tween the customer and the database source
electronic form through many or all stages of and do not maintain the database itself. On-
the publishing process. The primary final out- line database service companies actually main-
put has been in paper format. Advances in tech- tain copies of the databases online, so that
nology make it possible to disseminate the out- referral to the database source is not neces-
put information in a variety of electronic sary. Some database source companies do pro-
formats as well as paper. For some purposes vide for direct customer electronic access to
and some kinds of information, electronic for- the database, without going through a gate-
mats may be preferable to paper. This is espe- way or online services company. Companies
cially so for bibliographic, reference, statisti- that maintain online databases need:
cal, and bulletin board information where the
user may not want to see the whole document, a host computer and memory necessary
but is only interested in locating specific pieces for handling the volume of data and fre-
of information. The private sector information quency of use, and
industry has given high priority to computeriz- the necessary front-end processor and
ing access to these types of information, communications equipment for handling
whether the original source of the information remote inquiries and transmitting re-
is the government, academic, research, or com- sponses.
mercial sectors. The growth of the online information indus-
try has been phenomenal. From less than $500
This section discusses the technology and million in annual revenues in 1978, the indus-
application of online information retrieval sys- try has grown to about $2 billion total reve-
tems in the context of the private sector, since nues in 1986, $3 billion in 1987, and is projected
this is where much of the online activity is oc- to reach about over $4 billion by 1990.
curing. From a technical viewpoint, these pri-
vate sector applications are directly relevant A typical commercial online database serv-
to the Federal Government. ice charges about $40 to $80 per hour, of which
about 40 to 45 percent is for acquiring and pre-
paring the actual data, and another 40 to 45
The technology of online information re-
percent is for sales, marketing, and adminis-
trieval is well established. Customer access is
tration. About 6 to 9 percent is for communic-
typically via a microcomputer or terminal con-
ations (including the cost of customer prem-
nected to a modem. Residential customers nor-
ises equipment, e.g., computer terminal and
mally tie into the local telephone company
modem, local exchange access, and interex-
network (e.g., Bell Operating Companies, in-
dependent telephone companies) and, if access- change link if applicable), and about 6 percent
is for data processing (including the cost of
ing a database from long distance, then con-
hardware and software for database storage
nect to an interexchange carrier network (e.g.,
AT&T, MCI, U.S. Sprint) or a value-added net- and data communicaticm).l’)
work (e.g., Tymnet, Telenet). Business custom- “)Studies by Cuadra Associates and Elsevier Science pub-
ers can sometimes bypass the local telephone lishing cited in P. W’. Huber, The (leodesic ,Vetwork: 1987 Re-
port on Competition in the Telephone Industry, prepared for
company and connect directly to an interex- U.S. Department of Justice, Washington, DC, January 1987,
change or value-added network. At the other p, 7.13.
58

One implication of the above cost structure eral agencies. As currently planned, FTS 2000
is that substantial savings can result to the will include:
extent that the data are already in the appro- ● switched voice (up to 4.8 kilobits/second
priate electronic format. If as a result of elec- transmission capacity);
tronic publishing government statistical or ● switched digital integrated service;
reference reports were produced in electronic ● packet-switched services;
form as a matter of course (even if the ultimate ● video transmission (including graphics,
product is in paper format), then the electron-
facsimile, limited and full motion video);
ically formatted information could, at least
and
theoretically, also be made available as an on- ● dedicated voice or data transmission
line database. This could significantly reduce
circuits.
the cost of data acquisition and preparation.
This is a major cost element, regardless of The switched digital integrated service and
whether the government and/or commercial packet-switched service should be especially
firms disseminate the data. Sales, marketing, useful for online database retrieval or electronic
and administrative costs may not be as amena- document transmission. The switched digital
ble to reduction for commercial firms, unless integrated service is designed to be the equiva-
they are working under contract to the gov- lent of the Integrated Services Digital Network
ernment in such a way that the market was, (ISDN) for digitally integrating voice, data, im-
in effect, guaranteed. For the government, dis- ages, and video over the same transmission
tribution to information intermediaries (e.g., medium. As planned, the FI’S 2000 version will
libraries in the depository program) might help be consistent with ISDN international stand-
reduce marketing and other costs. Technology ards and will have a 1.544 megabit/second
is only a small part (perhaps less than 15 per- transmission capacity subdivided into 24 in-
cent) of the cost of online databases. dividual channels of 64 kilobits/second each.
For illustrative purposes, one 64 kilobit/sec-
ond channel can transmit about four pages of
Telecommunications text per second (at 250 words per page x 8 char-
Online information retrieval services and acters per word x 8 bits per character). The
several other kinds of electronic information planned IWS 2000 packet-switched service will
dissemination (e.g., electronic mail and facsi- be consistent with international standards for
mile) are dependent on telecommunication open systems interconnection and interoper-
technology and systems. A number of devel- ability with public data networks and public
opments are converging to facilitate and most electronic mail services. The FTS 2000 packet-
likely reduce the relative cost of data commu- switched service is planned as a 24 hours a day,
nication. One key trend is the transition from 7 days a week operation with 99.5 percent up-
analog to digital telecommunication networks time and 98 percent availability.
that are designed to transfer information much The basic concept of packet-switchingis that
more efficiently than the conventional analog data can be transmitted most efficiently when
telephone networks. A second trend is the rapid assembled into packets (or bunches) of bits of
movement towards national and international information. The U.S. packet-switching vol-
standards for data networks of all kinds. A ume for 1985 has been estimated by Interna-
third trend is the maturation of Ku-band sat- tional Resource Development at about 47 mil-
ellite, fiber optic, and FM subcarrier technol- lion kilopackets, of which 7 million kilopackets
ogies for data transmission. were for database access, 3 million for elec-
tronic mail, and 0.3 million for electronic data
The implementation of FTS 2000, the up- interchange. 11 Typical commercial rates for
graded Federal Telecommunication System, is
intended to make state-of-the-art data commu- 1lCited in P.W. Huber, Telephone Industry, op. cit., table
nication capability available to all major Fed- PA.1.
59

packet-switching have been estimated at about The trend towards so-called intelligent build-
$0.50 per page of text for local packet- ings will facilitate integration across different
switching and roughly 3 times that for national telecommunication technologies and services.
packet switching. ” While these rates com- Intelligent buildings are prewired during con-
pare favorably with electronic mail and may struction with local area networks (LANs) ca-
be acceptable for very short documents, the pable of handling digital data communication.
cost of packet-switching long documents would LANs can carry information much faster and
be quite high. Whether FTS 2000 will signifi- more efficiently than the conventional tele-
cantly reduce packet-switching costs in un- phone and PBX (private branch exchange) ana-
known at this time. log circuit and switching systems. The cost of
LAN installation is much reduced if completed
Data transmission networks of all kinds are
during building construction rather than ret-
expected to incorporate both satellite and fi- rofitted. The trend toward intelligent buildings
ber optic technology wherever appropriate. For is expected to accelerate in response to the
example, a high speed (56 kilobit/second) rapid increase in networking of microcomput-
packet-switched data transmission network ers, mainframe terminals, peripheral equip-
can incorporate both a fiber optic terrestrial
ment (including scanners, printers, and
component and a Ku-band (12-14 gigahertz) graphics workstations), and the like in the of-
satellite component. The Ku band permits use fice environment.
of lower-cost, very small aperture (VSAT) earth
stations with receiving disks that are 1.2 or A final telecommunication technology to be
1.8 meters in diameter. Such a system could
discussed in this section is FM (frequency
be used for such functions as transmitting data
modulated) radio subcarrier transmission. The
collected from remote locations.
FM subcarrier is an excess portion of the band-
Over the next few years, a balanced network width assigned to FM radio stations, and was
of satellite and fiber optic transmission links deregulated by the Federal Communications
is likely to evolve. Fiber optic links are likely Commission in 1983. The FM subcarrier ap-
to be used primarily for heavy volume, point- pears to be cost-effective for point-tOmulti-
to-point transmissions, while satellite links are point transmission of time-sensitive digital
expected to dominate for point-to-multipoint data traffic, such as news and public affairs
transmissions. Experimental tests of fiber op- information. For example, MultiComm Tele-
tics have attained transmission rates of 4 bil- communications Corp. (Arlington, VA) is using
lion bits/second over relatively short distances. Western Union’s Westar IV satellite to trans-
By comparison, this is more than a 1,000 times mit information to 90 participating FM radio
the 1.544 megabits/second transmission capac- stations, where the information is in turn
ity specified in the ISDN standard, and is retransmitted on the FM subcarrier to receiv-
equivalent to transmitting an entire 30-volume ing sites equipped with a special, low-cost FM
encyclopedia in 1 second. receiver. The information can be stored on a
microcomputer or printed out. MultiComm sells
The integration of fiber optics with satellite, the receiver/printer for $500 or leases the equip-
microwave, and copper wire circuits will be ment for a nominal fee of $25/month. The costs
facilitated by the continuing development of of the service per receive site range from 20
teleports, with respect to traffic between ma- cents per page of information transmitted for
jor U.S. metropolitan areas and overseas traf- immediate delivery (e.g., within 19 seconds),
fic. Teleports are essentially buildings and fa- 10 cents per page for delivery within 2 hours,
cilities that serve as a platform or bridge for and 5 cents per page for overnight delivery.
interconnecting different modes of telecommu-
nication all at one location. This is far cheaper than courier service, espe-
cially for shorter documents. The 90 partici-
pating FM stations broadcast to an estimated
“I bid., table PA.2 and accompanying text. 85 percent of the U.S. population. Ku-band
60

small satellite earth stations could be used to service. X.400 is based on the 0S1 (Open
reach rural and remote areas. MultiComm Systems Interconnection) model and will
offers a Federal News Service that transmits permit interconnection among various
transcripts of White House briefings, congres- electronic mail services.
sional testimony, and the like to hundreds of
newspapers and trade associations, and an In- Many electronic mail systems require a modem
fowire service for low-volume users who need (modulator/demodulator) at each end of the cir-
time-sensitive information on, for example, cuit, to convert the digital signals from the
White House and agency press releases, ad- sending computer into analog signals for trans-
mission over the telephone lines (at least in the
vance schedules of upcoming hearings, and the
like. Other private firms are using the FM sub- local exchange) and back again from analog
carrier to distribute such information as stock to digital at the receiving computer. However,
market quotes. modems are likely to be less of a constraint
in the future for at least two reasons. First,
the cost of modems continues to drop–a 300
Electronic Mail bits per second modem now costs $100 to $200,
Another technical option for online informa- a 1,200 bps modem (the de facto standard for
tion retrieval and two-way information trans- remote computer networking including elec-
fer is electronic mail. As discussed previously, tronic mail and bulletin boards) $200, and the
electronic mail capability is planned as part higher speed 2,400 bps modem about $300 to
of the FTS 2000 system. Electronic mail has $400. Second, in the future, all-digital data
grown more slowly than initial expectations, communication and telephone networks will
but appears to breaching a critical threshold eliminate the need for modems almost entirely.
of viability. Modems will be necessary only to the extent
analog phone systems are still used.
The outlook for electronic mail is being en-
hanced by several key trends: The cost of electronic mail varies according
to the length and the volume of the messages
Electronic mail is increasingly included as and the type of electronic mail system used.
a basic capability of office automation sys- For an inhouse personal computer or office
tems, such as those offered by Data Gen- automation-based electronic mail system, the
eral, DEC, IBM, Wang, and NBI; cost range has been estimated at roughly $1
Vendors are providing much improved ca- to $2 per 3-page message (7,500 characters) at
pacity for interconnections or gateways a monthly volume of 1,000 messages, and is
between electronic mail systems (e.g., estimated to drop to about $0.10 to $0.20 per
Wang and IBM, DEC and MCI Mail, IBM 3-page message at a monthly volume of 10,000
and Western Union Easy Link, MCI Mail messages. By comparison, electronic mail serv-
and CompuServe Easy Plex); ice bureaus typically charge in the range $1
Enhanced electronic mail capabilities are to 3 per 3-page message regardless of volume.
being developed that can handle graphics
and spreadsheets besides American Other alternatives for transmission and re-
Standard Code for Information Inter- ceipt of electronic mail include: electronic bulle
change (ASCII) text; and tin boards, digital facsimile services, and
There is growing acceptance of the CCITT videotext services. For discussion of these and
(Consultative Committee on International other related telecommunication technologies,
Telephone and Telegraph) X.400 standard see Communication Systems for An Electronic
for electronic mail and messagehandling Age (OTA, forthcoming, 1989).

61
— —

OPTICAL DISKS
For information that neither changes fre- hard disks, or about 10 of the 1,600 bits-per-
quently nor requires immediate, online remote inch magnetic computer tapes. A 12-inch (30
access, optical disk technology is a viable tech- cm) WORM (Write Once Read Many times) op-
nical option for purposes of information stor- tical disk can store up to 1 gigabyte (billion
age and dissemination and as an important bytes), which is roughly double the capacity
component of electronic publishing systems. of a CD-ROM. All of these storage capacities
(Other optical technologies not discussed here, are per single side, and would be doubled for
such as optical or laser cards on strips, could two-sided disks.
provide storage and dissemination of smaller
The total and per bit or byte manufacturing
amounts of information. ) While some stand-
ards issues still need to resolved, the signifi- costs of both 4.72-inch CD-ROMs and 12-inch
WORM optical disks are already quite low. CD-
cant technical advantages of optical disks are
ROMS can be mastered for $4,000 to $5,000
becoming more and more evident as a result
and can be reproduced in quantities ranging
of numerous development applications, proto-
from $30 per disk for 100 copies to $6 per disk
type tests, and, commercial offerings.
at volumes of several thousand. Some esti-
Optical disk technology uses a laser beam mates suggest per disk costs as low as $3 for
to record data on plastic disks by engraving volume runs. The 12-inch WORM disks are
pits in the surface. The disks can then be sub- more expensive to produce, at about $150 a
sequently read by a low-power laser beam to copy, but are still far cheaper per byte than
retrieve the data. There are several different floppy diskettes or hard disks. These costs do
types of optical disk, and some are further not include the cost of data acquisition and
along in terms of technology and standards preparation, which apply to any storage
than others. Standards are essential for opti- medium, and the cost of equipment needed to
cal technology to ensure compatibility among read the disks. All that is necessary to read
different types of disks and disk readers, and CD-ROMs is a CD-ROM reader, available from
to minimize the possible need for future several vendors in the $500 to $1,000 price
rerecording of data due to incompatible range, and a personal computer and screen.
equipment. Thus, for users already owning a microcom-
puter system, the incremental cost of CD-ROM
The major advantage of optical disk tech- equipment is in the same range as the medium
nology is the ability to store and disseminate to high-end consumer-oriented compact digi-
large amounts of information at very low cost. tal audio disk players. WORM readers are con-
For example, a 4.72 inch (12 centimeter) CD- siderably more expensive—several to tens of
ROM (Compact Disk-Read Only Memory) can thousands of dollars range-although this can
store up to roughly 540 megabytes (millions be modest for the institutional (corporate and
of bytes) of data. Assuming that one typewrit- government) users who are the likely clients
ten text page averages 250 words or about for 12-inch WORM disks.
2,000 bytes per page, one CD-ROM can store
up to 270,000 pages of typewritten text. Grol- Optical disks also offer other advantages:
lier has recorded its entire 20-volume Academic rapid access to stored data (i.e., in one
American Encyclopedia on about one-fifth of second),
one disk. One floppy diskette (single-sided, sin- the ability to use a microcomputer for data
gle density) can store about 360 kilobytes of access and retrieval,
data, which is equivalent to about 180 pages high levels of data integrity,
of double-spaced typewritten text. Thus, 1 CD- very minimal disk or equipment wear,
ROM can store the equivalent of about 1,500 convenience and portability, and
floppy diskettes, about 54 of the 10-megabyte relatively long media life.
62

The latter point is somewhat controversial as USGS is prototyping the use of CD-ROMs
initial manufacturer estimates of 10 to 20 years for the possible goal of providing all (or
have now been extended to 40 to 50 years. a large part) of USGS earth science infor-
Some suggest that 100 years is possible un- mation in CD-ROM format such as seis-
der ideal conditions. Disks could be recopied mic data from the National Earthquake
at periodic intervals if necessary. Information Center. USGS officials be-
lieve that CD-ROM offers the potential to
The high level of commercial and govern- make earth science data much more acces-
mental activity is indicative of the potential sible at lower cost.
for CD-ROM and WORM disks. Vendors (such The U.S. Navy’s Printing and Publica-
as Lockheed DIALOG, Cambridge Scientific tions Service is implementing a print-on-
Abstracts, Aide Publishing, and VLS, Inc.) are demand system for 1.2 million pages of
offering many new optical disk-based products military specifications and standards, in-
and services. Many of these include databases cluding text and graphics images. The
that originate in whole or in part from the Fed- Navy is using a 12-inch WORM optical
eral Government. disk unit to record the disks, which are
then placed on two 32-disk juke boxes.
Federal agencies are actively pursuing a wide More frequently requested documents are
range of development and prototype projects. concentrated on a few disks, and output
For example: is printed with Xerox 9700 laser printers.
● The Nuclear Regulatory Commission (in The system is intended to:
cooperation with the Smithsonian Insti- —reduce warehouse space and printing
tution) is developing a WORM optical disk costs,
system to keep track of submissions re- —improve response time,
garding nuclear waste disposal under the —eliminate dissemination of out-of-date
Nuclear Waste Policy Act of 1982. The documents, and
system uses personal computers, scan- —serve as a prototype for many other
ners, and 12-inch WORM optical disk applications-for example, technical man-
drives. uals, training materials, and handbooks.
● The Library of Congress is prototyping
The Navy intends to develop interactive
optical disk technologies for general re- applications for document updating, alter-
search, archival, and information retrieval
native storage media (such as CD-ROM
purposes, including the use of a 100-disk
and 5.25-inch WORM optical disks), and
optical jukebox for 12-inch WORM disks.
document search capability.
The jukebox has a potential storage ca-
Other Federal agencies actively pursuing
pacity of 200 gigabytes. optical disk technology include the: In-
● The National Library of Medicine is pro-
ternal Revenue Service, Patent and Trade
totyping various optical disk technologies
mark Office, National Archives and Rec-
for medical applications, research, archival,
ords Administration, National Aeronautics
and instructional purposes.
and Space Administration, Central Intel-
● The Bureau of the Census is prototyping
ligence Agency, and numerous compo-
the use of CD-ROMs for storing and dis-
nents of the Department of Defense in-
tributing maps that will result from the
cluding the National Security Agency.
1990 census. The Census Bureau is also
examining the potential of CD-ROM for Another popular optical disk technology is
a broad range of geographic and topo- the analog videodisk. This is heavily used for
graphic maps as part of the Topographic educational and training purposes, and can
Integrated Geographic Retrieval (TIGER) store up to 54,000 images per disk. Videodisks
project being conducted jointly with the are roughly similar to CD-ROMs in cost—about
U.S. Geological Survey. $2,000 to master, $18 per disk for the first 100
63
——— —.

copies, and under $10 per disk for runs of sev- only, character-encoded text for edit-
eral thousand. Videodisk readers cost in the ing/processing).
range of a few hundred to a few thousand
Optical disk technologies and applications
dollars.
continue to advance at a rapid pace. Double
sided 12-inch WORM disks are now available
Beyond the CD-ROM, WORM, and video- with up to 4 gigabytes storage capacity per
disk, there are several other optical disk tech- disk. The initial commercial 5.25-inch erasa-
nologies under active development and appli- ble disks already have been introduced, and
cation. The most noteworthy are: the CD-I CD-I disks now are in the prototype stage, with
(Compact Disk-Interactive) that combines commercial introduction expected in 1989 or
text/data, video, audio, and software storage, 1990. Some vendors have expanded the capac-
editing, and retrieval on one disk; and the eras- ity of CD-ROM disks up to 750 megabytes,
able 12-inch or 5.25-inch optical disk. CD-I is and others offer CD-ROM juke boxes that can
of particular interest because it will make pos- access up to 240 disks. Personal computer com-
sible such compact disk applications as: talk- pact disk readers are entering the market, as
ing CD books, “smart” CD books (using ex- are specialized PC-CD/ROM applications (e.g.,
pert systems), CD book (or library) of the using hypertext or hypergraphics software).
month, and interactive audio, video, and data- A fledgling CD-ROM service bureau industry
base software. CD- I will be a disk with power- is developing, not far behind and perhaps even-
ful capabilities including: tually to be integrated with the electronic pub-
lishing service bureau industry.
graphics (e.g., digital video still frames,
limited motion video, encoded colors, full A major critical path item for optical disk
screen animation); technology is the development of standards.
audio (e.g., digital audio, hi-fi, mid-fi, The current status and outlook are briefly dis-
speech quality); and cussed in a later section, along with consider-
text (e.g., bit-map text storage for display ation of other standards issues.

EXPERT SYSTEMS
Expert systems, sometimes known as knowl- and educational areas. For example, expert sys-
edge-based or rulebased systems, are typically tems have been used to:
computer soft ware packages that permit users
to have the benefit of expert knowledge in
● help make agricultural management de-
specified subject areas. The ‘expert inexpert cisions regarding pest control as a func-
system means that both the knowledge and tion of the type of crop, landscape, weather
rules (decision paths and criteria) built into the conditions, season, other vegetation, in-
software come from relevant subject matter festation history, and the like;
experts. Expert systems have advanced to the
● help students explore and master a sub-
point where widespread application to many ject or skill and even monitor the learn-
aspects of information dissemination is ing progress of the students (known as
likely-ranging from technical writing to in- “electronic or intelligent tutors”); and
formation access and retrieval to the manage- ● help technicians interpret technical data
ment of electronic publishing. from computer-assisted manufacturing
systems.
The expert systems applicable to informa-
tion dissemination are no different in princi- Expert systems can be tied into both online
ple from the systems that have been success- bibliographic and full text information retrieval
fully applied to various scientific, industrial, and to electronic publishing. For example, pro-
64

totype expert systems with sophisticated The Defense Technical Information Cen-
search strategies are being used to retrieve and ter has established an Artificial Intelli-
deliver full text information via electronic pub- gence/Decision Support Laboratory that
lishing systems. These kinds of information is working to apply the full range of ex-
retrieval expert systems could eventually work pert systems and even more powerful arti-
hand-in-hand with expert systems designed to ficial intelligence technologies to informa-
efficiently manage electronic publishing. One tion access and retrieval. The ultimate
can easily envision the day when expert sys- objective is to facilitate the capture and
tems will help optimize the electronic publish- transfer of knowledge from the experts to
ing and dissemination (paper and electronic) the users of DTIC (and other DoD) infor-
of information products (or packages of prod- mation systems, utilizing innovative in-
ucts), given the specific profile of the product formation display techniques and full
(number of pages, composition, type style, use integration with the DoD Gateway Infor-
of graphics, etc.), anticipated user needs (e.g., mation System that is interfaced with
size of demand by format), and the mix of dis- hundreds of online databases.
semination channels (initial press run of pa- The National Agricultural Library (NAL)
per copies, provisions for demand printing, on- has developed prototype expert systems
line database access, optical disk distribution, that query users on their information
etc.). needs and route them to the appropriate
Numerous expert system applications for in- bibliographic sources. The prototype was
provided to over 700 librarians in a floppy
formation search and retrieval are under de-
disk format that runs on a microcomputer.
velopment. For example:
NAL hopes to create a critical mass of ex-
● The National Records and Archives pert system users, and believes that ex-
Administration (NARA) developed a pro- pert systems could help free librarians
totype expert system to assist with rou- from the more routine ready reference and
tine inquiries from researchers. The objec- directional questions. NAL is also explor-
tives of the project were to evaluate the ing linking expert systems to other gov-
capability of an expert system to capture ernment and commercial online databases
the expertise of experienced archivists and and CD-ROM players. Expert systems
to relieve them of the significant expend- could be used to query the user on his or
iture of time needed to answer routine in- her information needs, help sharpen the
quiries. Test results indicated that if the request, and then route the request to an
prototype system were expanded to full online bibliography, a disk-based bibliog-
scale, the system could be expected to raphy, or a full text document on video-
agree with its human counterpart more disk or CD-ROM with electronic printing
than 90 percent of the time. NARA plans on demand. The possibilities are almost
to extend testing of expert systems to endless.
other areas of records management.

TECHNICAL STANDARDS
The pace of development and application of ● the International Committee on Consul-
several of the technologies discussed earlier is tative Telephone and Telegraph (CCITT),
dependent on the development of and agree- which is a unit of the International
ment on national and international standards. Telecommunications Union and whose for-
Standards-setting efforts are underway in all mal members are 160 governments;
critical areas, although the intensity of activ- ● the International Standards Organization
ity varies. The major standards organizations (1S0), whose members are the national
include: standards bodies of 89 countries;
65

the American National Standards Insti- software. However, this approach is consist-
tute (ANSI) that represents the United ent with the usual industry practice for disk
States in the IS0 and coordinates volun- drive standards. Standards for WORM, Erase-
tary standards activities in the United able, and CD-I disks are in earlier stages of
States; development.
the National Bureau of Standards (NBS), Another very important standards area in-
which is the lead U.S. government agency volves the page description and the text
in many standards areas; and markup languages used to code the format,
● the Federal Communications Commi ssion style, and composition of documents. If the
with respect to certain telecommunication text markup language used to prepare a docu-
standards. ment is not compatible with the language used
A new organization, the Corporation for Open by the composition and/or output devices, then
Systems (COS), was established in 1986 to pro- significant additional work is required to strip
mote open systems interconnection standards. the markup commands from the document and
COS members are primarily telecommunica- reinsert the commands in a compatible lan-
tion and information equipment and services guage. Sometimes it is easier just to rekey-
companies. board and recode the entire document, at sig-
nificant additional cost. Alternatively, a page
The following discussion highlights standards- description language can be used to make the
setting for optical disks, page description lan- conversion automatically, if there is page
guages, and test markup languages. For dis- description software compatible with the par-
cussion of other standards areas (e.g. electronic ticular text markup language and output de-
data interchange, integrated digital services), vices in use.
see Communication Systems for an Informa-
tion Age (OTA, forthcoming, 1989). One possible page description language is
the PostScript language (by Adobe Systems)
With respect to optical disk standards, the that is becoming a defacto standard at least
two leading manufacturers of optical disks— for desktop and WYSIWYG publishing sys-
Sony and Phillips-took the lead and developed tems due to the fact that both Apple and IBM,
a set of proposed standards for CD digital au- among others, use PostScript. This possibil-
dio, CD-ROM, and CD-I, known as the Red ity is under consideration by NBS, ANSI, and
Book, Yellow Book, and Green Book stand- 1S0. A related effort involves the development
ards, respectively. The proposed CD-ROM of a Standard Page Description Language
standards (Yellow Book) included detailed (SPDL). These approaches are intended to
technical specifications for CD encoding, match the applications software (e. g., for edit-
mastering, replication, decoding, and reading, ing and composition) to the output devices and
such that any CD-ROM disk can be read by eliminate the need for the so-called device
any CD-ROM disk drive, and have become de driver which is a separate set of instructions
facto industry standards. In addition, stand- needed to make the applications software com-
ards for the logical formatting of CD-ROMs municate with the output device.
were initially proposed by the so-called High
Sierra Group and subsequently adopted by Text markup standards are particularly im-
1S0 as an international standard. Data prep- portant to realize the full benefits of electronic
aration is the one area not fully specified by information dissemination. If government doc-
the proposed standards. While data must be uments (whether reports, pamphlets, manuals,
logically organized, formatted, and prepared other text, or text plus tabular and graphics
prior to conversion into opticaI disk format, material) are not prepared in a standardized
the standard permits use of a wide range of electronic format using standardized codes and
computer operating systems. Although the descriptors, substantial recoding and rekey-
other specifications insure that disks are phys- boarding may be necessary at later stages of
ically readable by any disk drive, the data may the dissemination process. Any significant
not be accessible except through proprietary recoding and rekey boarding is costly and can
66

offset some or all of the cost advantages of elec- facilitate private vendor repackaging or en-
tronic formats. hancing of agency documents, if the vendors
Text markup standards are intended to es- utilized the same standards.
tablish a consistent set of codes for labeling Three major approaches to text markup
key elements of a document–such as chapter standards are: 1) GPO’s logically structured
titles, paragraph indentations, tabular presen- full text database standard; 2) the Standard
tations, and the like. Such standards establish Generalized Markup Language (SGML) that
a logical structure for the elements of a docu- has been adopted by the Department of De-
ment, in a hierarchical order—such as chap- fense; 3) and the Office Document Architec-
ter, paragraph, line, word, and character. The ture standard.
elements are assigned codes (which can be a
letter, number, symbol, or combination thereof) The GPO standard is used almost exclu-
that are keystroked along with the text, ta- sively by GPO, congressional committees and
bles, and graphics included in the document. offices, and Federal agencies—primarily those
If these electronic codes are widely agreed upon agencies that submit magnetic tapes to GPO
and used (i.e., standardized), then the docu- for typesetting and printing. Full text data-
ments can be electronically transferred from base standard or specification is the applica-
one stage in the dissemination process to tion of a logical coding structure to the full text
another with little or no additional effort. or content of the document, including tables
as well as text. GPO staff recently completed
A wide range of information dissemination training on how to write software programs
functions would be facilitated by text markup that can translate from SGML to GPO’s full
standards, including: text database standard. GPO indicates that
● authoring it is prepared to write such software at cus-
—creating the document tomer request. The GPO standard is designed
—editing primarily to meet the needs of publishing
—revising professions. SGML, on the other hand, while
● archiving also meeting publishing needs, is recognized
— short-term as an international standard, endorsed by DoD
–long-term and some vendors, and is being issued by NBS
● disseminating in multiple formats as a Federal Information Processing Standard
—conventional printing (FIPS). SGML is a set of rules for developing
—electronic printing-on-demand the element codes for a document, whereas the
—online electronic GPO standard includes both the rules and the
—offline electronic (e.g., magnetic tape, specific codes themselves. Both SGML and
floppy disk, CD-ROM) GPO use a logical structure, so in principle
—microform SGML codes should be convertible to GPO’s
—specialized outputs (e.g., braille, foreign codes, and perhaps vice versa, although some
languages, voice) of these applications are still under devel-
● disseminating through multiple channels opment.
—agency clearinghouses and information
centers Office Document Architecture (ODA), a re-
lated protocol, is directed primarily to meet-
—governmentwide clearinghouses and
ing office, not publishing, needs, and thus the
sales programs
document complexity is reduced (due to fewer
—press, libraries, and commercial vendors
fonts, formats, etc.). ODA is a method of en-
For example, text markup standards would coding software that essentially converts doc-
help ensure that NTIS and/or GPO are able uments to a common code compatible with a
to efficiently reproduce and disseminate wide range of office automation systems. ODA
agency electronic documents. This would also is a protocol for converting the codes used to
67

format individual documents into a common need for rapid 0S1 implementation. In the
format for the interchange of the documents United States, NBS is coordinating an 0S1
among different systems. ODA was initially prototype system known as OSINET that is
defined by the European Computer Manufac- intended to be a test of the 0S1 reference
turers Association (ECMA) to be consistent model. The results are being made available
with the Open Systems Interconnection stand- to the standards-setting organizations.
ard developed by the 1S0 and has been issued
The Federal Government commitment to 0S1
as an international standard. Officials at NBS
is already significant, with a growing con-
believe that there may be a need for both ODA
sensus that 0S1 is necessary to move to in-
and SGML standards within the Federal Gov-
teroperability of the now confusing and largely
ernment.
incompatible range of equipment and software
Finally, there is intensive work by all major in the government inventory. Indeed a Fed-
standards organizations to refine and imple- eral interagency committee has recommended
ment the open systems interconnection (0S1) that 0S1 standards be mandatory for new Fed-
concept. An 0S1 reference model has been de- eral computer and telecommunication procure-
veloped under 1S0 auspices. The model serv- ments and be a first option for retrofits of
ices as a master standard for an integrated existing systems. The suggested 0S1 procure-
telecommunications-information systems envi- ment standard would be consistent with the
ronment. It also incorporates already estab- 1S0 reference model. This 0S1 procurement
lished standards such as those for packet- standard is being issued by NBS as a Federal
switched data networks and electronic mail. Information Processing Standard (FIPS).
Many vendors and users have recognized the
Chapter 4

Alternative Futures for the


Government Printing Office

Clockwise from top left: a hot type scene at GPO, circa 1940s; GPO operator using electronic photocomposition equipment;
GPO operators using keyboard input terminals; and the Congressional/ Record coming off the press (photo credit:
U.S. Government Printing Office).
CONTENTS
Page Table Page
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4-17, Origination Formats for Material .
Traditional GPO-Centralized . . . . . . . . . . . . 73 Submitted to GPO, as Percent of
Demand for Traditional GPO Services . . . 74 Total. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Impacts of Medium-Term Reductions in 4-18. Origination Formats, Including
Traditional Demand . . . . . . . . . . . . . . . . . 76 Camera-Ready, as Percent of Total
Traditional GPO–Legislative Branch and by Branch of Government. . . . . . . 86
Only . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 4-19. Civilian Department Agency Use of
Financial Impacts . . . . . . . . . . . . . . . . . . . . 81 Selected Electronic Technologies . . . . . 87
Labor Force Impacts . . . . . . . . . . . . . . . . . . 82 4-20, Estimated Use of Electronic
Other Vulnerabilities . . . . . . . . . . . . . . . . . . 83 Publishing Software, Calendar
Electronic GPO–Decentralized. . . . . . . . . . . 85 Year 1987 . . . . . . . . . . . . . . . . . . . . . . . . 87
Trends in Technology and Demand . . . . . . 86 4-21, Estimated Use of Other Electronic
Opportunities and Challenges. . . . . . . . . . . 89 Publishing Technologies, Calendar
Year 1987 . . . . . . . . . . . . . . . . . . . . . . . . 87
Tables 4-22, Depository Library Demand for
Federal Information, by Type and
Table Page Format , . . . . . . . . . . . . . . . . . . . . . . . . . . 88
4-1, GPO Revenues for Printing Services - 4-23, Scientific and Technical Association
and Publications Sales, Fiscal Years Demand for Federal Information, by
1978-87 ....., . . . . . . . . . . . . . . . . . . . . . 74 Type and Format . . . . . . . . . . . . . . . . . . 89
4-2. GPO Billings and Labor Force 4-24. Agency Activities and Plans for
Breakdown, Fiscal Year 1987 . . . . . . . . 75 Electronic Information Dissemination,
4-3. GPO Workload Distribution, Fiscal by Type and Format . . . . . . . . . . . . . . . 90
Year 1987 . . . . . . . . . . . . . . . . . . . . . . . . 75 4-25. GPO Pages Produced from Databases
4-4. Electronic Potential for Main Plant and Direct Drive Magnetic Tapes,
Products . . . . . . . . . . . . . . . . . . . . . . . . . 76 Fiscal Years 1983-87 . . . . . . . . . . . . . . . 91
4“5. Cost of GPO Work, 20 Sample Jobs . . 77 4-26. GPO Dial-Up Electronic Transmission
4-6. GPO Cost Structure, Fiscal Customers, January 1988 . . . . . . . . . . . 92
Year 1987 . . . . . . . . . . . . . . . . . . . . . . . . 78 4-27, Departmental Applications of GPO
4-7. GPO Main Plant Cost Structure, Structured Full Text Database
Fiscal Years 1984 and 1985 . . . . . . . . . 78 Standard, as of November 1987 . . . . . . 93
4-8. Major Changes in GPO Labor Force, 4-28. Ten Largest GPO Printing Customers,
Fiscal Years 1975-87 . . . . . . . . . . . . . . . 79 Fiscal Year 1986 . . . . . . . . . . . . . . . . . . . 94
4-9. Selected Major Equipment, 4-29. Distribution of GPO Defense
Acquisitions, GPO Press and Bindery, Customers by Procurement and
Since 1977, as of September 1987 . . . . 80 Printing Offices, Fiscal Year 1986 . . . . 94
4-1o. Hypothetical Calculation of Financial 4-30. Federal Agency Electronic Publishing
Impact of Legislative Branch GPO . . . 82 Activities and Plans, as of 1986 in
4-11. Hypothetical Main Plant Total Labor Percent of Agencies Responding . . . . . 96
Force Reductions . . . . . . . . . . . . . . . . . . 83 4-31 0 Depository Library Demand for
4-12. Hypothetical Main Plant Divisional
Federal Information Electronic
Labor Force Reduction ... , ., . . . . . . . 83 Formats . . . . . . . . . . . . . . . . . . . . . . . . . . 98
4-13 ( GPO Main Plant Composers, Years of 4-32. Scientific, Technical, and General
Service and Retirement Eligibility, Association Demand for Federal
Fiscal Year 1987 . . . . . . . . . . . . . . . . . . . 84 Information Electronic Formats. . . . . . 99
4-14. Changes in GPO Main Plant Volume 4-33. Library and Association Access to
for Eight Principal Products, Information Dissemination Technology 99
1975-84 . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 4-34, GPO Union Bargaining Units, as of
4-15. Congressional Printing and Binding April 1987 . . . . . . . . . . . . . . . . . .......101
Billings Selected Items, Fiscal Years 4-35. Selected GPO Electronic Composition
1975, 1983, and 1984 ........, . . . . . . 85 Equipment, Fiscal Year 1987 ., .. ....103
4-16. Congressional Printing and Binding
Billings, Selected Items, Fiscal Years
1975 and 1987 . . . . . . . . . . . . . . . . . . . . . 85
Chapter 4

Alternative Futures for the


Government Printing Office

SUMMARY
This chapter along with chapter 5 examines On one hand, near-term (l-3 years) future de-
in detail selected alternatives for the future of mand for traditional GPO services is likely to
the U.S. Government Printing Office (GPO) be stable, absent a severe governmentwide fis-
and National Technical Information Service cal crisis and assuming executive agencies con-
(NTIS). Chapter 4 focuses on GPO, and chap- tinue to be required to obtain printing from
ter 5 on NTIS and NTIS/Superintendent of or through GPO. GPO’s greatest assets are
Documents (SupDocs) cooperation. The intent its traditional ink-on-paper printing facilities
is to gain a full understanding of the strategic and experienced labor force, coupled with a
outlook for both GPO and NTIS as an input substantially automated prepress capability,
to congressional decisions on their future direc- including electronic input, photocomposition,
tion. This strategic outlook should be relevant and typesetting. GPO has invested heavily
to congressional consideration of any alterna- over the past decade in upgraded prepress,
tives for GPO and NTIS, not just the ones ex- press, and bindery facilities. Despite the pos-
plicitly discussed here, and should be read in sibly misleading external appearance, the GPO
the context of the trends discussed in chap- main plant compares favorably with even the
ters ‘2 and 3 of this report. The results of this largest and best equipped commercial print-
strategic analysis are highlighted below. Pol- ing plants.
icy implications are discussed in chapters 11
and 12. On the other hand, GPO is particularly vul-
nerable in the medium-term (3-5 years) to
First, General Accounting Office (GAO) sur- changes in demand for paper formats that
veys of Federal agencies (chapter 2) and Fed- might reduce executive agency need for GPO
eral information users (this chapter), coupled procured printing (which accounts for about
with a review of agency automation plans and three-quarters of GPO’s total printing work)
activities, suggest the following overall pro- and for GPO main plant inplant printing
jections: (where the bulk of GPO overhead and labor
c 1 to 3 years-steady state in demand for costs are located). The future of GPO printing
paper formats; rapid growth in electronic depends in large measure on the plans and
formats, but still a very small percentage activities of Department of Defense (DoD)
of total demand; agencies, that collectively account for roughly
3 to 5 years-demand for paper formats one-third of all GPO billings. The defense agen-
may start to decline; demand for electronic cies are determined to reduce drastically their
formats likely to reach critical thresholds dependence on paper formats within the next
for several types of Federal information; few years. If traditional ink-on-paper defense
● 5 to 10 years-demand for paper formats work was phased out, almost all of the GPO
likely to decline markedly in some cate- regional and satellite procurement offices
gories, but would still be significant for would lose at least half and some over 90 per-
traditional government books, reports, cent of their work, and two of the GPO regional
and publications; electronic formats likely printing plants would lose perhaps one-third
to dominate for many types of infor- to one-half of their work. Realistically, the de-
mation. fense conversion from paper to electronic for-

71
72

mats could take longer than planned, but war- ing process by encouraging electronic sub-
rants careful monitoring due to the highly missions, already at high levels, and dial-
leveraged effects on the GPO regional work up composition services where appro-
load. The GPO main plant is not dependent priate;
on military work, but is vulnerable due to the ● encourage adoption of governmentwide
combination of potential electronic competi- structured database standards for elec-
tion for some major product lines (such as the tronic (as well as conventional) printing;
Congressional Record and Federal Register) ● facilitate mechanisms for training and
and a high overhead cost structure necessi- education about electronic publishing;
tated largely by GPO’s current diversified ● establish an electronic publishing labora-
responsibilities. tory and innovation center, open to agency
personnel; and/or
In the hypothetical case that GPO were to ● increase SupDocs dissemination of elec-
be limited to legislative branch printing (plus tronic formats.
some key governmental process work such as
the Federal Register, passports, and postal Up to now, GPO has participated in only a
cards), the main plant printing operations handful of agency automation programs. From
would switch from a net surplus of several mil- a strategic perspective, GPO would benefit from
lion dollars to a net loss of several tens of mil- staying abreast of agency applications and,
lions of dollars, assuming no changes in de- thereby, be in a much better position to identify
mand and overhead, cost, and labor force opportunities to meet agency needs. In a decen-
structure (other than those associated with tralized and competitive electronic information
transferring the GPO printing procurement environment that increasingly characterizes the
program to an executive agency such as the Federal Government, GPO will have to be inno-
General Services Administration (GSA)). Re- vative in matching its expertise to agency needs
storing GPO to breakeven operations under that are likely to vary widely and change at an
this scenario could necessitate up to a 40 per- increasingly rapid pace.
cent reduction in the main plant labor force
and/or a significant increase in rates. For the legislative branch, GPO already has
a central role in many traditional publishing
GPO is faced with several challenges and op- activities and several electronic publishing pi-
portunities concerning electronic publishing lot projects. GPO could develop plans for an
and dissemination of electronic formats. Fed- expanded role for congressional committees
eral executive agencies are rapidly increasing and offices, including electronic search, re-
their automation activities, and have already trieval, and printing-on-demand of congres-
invested, collectively, an estimated $400 mil- sional documents. Any detailed planning
lion in electronic publishing systems. GAO sur- would need to take into account the related
veys found dramatic increases in the percent- roles of the Library of Congress and Congres-
ages of both agencies and information users sional Research Service, House Information
that anticipate use of electronic formats (espe- Systems Office, and Senate Computer Center.
cially online data bases, electronic mail or bulb
With respect to GPO SupDocs sales of elec-
tin boards, floppy disks, and compact optical
tronic formats, SupDocs would be operating
disks) over the next 3 years. in a more competitive environment than has
For the executive branch, several roles for traditionally been the case with respect to pa-
GPO are emerging beyond the continued pro- per formats. For many types of Federal infor-
mation, individual agencies and/or private
vision of traditional printing services. GPO
could: vendors might decide to market electronic for-
mats. SupDocs would need to decide which
● continue to improve cost-effectiveness at electronic items would be cost-effective and
the input and prepress end of the print- competitive if included in the sales program.
73
.

Three policy issues would need resolution. however that role may be defined. With respect
to capital investment, it would seem prudent
● First is that significant SupDocs sales of for GPO to carefully reevaluate its capital in-
magnetic tapes and floppy disks (and, po-
vestment plans in light of possible adjustments
tentially, CD-ROMs and electronic print-
to traditional printing services and possible
ing-on-demand) could overlap and dupli-
new electronic initiatives. Staffing and capi-
cate the NTIS sales program, absent a
tal investment decisions are best made within
consolidation of or close cooperation be-
an overall strategic framework for the future
tween NTIS and SupDocs.
of GPO.
● Second is that SupDocs sales of online
databases could overlap and duplicate The discussion in this chapter focuses on
offerings by individual agencies, agency technical, financial, organizational, labor, and
online gateways (such as the National Li- demand questions and not on the broad pol-
brary of Medicine), and/or private (or non- icy issues addressed in chapters 11 and 12. The
profit) online gateways or database pro- alternatives presented here assume that Con-
viders that include Federal agency data- gress would take whatever policy actions were
bases, absent agency agreements. While necessary to implement the particular alter-
multiple government sales outlets for the native. In other words, this chapter is intended
same tapes and disks may be cost-effec- to probe more deeply into several of the possi-
tive, given the small breakeven volumes, ble alternatives for GPO. Also, although the
multiple government outlets for sales of alternatives are framed in terms of GPO, the
online services may be hard to justify, discussion applies equally regardless of the
considering the more substantial develop- name, for example, Government Information
ment, staff, and capital investment re- Office or Government Publications Office.
quirements. The alternatives explored in this chapter are:
● Third, sales of electronic formats could in-
volve heavier demands for user support . Traditional GPO—centralized
and generate the need for far more sophis- Traditional GPO—legislative branch only
ticated and extensive customer service— ● Electronic GPO—decentralized
from GPO, agencies, and/or vendors—
Each of these alternatives is defined and dis-
than is expected for paper (or microfiche)
cussed below. The order of discussion does not
formats.
imply an order of preference or priority, but
GPO faces two major challenges with respect was chosen to facilitate the presentation. Key
to staffing:retaining the necessary skilled labor facts and analyses are incorporated at the first
force to maintain traditional printing services appropriate place and then referenced in sub-
at a level commensurate with demand, and ob- sequent discussion rather than repeated. Dis-
taining personnel with the new skills needed to cussion relevant to SupDocs is also found in
implement GPO’s future role in electronic pub- chapter 5, which focuses on an electronic NTIS
lishing and electronic information dissemination, and NTIS/SupDocs cooperation.

TRADITIONAL GPO–CENTRALIZED
Under this alternative, GPO would continue pository Library Program (DLP). GPO would
to provide centralized conventional printing do very little electronic dissernination, as is the
services (that is, Federal Government ink-on- situation today. The mission agencies would
paper printing would be obtained from or handle electronic dissemination themselves, in-
through GPO), disseminate paper formats on eluding direct distribution to the depository
a sales basis through SupDocs, and dissemi- libraries to the extent needed. Note that print-
nate paper and microfiche formats to the De- ing services are defined to include composition,
74

printing, binding, blank paper sales, and re- However, in the longer term, significant re-
lated activities. ductions in paper formats could occur. OTA’S
independent printing consultant reviewed
Demand for Traditional GPO Services GPO’s current product line and, using several
different methodologies, concluded that about
A logical starting point for the analysis is 60 percent of GPO’s current products could
to examine demand for traditional GPO serv- potentially be suited for electronic formats, al-
ices. The two major components of demand are though realistically perhaps only one-half of
printing services, and publication sales. In fis- this amount (or 30 percent) would be suited
cal year 1987, these accounted for 88.7 percent for electronic dissemination and even this
and 8.4 percent of total revenue, respectively. would not automatically mean that paper dis-
The 10-year trend data for these two items are semination would be eliminated.1 Any actual
shown in Table 4-1. Clearly, other than the tem- switch from paper to electronic formats would
porary decline in printing services during fis- most likely take place gradually, since elec-
cal year 1981 and fiscal year 1982 (almost half tronic dissemination requires that the recipi-
of which was due to reductions in congressional ent (user) as well as the sender have the neces-
work), the overall trends show a gradual in- sary equipment and knowhow. On the other
crease in printing services and a rather steady hand, results of the GAO survey of Federal
increase in sales of publications. Even account- information users, highlighted later in this
ing for inflation, there is no historical evidence chapter, indicate that many users desire to in-
of weakness in the demand for traditional GPO crease dramatically their use of electronic for-
services. mats within the next 3 years. These findings,
Looking to the future, most independent pro coupled with the ambitious automation plans
jections suggest that overall general demand and activities of many Federal agencies, sug-
for paper formats will continue for at least 5 gest the following projections:
years at a slow growth or, at worst, steady ● 1 to 3 years-steady state in demand for
state level—even in the face of rapid growth paper formats; rapid growth in electronic
in electronic formats. This projection should formats, but still a very small percentage
apply to the Federal Government as well, short of total demand;
of a severe fiscal crisis. There is no evidence ● 3 to 5 years-demand for paper formats
that agency budgetary restraints in the past may start to decline; demand for electronic
few years have translated into a significant re- formats are likely to reach critical thresh-
duction in actual printing services obtained olds for several types of Federal infor-
from GPO. mation;
● 5 to 10 years-demand for Paper formats
are likeiy to decline markedly {n some cat-
Table 4“1 .—GPO Revenues for Printing Services and egories but would still be significant for
Publications Sales, Fiscal Years 1978-87 traditional government books, reports,
(in millions of dollars)
and publications; electronic formats are
Sales of likely to dominate for many types of in-
Fiscal year Printina services Duplications formation.
1978 . . . . . . . . . . . $499 $44.4
1979 . . . . . . . . . . . 606 44.4 However, even though there is not likely to
1980 . . . . . . . . . . . 672 47.6 be a precipitous near-term decline in overall
1981 . . . . . . . . . . . 644 51.3 demand for paper formats, GPO is particularly
1982 . . . . . . . . . . . 608 55.0
1983 . . . . . . . . . . . 637 57.1 vulnerable to changes in demand for products
1984 . . . . . . . . . . . 739 59.4
1985 . . . . . . . . . . . 771 59.3
1986 ... , . . . . . . . 737 62.9 ‘Frank Romano,, “Decision Analysis Framework for GPO
1987 . . . . . . . . . . . 773 73.5 Strategic Alternahves,” contractor report prepared for OTA,
SOURCE: U S Government Printing Office, 1988 January 1988.
75
-.

that are printed at the main GPO plant in As shown in Table 4-3, about 80 percent of all
Washington, DC. This is because the bulk of legislative branch work is done inplant, while
GPO overhead and labor are located at the about 85 percent of all executive branch work
main plant, and also because about three- is contracted out. Of the 20 percent (or $23 mill-
quarters of GPO’s total printing work is con- ion worth) of legislative branch printing that
tracted out (known as procured printing). is procured, only about $1 million is for Con-
These figures are highlighted in Table 4-2 for gress itself, with the remainder for legislative
fiscal year 1987. branch agencies and extra copies of agency doc-
uments for SupDocs and DLP. Also, about 45
In addition to the 3,500 personnel allocated
percent of inplant work is legislative, while
in Table 4-2 to procured, main plant, and re-
about 95 percent of contracted work is for the
gional printing, there are 692 administrative
executive branch. Judicial branch work is split
and support personnel located primarily at the
about 50-50 between inplant and procured
main plant and 930 personnel assigned to the printing, but represents only a fractional per-
SupDocs office. The SupDocs personnel are centage of total GPO work compared to about
supported through sales revenues, appropria- 15 percent for the legislative branch and 85
tions (for DLP and by law dissemination), and
percent for the executive branch. Also, over
agency reimbursements (for reimbursable dis-
90 percent of inplant work is done at the main
semination), and are not counted as part of plant, with the remainder at GPO regional
GPO overhead. However, the administrative plants. Complete fiscal year 1987 workload
and support personnel, plus main plant main- data are presented in Table 4-3. Again, note
tenance, utilities, and the like, are included in that the term “printing’ is defined to include
general overhead, which is allocated across all composition, layout, printing, binding, blank
major GPO activity centers. paper sales, and other associated services in
Any reduction in the GPO work load would addition to printing.
result, at least in the short run, in spreading An analysis of fiscal year 1987 billing data
the general overhead over a smaller base and, for the GPO main plant indicates that a sig-
thereby, increasing unit costs. Moreover, re- nificant portion could be suitable for electronic
ductions in the main plant work load would dissemination or could be vulnerable to com-
have a magnified impact since the high costs petition from electronic formats. The major
of main plant operations would be allocated items are listed in Table 4-4 with fiscal year
over a smaller base of main plant work thus 1987 billing amounts indicated. Other signifi-
driving up the unit costs even further, all other cant main plant billing items which are judged
things being equal. as not suitable for electronic formats include
Main plant operations are particularly vul- such things as envelopes, books, letter head
nerable to changes in the legislative branch stationery, note pads, passports, and postal
work load, which is concentrated at that plant. cards. Some main plant billings are for person-
nel services only (e.g., Congressional Record
Table 4-2.—GPO Billings and Labor Force Breakdown,
Fiscal Year 1987
Table 4-3.—GPO Workload Distribution,
Main Regional Fiscal Year 1987 (in millions of dollars)
Procured plant plant
printing printing printing Main Regional
Billings ($ millions) . . . . . . . $576 $180 $14 Procured plant plant
(Percent of total). . . . . . . . . . 74.8°\0 23.40/o 1.80/0 mintina Drintino ~rintino Totals
Labor force assigned Legislative branch . . 23 90 – 113
(persons) . . . . . . . . . . . . . . 637 2,619 a 244 Judicial branch . . . . . 1 l — 2
(Percent of total). . 18.2°/0 74.80/o 7,0 ”/0 Executive branch . . . 552 90 14 656
a
Excludes 692 admtnlstratwe and support personnel and 930 SupDocs personnel Totals . . . . . . . . . . . $576 $181 $14 $771
SOURCE U S Government Prlntlng Off Ice 1988 SOURCE. U.S Government Printing Off Ice, 1988
76

Table 4=4.—Electronic Potenfial for Main Plant Products

Fiscal year 1987


billings
Major product (in $ millions) Electronic potential
U.S. Code . . . . . . . . . . . . 1.4 Online and CD-ROM distribution could reduce demand for paper format.
Forms . . . . . 5.8 Electronic filing and electronic printing-on-demand could reduce demand for
paper formats in medium to long-term.
Pamphlets . . . . . . . 11.5 Online, CD-ROM, diskette, and electronic mail/bulletin board distribution
along with electronic printing-on-demand could reduce demand for paper
formats in medium-term.
Bills, resolutions,
amendments . . . . . . . 11.0 Online authoring, editing, publishing, and status systems along with online
and CD-ROM distribution could significantly reduce demand for paper
formats.
Calendars . . . . . . . . . . . . 2,0 Online systems could reduce demand for paper formats.
Codeof Federal
regulations . . 8,6 Online and CD-ROM distribution could reduce demand for paper formats.
Committee prints . . . . 3.5 Online, CD-ROM, and diskette distribution along with electronic printing-on-
Committee reports . . . 4,2 demand could reduce demand for paper formats.
Congressional Record-
Daily . . . . . . . . . . . . . . . 13,0 Online and CD-ROM distribution could sharply reduce demand for paper
formats.
Federal Register ..., . 17.7 Online and CD-ROM distribution could sharply reduce demand for paper
formats.
Hearings . . . . . . . . . . . . . 17.6 CD-ROM distribution could reduce demand for paper formats,
Total, . . . . . . . . . . . . . 96.3
SOURCE Bllltngs from GPO: Electronic Potential from OTA, 1988

indexers at $0.75M, details to congressional run (e.g., a few thousand copies) of paper co-
committees at $6 M). pies and possibly some microfiche copies (on
a transitional basis). Any subsequent dissem-
Overall, just over half of the main plant work ination of paper copies could be on a printing-
could be affected by electronic formats. The on-demand basis for complete copies or, prob-
vulnerable congressional work is particularly
ably more common, printing of selected pages.
significant and amounts to about 45 percent
of main plant billings if the Federal Register
and Code of Federal Regulations (CFR) are in- Impacts of Medium-Term Reductions
cluded. In principle, reductions in inplant work in Traditional Demand
could be offset by shifting some procured work
to in-plant printing. However, the congres- Realistically, any significant transition from
sional work is quite unique and specialized, paper to electronic formats would take place
with little work of a similar nature currently over several years, so GPO would have time
being procured, unlike forms and pamphlets to adjust. Basically, GPO could make up for
where work could be rather easily shifted from any shortfall by transferring a portion of pro-
procured to in-plant printing. Also, once ini- cured printing (primarily executive branch
tial press runs reached small enough levels, cer- work) to the main plant or reducing main plant
tainly in the few hundreds of copies and possi- operating costs or some combination of the
bly in the few thousands, fully electronic two. The major drawback of transferring more
composition and printing would likely be cost- work in-house is that the main plant work costs
effective. In other words, primary dissemina- significantly more than procured work. Thus,
tion could shift from paper (or paper and mi- either the executive agency customers would
crofiche) to electronic with a small initial press pay considerably more than they do now, or
77
—————. .

GPO would have to charge considerably less There are several reasons for the higher main
than full cost to offer a competitive price. plant costs. First and foremost, GPO is a
unique printing facility in terms of product
GPO Cost and Labor Force Structure mix, schedule requirements, and customer
The cost of GPO work for 20 sample print- base. GPO produces a much more diversified
ing jobs is shown in Table 4-5 in cost per 100 set of printing products than any single pri-
pages and total cost, averaged over all 20 jobs. vate sector printing facility. As a consequence,
For these 20 sample jobs, the average main GPO operates and maintains a much wider
plant regular rate cost was more than double range of equipment than do private printing
the procured cost. While these 20 jobs do not companies. Most private firms specialize in a
constitute a statistically valid sample of all small number of products, to keep overhead
GPO work, the sample jobs were selected by down and maximize economies of scale. Sec-
GPO as being reasonably representative. (See ond, to provide quick turnaround of congres-
ch. 11 for further details. ) sional work and overnight printing of the daily
Congressional Record and Federal Register,
In recent years, GPO has been offering a spe- GPO operates on a three shift basis. This re-
cial rate for some executive branch printing sults in significantly higher costs for staffing,
jobs done at the main plant. The rate is based supervision, maintenance, and general over-
on the tenth lowest bid for comparable pro- head. The overnight operations are so impor-
cured work plus ten percent. For the 20 sam- tant that, in 1987, GPO designated an Assis-
ple jobs, the average main plant cost using the tant Public Printer to provide overall on-site
special rate was about 45 percent higher than management of the night operations. Third,
the procured cost. The special rate is intended as a government agency, GPO provides such
to recover variable costs and make some con- services as employee and congressional rela-
tribution to general overhead. This would ap- tions, public affairs, inspector general, equal
pear to be the case, since the special rate for employment, labor relations, safety and health,
the 20 sample jobs averaged about 68 percent and the like, many of which contribute to
of the regular rate that presumably covers full higher general overhead than in private com-
costs. According to GPO, the average direct panies.
labor rate is about 30 percent of full costs. So
the special rate does appear to more than cover Overall, GPO is a labor intensive organiza-
tion. After deducting the cost of procured
direct labor. If one assumes direct labor to be
printing and sales of publications and the sur-
a fixed cost, at least in the short-term, the spe-
plus (net profit), about two-thirds of the re-
cial rate appears to easily cover the cost of ex-
maining costs are for labor, about one-fifth for
pendable (e.g., paper, ink) PIUS make a Con-
supplies and materials, and one-tenth for util-
tribution to overhead. Of course, on the other
hand, the greater the use of the special rate, ities and the like. The GPO cost structure,
based on fiscal year 1987 data, is shown in Ta-
the greater the overhead rate will be for the
ble 4-6. Data for main plant costs shown in Ta-
balance of the work, all other things being
ble 4-7 confirm the general cost structure noted
equal.
above.
With respect to the total GPO labor force,
Table 4-5.—Cost of GPO Work, 20 Sample Jobs a significant downsizing has already taken
Main plant Main plant Main plant place. Over the past 12 years, total GPO em-
procured a regular rateb special ratec ployment has declined by about 3,500 persons
T o t a l c o s.-t , . $1oo,o17 $213,281 $144,881 or 40 percent (from 8,632 in fiscal year 1975
to 5,122 in fiscal year 1987). As shown in Ta-
a
Procured esttmates based on general usage contracts us I ng the average pr!ce
of the f!rst 5 lowest b!dders
bMaln plant regular rate estimates based on the GpO Price scale as of Dec
1
ble 4-8, the reductions have been spread across
1987
cMaln plant special rate estimates based on the 10th lowest b!d plus 10°. several GPO activities, but with the highest
SOURCE U S Government Prlntlng Off Ice 1988 absolute and percentage reductions occurring
78

Table 4-6.—GPO Cost Structure, Fiscal Year 1987 severe fiscal crisis on the part of the customer
agencies or some other circumstance that
Percent of
fiscal year 1987 would precipitate a rapid decrease in conven-
Cost element revenue dollar tional printing activity, and assuming the ex-
Procured printing . . . . . . . . . . . . . . . . . . . . . . 61.5 ecutive agencies continue to be required to ob-
Sales of publications . . . . . . . . . . . . . . . . . . 2.3 tain printing from or through GPO.
Surplus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6
Subtotal . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66.4 On the other hand, the GPO main plant ap-
Labor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22.4 pears to be vulnerable in the medium-term (3-
Supplies and materials . . . . . . . . . . . . . . . . . 6.1
Rents, communications, and utilities . . . . . 3,3 5 years) and beyond due to the combination
Capital expenditures . . . . . . . . . . . . . . . . . . . 0.8 of electronic competition for some major prod-
Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.0 uct lines (such as the Congressional Record and
Subtotal . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33.6
Federal Register), and a high overhead cost
Percent of fiscal year 1987
revenue dollar less cost structure necessitated largely by GPO’s cur-
of procured printing and rent responsibilities. Additional executive
sales plus surplus agency work could be shifted from private
Labor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66.6 printing companies to the GPO main plant, but
Supplies and materials . . . . . . . . . . . . . . . . . 18.2
Rents, communications, and utilities. . . . . 9.8 this would likely increase the cost to the agen-
Capital expenditures . . . . . . . . . . . . . . . . . . . 2.4 cies. GPO could charge a special, lower rate
Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.0 for most agency work, but this would mean
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100,0 some portion of overhead would be uncovered
SOURCE” US Government Printing Office and Office of Technology Assessmen~
1988 and have to be paid out of direct appropria-
tions or, possibly, reimbursed from net reve-
nues on sales of publications. These latter alter-
Table 4.7.—GPO Main Plant Cost Structure, natives would appear to require amendment
Fiscal Years 1984and 1985
of the relevant provisions of Title 44 of the U.S.
Percent of total costs Code. On the other hand, it is possible that cost
Fiscal year Fiscal year reductions resulting from the provision of elec-
Cost element 1984 1985 tronic alternatives to the Record and Regis-
Labor. . . . . . . . . . . . . . . . . . . . . . . . 65.0 66.8 ter (among other publications) could offset any
Supplies and materials . . . . . . . . . 26.2 25.0
Rents, communications, and
cost increases that might result by shifting
utilities . . . . . . . . . . . . . . . . . . . 4,3 4.4 more executive agency work from procured to
Depreciation . . . . . . . . . . . . . . . . . . 3.3 2.6 inhouse printing.
Other . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 1.2
100.0 100.0 GPO could attempt to further reduce oper-
SOURCEUS Government Printing Office and Office ofTechnology Assessment, ating costs, but this may be difficult given the
1988
already substantial labor force reductions ob-
tained since the mid 1970s, absent a basic res-
in the composition, printing, binding,procure- tructuring of GPO responsibilities and opera-
merit, persomel management, and documents tions. Some additional labor cost reductions
sales/distribution areas. Most of these labor are likely to occur as agencies assume greater
force reductions resulted from advances in responsibility for composition and other pre-
printing technology and improvements in man- press functions, as a consequence of desktop
agement efficiency. The reduction in person- and high-end electronic publishing capabilities.
nel management in part reflects the reassign- However, in other production areas, such as
ment of apprentices from personnel to the press and binding, projected retirements could
appropriate operating units. create an actual labor shortage, according to
GPO. OTA’S independent labor consultant
Medium-Term Outlook concluded that GPO workforce is relatively
To sum up,near-termdemandfortraditional old (average age of 45.2 years) and that 13 per-
GPO services appears to be stable, absent a cent of the work force (687 persons) is eligible
79

Table 4-8.–Major Changes in GPO Labor Force, Fiscal Years 1975-87



Fiscal year Fiscal year Net change
Selected labor force categories’ 1975 1987 Number - Percent
Document sales/distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,833 930 – 903 – 49:3
Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,632 616 – 1,016 –62.3
Binding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,166 630 – 536 –46.0
Press (includes prepress) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,006 701 –305 –30.3
Printing procurement (excludes regional) . . . . . . . . . . . . . . . . . . 718 432 –286 –39.8
Engineering and facilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490 359 –131 –26.7
Personnel (includes apprentices in fiscal year 1975, but not in
fiscal year 1987) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 104 –228 –68.7
Financial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 250 –117 –31.1
Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 205 –64 –23.8
Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 79 –45 –36.3
‘L’b~ r force ~ategor,e~ ~ele~ted and defined to permit FY75-87 comparison; may not correspond exactly with current labor force cate90rles
bAsof Aug 5, 1987
SOURCE US Government Pnntlng OffIce and OffIce of Technology Assessment 1988

to retire immediately. About 35 percent of the would require much more acreage and would
GPO work force has over 20 years of service.’ probably have to be located much further away
There is also the possibility of reducing GPO’s from GPO’s customer base in Congress and
overhead costs, discussed later. the executive agencies. Overall, OTA’S print-
ing consultant concluded that the GPO main
GPO Plant and Equipment plant is equal to most commercial printing fa-
cilities handling a comparable volume of work.
Other areas of possible cost savings include
the purchase or construction of a new main However, if viable options become available
plant building, and the upgrading of conven- to GPO, a detailed evaluation would be war-
tional prepress, press, and binding technology. ranted, taking into account all the factors
A 1982GA0 study identified numerous ineffi- mentioned above and others, especially any
ciencies in GPO’s facilities-including mate- strategic decisions on GPO’s future role in con-
rials handling, storage, and production flow ventional printing and electronic publishing
problems at the main plant. OTA’S inde- and dissemination. One GPO building option
pendent printing consultant examined al lof currently under consideration involves a plan
these areas and concluded, first, that there is to transfer GPO’s main plant to GSA in ex-
no compelling need for a new plant.Thepresent change for the construction of a new plant on
building was specifically built to handle the property in the Washington Navy Yard, and
load factors of the printing process, whereas to relocate SupDocs to a site in Suitland, Mary-
very few commercial printing facilities were land. As noted in the 1982 GAO report, cost-
originally designed for printing. Most new benefit analyses of all serious alternatives are
plants are on one floor, rather than a multi- warranted, including continued renovation of
floor facility such as GPO’s, and do offer some the existing main plant building, as well as
construction of a new building. Cost-benefit
production efficiencies not currently available
studies would appropriately include consider-
to GPO. However, continued renovation and
upgrading of the main plant elevators should ation of the impacts on the cost, quality, and
help compensate. Also, a single level building timeliness of GPO main plant services, produc-
tivity of GPO main plant operations, and, to
‘Gregory Giebel, *’Technological Changes at the Government the extent possible, GPO’s general morale,
Printing Office, ’ contractor report prepared for OTA, January sense of direction, and strategic outlook.
1988.
‘]U. S. General Accounting Office, Report to the Chairman, With respect to printing technology, OTA’S
Joint Committee on Printing, GPO Needs to Analyze Alterna-
tives to Overcome Limitations in Government Printing Opera- printing consultant concluded that GPO tech-
tions, PLRD-82-20, Jan. 4, 1982. nology at the main plant was generally on a par
80

with or exceeded the top fifth of the commer- 1972, over 1 million pages in fiscal year 1976,
cial printing industry. GPO was found to be on and over 2.75 million in fiscal year 1980. As
a par with the top 5 percent of private firms of fiscal year 1986, about 3.7 million pages per
with respect to composition technology, the year were being phototypeset. Another exam-
top 11 percent for press technology, and sub- ple is the rapid increase in electronic input to
stantially ahead for bindery technology (a more the GPO printing process over the past sev-
exact estimate here was not possible given the eral years. As of fiscal year 1987, about three
differences between the GPO product mix and quarters of material phototypeset at the GPO
that of typical commercial firms). GPO has main plant was received in electronic form.
stayed abreast of the private sector with re-
spect to conventional technology as a result With respect to conventional press and bind-
of gradual, but continuous equipment up- ery equipment, GPO has nearly completed a
grades. As long as GPO provides a substan- major equipment upgrade stretching over the
tial volume of inplant printing services, peri- past decade. Selected major equipment acqui-
odic equipment upgrades are likely to be sitions are listed in Table 4-9 along with the
cost-effective. acquisition date and cost for each item. Since
1977, GPO has invested almost $15 million in
Perhaps the best example of GPO’s perform- major press equipment, and over $10 million
ance in adopting new technology is the now in bindery equipment. Actual totals are higher
fully completed transition from hot type com- than shown, since a large number of small
position to electronic photocomposition at the equipment items plus furniture, vehicles, and
main plant. This transition took place largely extensive renovations are not listed here.
during the 1970s. In fiscal year 1968, only
40,000 pages were phototypeset. This in- Based on all of the above, OTA has con-
creased to over 700,000 pages in fiscal year cluded that, despite the possibly misleading

Table 4-9.–Selected Major Equipment, Acquisitions, GPO Press and Bindery, Since 1977, as of September 1987

Acquisition Acquisition Acquisition Acquisition


Item year cost in dol tars Item year cost “in dol Iars
Press Division Shredder . . . . . . . . . . . . . . . . . . . 1987 19,272
Letter Press-Envelope . . . . . . . . 1986 48,500 Perforator (2 units). . . . . . . . . . . 1984 15,600
Letter Press-Auto Feed Nipping Machine . . . . . . . . . . . . 1985 13,900
Dryer . . . . . . . . . . . . . . . . . . . . 1987 98,303 Paper Cutter . . . . . . . . . . . . . . . . 1986 47,139
Offset Press-Harris . . . . . . . . . . 1986 40,663 Trim Paper Collection . . . . . . . . 1987 39,495
Offset Press (5 units) . . . . . . . 1979 2,025,000 Waste Paper System . . . . . . . . . 1986 40,582
Offset Press 35x50 (2 units). . 1981 6,264,000 Perforator (2 units). . . . . . . . . . . 1982 17,200
Copier-Xerox 9200 II . . . . . . . . . 1979 62,530 Passport Machine . . . . . . . . . . . 1987 1,213,650
Offset Press-Miehle 43x60 Nipping Machine . . . . . . . . . . . . 1983 12,300
(7 units) . . . . . . . . . . . . . . . . . . 1977 1,918,000 Sewing Machine-Smyth
Offset Press-Harris (2 units). . . 1980 66,000 No. 12. . . . . . . . . . . . . . . . . . . 1980 17,355
Offset Press-Web (3 units) . . . . 1979 2,136,000 Cutter Spacer-Lawson . . . . . . . . 1984 70,000
Offset Press-5 Color Postal Eyelet Attacher Machine . . . . . 1978 5,045
Card . . . . . . . . . . . . . . . . ... 1986 1,104,674 Wrapping Package Machine. . . 1979 37,972
Cut-Pack System-Postal Casemaking Machine-Smyth 1979 25,138
Card . . . . . . . . . . . . . . . . . . . . 1987 970,084 Strapping Machine (4 units) . . . 1982 20,000
Total Press Division . . . . . . . 14,733,754 Strapping Machine-Signode
(3 units) . . . . . . . . . . . . . . . . . . 1984 63,000
Binding Division Folding Endsheet Machine . . . 1983 4)950
Paper Cutter-71 inch . . . . . . . . . 1986 75,237 Casing-In Machine-
Strapping Machine. . . . . . . . . . . 1987 2,623 Versamatic . . . . . . . . . . . . . . . 1983 23,100
Machine Wrap-Stretch Plastic Copier-Xerox 9500 VR . . . . . . . . 1986 12,564
(2 units) . . . . . . . . . . . . . . 1979 23,000 Adhesive Binder (2 units) . . . . . 1983 6,343,347
Inserter-Stitcher Complete . 1987 326,400 Adhesive Mailer (2 units) . . . . . 1983 977,498
Trimmer-3 Knife (2 units) . . . . . 1981 243,000
Folding Machine (6 units). . . . . 1985 420,000 Total Binding Division. . . . . . . . 10,113,678
Labeler Machine. . . . . . . . . . . . . 1986 4,311
SOURCE U S Government Printing Off Ice, 1988
81

external appearance, the GPO main plant is, GPO can remain competitive and maintain or
overall, essentially up-to-date with respect to improve cost-effectiveness in the face of pos-
conventional printing technology and already sible future reductions in the demand for pa-
makes very extensive use of electronic input per formats (and especially demand for major
and photocomposition. There will, of course, products produced at the main plant), increases
be opportunities for future technology up- in demand for electronic formats, and rapid
grades as the need arises. Overall, however, progress in agency automation (including use
GPO is well positioned technologically to carry of desktop and high-end electronic publishing).
on its traditional printing responsibilities. The These possibilities are discussed later in this
strategic challenge arises with respect to how chapter.

TRADITIONAL GPO–LEGISLATIVE BRANCH ONLY


Under this alternative, GPO would continue from about $771 million to about$113 mil-
to provide centralized conventional printing lion or an 85 percent reduction;
services, but only for the legislative branch. the total workload at the GPO main plant
The printing procurement program would ei- would decrease from about $180 million
ther be transferred to the executive branch to $113 million ($90 million main plant
(e.g., to GSA) or dispersed among individual work plus $23 million previously procured)
agencies. Responsibility for the DLP could be or a 37 percent reduction (this assumes
retained at GPO, as could the sale of paper for- the $23 million in procured printing for
mats by the SupDocs, or these functions could the legislative branch would be shifted to
also be transferred to the executive branch. the main plant);
GPO would do little electronic dissemination. the total labor force of GPO would decline
by about 881 persons or about 17 percent
Analysis of this alternative is illustrative of (637 printing procurement staff and 244
one extreme on the spectrum of alternatives regional printing plant staff, all presum-
available and provides further insights into the ably transferred to GSA or elsewhere);
functioning of GPO. The discussion here em- the net income of GPO would decrease by
phasizes financial and labor impacts (see ch. about $4.6 million due to transfer of the
11 and 12 for other implications). printing procurement program which has
To keep this in perspective, it is important operated at a net surplus for the past sev-
to note that GPO was originally established eral years (presumably this net income
in 1860 primarily to serve the printing needs would accrue to GSA, assuming the pro-
of Congress and to eliminate the corruption curement program was kept intact and re-
in printing procurement that had become wide- tained its effectiveness);
spread. Over the following decades, executive the net income of GPO would increase by
branch printing needs grew much faster to the about $1 million due to transfer of the re-
point where, for fiscal year 1987, 85 percent gional printing plants (which have oper-
of GPO work is for the executive branch. ated at a net loss for the last several years),
all other things being equal; and
Financial Impacts the net income less expenses at the GPO
main plant would change from a surplus
Using fiscal year 1987 data, the restriction of several million dollars to a potential loss
of GPO to conventional printing for the legis- of several tens of millions.
lative branch would have the following first
order impacts, all other things being equal: These figures highlight how the GPO main
plant operation is dependent on executive
c the total workload of GPO would decrease branch work to help spread the costs of gear-
82

ing up to meet the quick turnaround and di- Labor Force Impacts
verse needs of the legislative branch (and some
executive branch work, such as the Federal In order to return GPO to break even opera-
Register). The executive branch work helps fill tions, it would be necessary, using this hypo-
in the valleys between the peaks of the con- thetical calculation, to cut costs and/or increase
gressional work load and utilizes labor and revenues by a total of $37 million. Any signif-
plant capacity that would otherwise be underu- icant cost reductions would probably necessi-
tilized. Both in-plant and procured executive tate labor force reductions, since further cuts
branch work help cover GPO overhead ex- in the other, much smaller cost categories
penses and are sources of net income. would have marginal effects at most.
The role of executive branch work can be il- Recovering the entire hypothetical deficit
lustrated using the assumptions about the would necessitate roughly a one-third reduc-
main plant cost structure presented earlier. tion in the main plant labor force, assuming
Starting with $180 million in main plant gross that this could be accomplished without
revenues and assuming a 2 percent profit or jeopardizing the main plant’s capacity to do
surplus, the total GPO main plant expenses the $113 million in legislative branch work. Re
would be $176.4 million. The cost breakout for organization of the production processes might
the main plant would be as in Table 4-10. Now be necessary-for example, scaling back or
if gross revenues drop by 37 percent to $113 eliminating the night shift. Alternatively, some
million due to the exclusion of executive branch or all of the hypothetical deficit could be off-
work, total expenses would decrease by only set through increased appropriations and/or
13 percent to about $154 million, if materials user fees.
and supplies are assumed to be variable costs
but labor, rent, and depreciation are assumed If the hypothetical deficit was to be recov-
to be almost entirely (95 percent) fixed costs ered through labor force reductions, a total re-
in the short run. The result is a swing from duction in force of about 1,100 employees
a net surplus of about $4 million to a net loss would be needed, assuming a total main plant
on main plant operations of about $4 I million. work force of 3,311, calculated as shown in Ta-
If printing procurement were transferred out, ble 4-11. The main plant labor force of 3,311
there would be no net surplus from procure- persons is estimated by deducting the Sup-
ment to even partially offset this loss. Docs staff (which operates on a breakeven ba-

Table 4.10.— Hypothetical Calculation of Financial Impact of Legislative Branch GPO

Main plant Main plant


Executive and legislative work Legislative work only
Gross ;evenue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $180.0 million $113.0 million
Less net surplus (assumed 2°\o). . . . . . . . . . . . . . . . . ... . –3.6
Total expenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176.4 million
Assumed cost structure Assumed cost
(as percent of total expense) reduction
Labor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67°/0 $118.2 million –50/0 $112.3 million
Materials and supplies . . . . . . . . . . . . . . . . . . . . . . . 25°\0 44,1 – 370/0 27.8
Rent, communications, and utilities. . . . . . . . . . . . . 4,40/0 7.8 –50/0 7.4
Depreciation and other . . . . . . . . . . . . . . . . . . . . . . ; . 3.60/o 6,3 – 50/0 6.0
$176.4 million $153.5 million
Net income or (loss) . . . . . . . . . . . . . . . . . . . . $ 3.6 million ($ 40.5 million)
SOURCE Office of Technology Assessment, 1988
83

Table 4-11 .—Hypothetical Main Plant Table 4-12.— Hypothetical Main Plant Divisional
Total Labor Force Reductions Labor Force Reduction

Fiscal year 1987 total GPO labor force . . . . . . . . . 5,122 Persons


Less SUP D OCS staff . . . . . . . . . . . . . . . . . . . . . . . – 930
Electronic Photocomposition Division . . ~. 616
4,192 Press Division . . . . . . . . . . . . . . . . . . . . . . . . . . . 701
Less regional printing staff –244 Binding Division . . . . . . . . . . . . . . . . . . . . . . . . . . . . +630
3,948 Current FY87 labor force . . . . . . . . . . . . . . . . . . . 1,947
Less printing procurement staff. . . . . . . . . . . . . –637 x .78
Current main plant labor force . . . . . . . . . . . . 3,311 Reduced labor force . . . . . . . . . . . . . . . 1,519
Less 1/3 reduction in force . . . . . . . . . . . . . . . . . 1,100 Staff reductions. . . . . . . . . . . . . . . . . . . . . 428
Reduced main plant labor force . . . . . . . . . . 2,211 Executive Office . . . . . . . . . . . . . . . . . . . . 693
SOURCE Off Ice of Technology Assessment 1988 Operations Division . . . . . . . . . . . . . . . . . . . . . . . . . 359
Production Division . . . . . . . . . . . . . . . . . . . . . . . . 313
sis, financially separate from the main plant) Current FY87 labor force . . . . . . . . . . . . . . . . . . . 1,364
x .50
and the regional printing and printing procure
Reduced labor force . . . . . . . . . . . . . . . . . . . . . . . 682
ment staffs (which would, in this hypotheti- Staff reductions. . . . . . . . . . . . . . . . . . . . . . . . . . . 682
cal example, be transferred to GSA). The hypo- Total staff reductions. . . . . . . . . . . . . . . . . . . . . ., . 1,110
thetical one-third labor force reduction is Total remaining labor force . . . . . . . . . . . . . . . . . . . 2,201
calculated by dividing the net loss ($37.3 mil- SOURCE Off Ice of Technology Assessment 1988 -
lion) into the main plant labor costs ($110.0
million). percentages. For example, of the 547 com-
GPO has previously estimated that about posers (International Typographical Union,
78 percent of the employees in the main plant Local No. 101), 176 (or 32 percent) are eligible
Photocomposition, Press, and Binding Divi- for retirement and 326 (or 60 percent) have 21
sions would be required to perform the legis- or more years of service, as shown in detail in
lative branch work (GPO defines this as con- Table 4-13.
gressional work plus the Federal Register,
postal cards, passports, CFR, and OMB and If three years were available to make the
Presidential documents). This means that the transition to a legislative branch GPO, the 22
fiscal year 1987 staffing level of 1,947 persons percent reduction in the photocomposition,
for these divisions could be reduced by only press, and binding labor force probably could
428 persons (22 percent) in order to maintain be made mostly through natural attrition
the necessary capacity. The remaining reduc- (averaging 5-10 percent per year at the main
tion of 672 persons (to provide a total of 1,100) plant). However, the 50 percent reduction in
would have to come from the Executive Of- the executive office, operations, and produc-
fice, Operations, and Production Divisions. tion labor force probably could not be made
These divisions had a fiscal year 1987 com- over this period of time through natural attri-
bined staffing level of 1,364 persons, which tion, and some early retirement buyouts might
would translate into a roughly 50 percent staff be necessary. Of course, the hypothetical tran-
cut (672 out of 1,364) for these areas. The cal- sition would be easier if more time were
culations are shown in Table 4-12. available.

Options available to GPO for handling these Other Vulnerabilities


hypothetical reductions would depend on how
fast they had to be made. Overall, GPO has As a final note, the “traditional GPO-legis-
a relatively old labor force with about 13 per- lative branch only” alternative would be espe-
cent of its employees eligible for retirement, cially vulnerable to any significant future re-
and about 35 percent having 21 or more years ductions in the demand for paper formats.
of service. Some craft units have even higher Prior GPO analyses have, indeed, documented
84

Table 4-13.—GPO Main Plant Composers, Table 4-14.—Changes in GPO Main Plant Volume
Years of Service and Retirement Eligibility, for Eight Principal Products a, 1975”1984
Fiscal Year 1987
1975 1984 Percent
Number of number number change
employees Titles . . . . . . . . . . . . . . . . . . 28,893 13,854 –55
Years of service Original pages (inmiliions) . 1.048 0.886 –16
o-5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Copies (in millions) . . . . . . . . . 134.6 48.7 –64
6-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Pages per copy (average) . . . . 36 64 +78
11-15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Printed pages (in trillions) . . . . 4.85 3.12 –36
16-20 . . . . . . . . . . . . . . . . . . . . . . . . . . 161 aDailY cO~gfeSS/Ona/ R e c o r d , F e d e r a l Ffeg/ster, bills, reSOILItl OnS, a n d
193 amendments, committee heari rigs, committee reports, committee pr! nts,
21-25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . calendars, and Code of Federa/ Regulatw?s
26-30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
SOURCE U S Government Prlntlng Office and Off Ice of Technology Assessment,
31-35 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 1988
36-40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
41+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547
Retirement eligibility
Age 55/30 years service . . . . . . . . . . . . . . . . . . 32 were not equal. During this period, the GPO
Age 60/20 years service . . . . . . . . . . . . . . . . . . 85 labor force decreased by about 34 percent, and
Age 62/5 years service . . . . . . . . . . . . . . . . . . . 59
176
more productive, less labor-intensive equip-
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SOURCE US Government Pnntlng Office, 1988 ment was deployed. Some of the difference was
also made up in price increases. However, as
the trend data indicate (Table 4-15), billings
for key congressional printing and binding
significant reductions over the 1975-1984
items remained remarkably stable, increasing
period.
by only 1.4 percent through fiscal year 1983
Thecombinedtotalsforeightprincipalmain and by about 17 percent through fiscal year
plant products (daily Congressional Record; 1984.
Federal Register; bills, resolutions, and amend-
Trend data for the entire fiscal year 1975-
ments; committee hearings; committee re-
1987 period show only minor changes in GPO
ports; committee prints; calendars; and the
billings for hearings, committee prints and
Code of Federal Regulations) showed declines reports, and calendars. As indicated in Table
of 64 percent in total number of copies printed. 4-16, billings for bills, resolutions, and amend-
However, this reflected primarily a 55 percent
ments were up significantly, although this may
reduction in the number of titles, which is a
reflect a fiscal year 1987 anomaly since fiscal
function of a lower overall level of congres-
year 1986 billings were $8.41 million, up only
sional activity rather than an indication of
marginally from the $7.97 million expended in
lower demand. The number of pages declined
fiscal year 1975. The only dramatic changes
by only16percent,whichmeans that theaver-
were for bills, resolutions, and amendments (up
age number of pages per title must have in-
35 percent), the Congressional Record (up 71
creased significantly over this period of time
percent) and the Federal Register (up 128 per-
(for example, fewer but longer reports and
cent), as shown below. These latter two items
bills). Indeed, as shown, the average number
are: among the biggest work orders at the main
of pages per copy almost doubled, from 36 to
plant, very labor intensive, the primary rea-
64 pages. Nonetheless, over the 1975-1984
son (along with congressional bills and reports)
period, the total number of pages printed at
for overnight operations at the main plant, and
the GPO main plant for these eight products
among the more vulnerable main plant prod-
declined by about 36 percent. The statistical
ucts with respect to competition from elec-
results are shown in Table 4-14.
tronic formats. As mentioned elsewhere in this
This volume reduction would be expected to chapter, the Record and Register are both
increase drastically per unit costs, all other highly suited to online and CD-ROM electronic
things being equal. However, all other things formats.
85

Table 4-15.—Congressional Printing and Binding Billings Selected Items,


Fiscal Years 1975, 1983, and 1984

Fiscal year Fiscal year Fiscal year “


1975 1983 1984
Item (in thousands of dollars)
Hearings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $17,746 $16,684 $22,304
Miscellaneous printing and binding . . . . . . . . . 9,776 8,720 10,042
Bills, resolutions, and amendments . . . . . . . . . 7,965 7,552 6,827
Miscellaneous publications . . . . . . . . . . . . . . . . 3,680 4,130 4,585
Committee prints. . . . . . . . . . . . . . . . . . . . . . 4,372 2,956 3,065
House and Senate calendars. . . . . . . . . . . . . . . 1,720 1,256 2,138
Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466 1,571 958
Committee reports . . . . . . . . . . . . . . . . . . . . . . . 2,644 2,827 3,048
Franked envelopes . . . . . . . . . . . . . . . . . . . . . . . 815 759 1,111
Congressional Record (daily) . . . . . . . . . . . . . . . 8,287 11,794 13,352
Totals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $57,471 $58,249 $67,430
SOURCE U S Government Printing Office and Offtce of Technology Assessment, 1988

Table 4.16.—Congressional Printing and Binding Billings, Selected Items,


Fiscal Years 1975 and 1987

Fiscal year Fiscal year Percent


1975 1987 change
Item (in thousands of dollars)
Hearings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $17,746 $16,835 –5.1 “/0
Bills, resolutions, and amendments . . . . . . . . . 7,965 10,830 + 36.0
Committee prints and reports . . . . . . . . . . . . . . 7,016 7,247 +3.3
House and Senate calendars. . . . . . . . . . . . . . . 1,720 1,543 – 10.0
Congressional/ Record (daily) . . . . . . . . . . . . . . . 8,287 11,173 + 35.0
a
Federal Register (daily). . . . . . . . . . . . . . . . . . . . 7,776 17,697 + 128.0
aln~ludes billings for Congressional copies only and thus understates total billin9s
SOURCE U S Government Prlntlng Off Ice and Off Ice of Technology Assessment, 1988

ELECTRONIC GPO–DECENTRALIZED
Under this alternative, the GPO would con- thermore, electronic formats selected for inclu-
tinue to provide centralized conventional print- sion in the DLP would be distributed to the
ing services (that is, Federal Government ink- depository libraries either directly by the agen-
on-paper printing would be obtained from or cies or via GPO. Also, this alternative assumes
through GPO), expand the range of electronic that GPO would develop and maintain a
publishing services available to agencies, dis- governmentwide information index in cooper-
seminate selected electronic formats on a sales ation with NTIS and would actively partici-
basis through SupDocs as well as traditional pate in governmenttide standards-setting and
paper formats, and disseminate selected elec- innovation activities concerning electronic
tronic as well as paper and microfiche formats printing, publishing, and information dissem-
to the DPL. However, government dissemina- ination.
tion of electronic formats would not be central-
ized solely via GPO. Mission agencies could, This alternative, labelled for convenience
at their discretion, disseminate their own elec- “Electronic GPO-Decentralized,” most closely
tronic formats, or they could opt to utilize Sup- aligns with the current development path of
Docs, or both. SupDocs could, at its discretion, GPO. GPO is conducting a number of relevant
select those electronic formats judged to be pilot projects, and is experimenting with and
suitable for inclusion in the sales program. Fur- occasionally implementing precursor electronic

86

applications. This alternative is conceptually the original material is being provided to GPO
viable, and, therefore, warrants careful consid- in a variety of formats, primarily electronic,
eration, due to the convergence of several key as shown in Table 4-17.
trends in electronic technology and demand for
Federal information in electronic formats. The overall picture that emerges is as fol-
lows. Almost all executive agency material is
being provided to GPO in camera-ready or elec-
Trends in Technology and Demand
tronic formats, with very little material requir-
Technology Trends ing GPO keyboarding. Almost all agency elec-
tronic input is via magnetic tape. On the other
One key technology trend is the rapid in- hand, roughly one half of all legislative branch
crease in agency automation, which means that material requires GPO keyboarding, roughly
most agencies already are creating their origi- 10 percent is scanned, and the remaining 40
nal information products in electronic form, percent of electronic input is split between
and many are also converting this material to magnetic tape and fiber optic cable transmis-
a camera-ready format. OTA’S independent sion. The distribution of origination formats
printing consultant estimated that about 25 is shown in Table 4-18 for camera-ready, man-
percent of the original material is being pro- uscript, scanned, and electronic input as a per-
vided by Federal agencies to GPO in camera- centage of total input and total by branch of
ready format. For these pages, no typesetting government.
or page composition by GPO is required.
Almost all (98 percent) of this camera-ready The executive branch agencies are able to
material is estimated to originate from execu- capture their own electronic keystrokes and,
tive branch agencies. The other 75 percent of increasingly, do their own electronic composi-

Table 4-17.—Origination Formatsa for Material Submitted to GPO,


as Percent of Total

Executive Legislative
Format branch branch Totals
Manuscript Copyb . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 0/0 23.4 0/0 260/o
Scanned entry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.5 4.5 5
Magnetic tape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.6 10.4 52
Floppy disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.96 0.04 2
Fiber optic cable. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 10.0 10
Other electronic transmission . . . . . . . . . . . . . . . . . . . . . . . . . . 4.25 0.75 5
100’Y”
aexcluding camera-ready cOPY.
brequlring keyboarding.
SOURCE U S Government Printing Office, F R. Romano, and Office of Technology Assessment, 1988

Table 4-18.—Origination Formats, Including Camera. Ready, as Percent of Total


and by Branch of Government

Percent of all Percent of all


Government totals branch totals
Executive Legislative Executive Legislative
Format branch branch Totals branch branch
Camera-ready. . . . . . . . . . . . . . . 24.5 0.5 25 39.1 1.3
Manuscript . . . . . . . . . . . . . . . . . 1.95 17.55 19.5 3.1 47.0
Scanned entry . . . . . . . . . . . . . . 0.375 3.375 3.75 0.6 9.0
Electronic input . . . . . . . . . . . . 35.86 15.89 51.75 57.2 42.6
Totals . . . . . . . . . . . . . . . . . . . 100 100 99.9
SOURCE: U S. Government Printing Office, F R. Romano, and Office of Technology Assessment, 1988
87

tion, largely because of the widespread penetra- Table 4.20.—Estimated Use of Electronic
tion of computer and word processing technol- Publishing Software, Calendar Year 1987
ogies and, recently, the rapid increase in the Total Estimated
use of desktop and high-end electronic publish- units in units in
ing. As noted in chapter 2, the GAO survey Software/Vendor United States U.S. Government
of 114 civilian departmental agency compo- Page makeup software
Aldus Pagemaker. . . . . . . . 115,000 6,000
nents indicated that many are using and/or Xerox Ventura . . . . . . . . . . 85,000 12,000
testing relevant technologies, as summarized All others ... , ... , . . . . . . 66,000 3,000
in Table 4-19. Totals. . . . . . . . . . . . . . . . 226,000 21,000
Page description software
The GAO survey did not ask for estimates Hewlett. Packard PCL . . . . 210,000 29,000
of the absolute number of these technologies Postscript . . . . . . . . . . . . . . 420,000 14,000
in use. However, these estimates can be devel- Proprietary for printer . . . . 790,000 67,000
Typesetter. . . . . . . . . . . . . . 125,000 8,700
oped from other relevant indicators, including Other laser printer. . . . . . . 45,000 11,000
the use of page makeup and page description Totals. . . . . . . . . . . . . . . . 1,590,000 129,700
software. OTA’S independent printing consul- SOURCE: TypeWorld, F J. Romano, 19S8
tant has estimated that, as of year end 1987,
there were already over 20,000 units of page
Table 4-21 .—Estimated Use of Other Electronic
makeup software in use in the Federal Gov- Publishing Technologies, Estimated,
ernment, and over 125,000 units of page de- Calendar Year 1987
scription software. The detailed breakout is in
Table 4-20. Estimated Estimated
units in cost
OTA’S printing consultant estimates that, Technology U.S. Government per unit
in addition, there are 200-350 high-end elec- High-end electronic
publishing software . . . loos $40K
tronic printers (Xerox 9700 class) in use in the (e.g., Interleaf)
Federal Government. Just these three items High-end workstation . . . 1 ,000s $20K
alone account for over $200 million in installed (e.g., Sun)
Low-end laser printers . . 10,000s $ 3K
base of electronic publishing technology (21K (e.g., HP Laserjet)
units page makeup at $600/unit + 127K units SOURCE Off Ice of Technology Assessment, 1988
page description at $750/unit + 275 high-end
electronic printers at 400 K/unit = $12.6M +
equipment (assuming 300 units high-end soft-
$95.3M + $11OM = $217.9 M). This does not
ware at $40 Khmit + 3000 units highend work-
include high-end workstations and low-end la-
stations at $20 K/unit + 30,000 units low-end
ser printers, among other relevant technol- laser printers at $3 K/unit = $12M + $60M
ogies. Rough estimates for the latter are shown + $90M = $162 M). These estimates suggest
in Table 4-21.
that the Federal Government, and primarily
These technologies represent, conserva- the executive agencies, have already invested
tively, roughly another $160 million in installed about $350-$400 million in electronic publish-

Table 4“19.—Civilian Department Agency Use of selected Electronic Technologies


(percent of agencies responding)

Currently in Currently prototyping


Technology operational use or pilot testing Totals
Computer-aided page makeup . . . . . . . . 50.0 8.8 58.8
Computer graphics . . . . . . . . . . . . . . . . . . 65.8 7.9 73.7
Electronic photocomposition . . . . . . . . . 43.9 7,9 51.8
Laser and other non-impact printing . . . 64.0 1.8 65.8
Desktop publishing system. . . . . . . . . . . 34.2 14.9 49.1
Electronic publishing system . . . . . . . . . 21.1 10.5 31.6
SOURCE U S Government Printing Office, F R Romano, and Office of Technology Assessment, 19S8
88

ing and related technologies, with no end to Federal libraries, indicated a strong preference
agency procurements in sight. Also, these to obtain increasing percentages of Federal in-
figures do not include any allocation of the over formation in electronic form and declining per-
25,000 mainframe computers and the esti- centages in paper and microfiche. The results
mated 500,000 microcomputers in the Federal for 318 depository libraries responding out of
Government that are used in part for electronic a sample of 450 (34 out of 50 regional deposi-
publishing applications. Finally, these figures tory libraries and 284 out of 400 selective de-
do not reflect the rapidly growing agency pi- pository libraries) are highlighted in Table 4-
lot tests and operational applications for di- 22. Only the most significant changes are in-
rect electronic dissemination via bulletin cluded here. These results show that the library
boards, electronic mail, CD-ROM, and the like. community desires or anticipates decreases in
use of paper and microfiche formats, signifi-
Trends in Demand cant increases in electronic mail or bulletin
This high level of agency activity reflects, boards and floppy disks, and substantial in-
in part, opportunities presented by advancing creases in online databases and compact opti-
technology and the overall drive to automate cal disks.
agency functions. However, agency activities Trends for other segments of the Federal in-
also reflect growing demand on the part of Fed- formation user community are not so dramatic
eral information users to receive information but show a similar pattern. The results for 109
in electronic formats. scientific and technical associations respond-
ing to the GAO survey (out of a sample of 250)
The results of the GAO survey of Federal
are highlighted in Table 4-23.
information users document this demand. For
example, the depository library community, The GAO survey of Federal agencies indi-
which heavily reflects university, research, and cates that agency use of electronic dissemina-

Table 4“22.—Depository Library Demand for Federal Information, by Type and Format

Number of libraries using


Next Percent
Type of information Format Now 3 years change
Congressional record/hearing/ paper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 234 – 13.7
reports/bills microfiche . . . . . . . . . . . . . . . . . . . . . . . . . 274 225 – 17.9
online database. . . . . . . . . . . . . . . . . . . . 59 132 + 124.0
floppy disk . . . . . . . . . . . . . . . . . . . . . . . . . 0 27 large increase
compact optical disk . . . . . . . . . . . . . . . . 3 112 + 3,633.0
Scientific and technical paper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 172 – 17.2
reports/information microfiche . . . . . . . . . . . . . . . . . . . . . . . . . 212 159 –21 .7
online database. . . . . . . . . . . . . . . . . . . . . 76 95 +25.0
floppy disk . . . . . . . . . . . . . . . . . . . . . . . . . 1 27 + 2,600.0
compact optical disk . . . . . . . . . . . . . . . . 9 78 + 767.0
Press releases/bulletins paper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 183 –25.6
microfiche . . . . . . . . . . . . . . . . . . . . . . . . . 39 35 –10.3
electronic mail or bulletin board . . . . . . 9 51 +467.0
online database. . . . . . . . . . . . . . . . . . . . . 24 50 + 108.0
compact optical disk . . . . . . . . . . . . . . . . 1 18 + 1,700.0
Statistical data paper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 270 – 12.6
microfiche . . . . . . . . . . . . . . . . . . . . . . . . . 241 134 –44.4
electronic mail or bulletin board . . . . . . 12 27 + 125.0
online database. . . . . . . . . . . . . . . . . . . . . 103 158 + 53.4
magnetic tape/disk . . . . . . . . . . . . . . . . . . 11 25 + 127.0
floppy disk . . . . . . . . . . . . . . . . . . . . . . . . . 12 65 +442.0
videodisk . . . . . . . . . . . . . . . . . . . . . . . . . . 0 12 large increase
compact optical disk . . . . . . . . . . . . . . . . 15 140 +833.0 —
SOURCE: GAO Survey of Federal Information Users, 1988,
89

Table 4-23.—Scientific and Technical Association Demand for Federal Information, by Type and Format

Number of associations using


Next Percent
Type of information Format Now 3 years change
Congressional record/hearings/ paper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 57 –6.6
reports/bills electronic mail or bulletin board . . . . 1 18 + 1,700.0
online database. . . . . . . . . . . . . . . . . . . . . 4 22 + 450.0
floppy disk . . . . . . . . . . . . . . . . . . . . . . . . . 0 13 large increase
Scientific and technical paper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 75 – 15.7
i nformat ion microfiche . . . . . . . . . . . . . . . . . . . . . . . . 8 5 – 37.5
electronic mail or bulletin board . . . . . . 6 24 + 300.0
online database. . . . . . . . . . . . . . . . . . . . 9 30 + 233.0
floppy disk . . . . . . . . . . . . . . . . . . . . . . . . . 6 20 + 233.0
compact optical disk . . . . . . . . . . . . . . . . 1 3 + 200.0
Press releases/bulletins paper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 60 – 22.1
electronic mail or bulletin board . . . . . . 3 26 + 767.0
online database. . . . . . . . . . . . . . . . . . . . . 5 16 + 220.0
floppy disk . . . . . . . . . . . . . . . . . . . . . . . . . 1 10 + 900.0
Statistical data paper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 49 –18,3
electronic mail or bulletin board . . . . . . 1 11 +1,000.0
online database. . . . . . . . . . . . . . . . . . . . . 6 23 +283.0
floppy disk . . . . . . . . . . . . . . . . . . . . . . . . . 5 23 +360.0
compact optical disk . . . . . . . . . . . . . . . . 1 5 +400.0
SOURCE GAO Surveyof Federal Information Users 1988

tionis already significant forsometypesof ber of challenges and opportunities. Principal


information and is projected to expand con- among these are: electronic input, structured
siderably over the next 3years. For the 114 database standards, electronic publishing sup-
civilian departmental agency components re- port, dissemination of electronic formats, staff-
spending, some key trends are in dicated inTa- ing, and capital investment.
ble 4-24.
While generalizations are difficult, the sur- Electronic Input
vey data suggest that, for several major types
of Federal information, especially statistical As noted earlier, most Federal agencies al-
data, scientific and technical reports/informa- ready have the technology needed to originate
tion, administrative reports, and press re- materials in electronic form and capture the
leases, about one-fifth to one-third of the key strokes needed to initially enter the data
executive branch agencies expect to have elec- and make subsequent revisions. Once the ma-
tronic dissemination available within 3 years. terial is ready for composition and layout, and
The dominant electronic formats vary by type assuming the originating agency is not per-
of information. Probably one-tenth to one- forming these functions, it is generally more
eighth of the agencies expect to have electronic cost-effective to transmit the data in electronic
dissemination of other types of Federal infor- form to GPO so as to minimize any necessary
mation (e.g., pamphlets, manuals, regulations, rekeyboarding. The cost savings can be sub-
directories). Overall, however, the survey data stantial. GPO estimates that rekeyboarding
suggest that despite dramatic increases in elec- costs on average $35 to $50 per page, whereas
tronic formats, paper will still be the dominant electronic input costs $1 to $15 per page, de-
format for the near- to medium-term. pending on how much recoding and reformat-
ting are needed. For electronic input materials
Opportunities and Challenges using the GPO structured database standards,
the average cost is $1 to $2 per page, since no
These trends in technology, user demand, rekey boarding and minimal recoding or refor-
and agency activities present GPO with a num- matting are needed.
90

Table 4-24.—Agency Activities and Plans for Electronic Information Dissemination, by Type and Format

Percent of agencies using


Next Percent
Type of information Format Now 3 yearsa change
Administrative reports Electronic mail . . . . . . . . . . . . . . . . . . . 14.0 ”/0 ~b.~ “/0 +87.90/o
Electronic bulletin board . . . . . . . . . . 0.9 8.8 + 878.0
Electronic data transfer . . . . . . . . . . . 12.3 21.1 +71 .5
Magnetic tape/disk . . . . . . . . . . . . . . . 6.1 11.4 +82.0
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 7,9 16.7 + 111.4
Compact optical disk . . . . . . . . . . . . . — 2.6 +
Scientific and technical reports/ Electronic mail . . . . . . . . . . . . . . . . . . . 6.1 15.8 + 159.0
information Electronic bulletin board . . . . . . . . . . 6.1 10.5 +72.1
Electronic data transfer . . . . . . . . . . . 14.9 18.4 +23.5
Magnetic tape/disk . . . . . . . . . . . . . . . 14.0 16.7 + 19.3
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 8.8 16.7 + 90.0
Compact optical disk . . . . . . . . . . . . . — 8.8 +
Press releases Electronic mail . . . . . . . . . . . . . . . . . . . 13.2 28.1 + 112.9
Electronic bulletin board . . . . . . . . . . 5.3 12.3 + 132.1
Electronic data transfer . . . . . . . . . . . 7.0 13.2 + 88.5
FIoppy disk. . . . . . . . . . . . . . . . . . . . . . 3.5 7.0 + 100.0
Video tape . . . . . . . . . . . . . . . . . . . . . . 6.1 8.8 +44.3
Pamphlets/bulletins Electronic mail . . . . . . . . . . . . . . . . . . . 8.8 18.4 + 109.0
Electronic bulletin board . . . . . . . . . . 3.5 10.5 + 200.0
Electronic data transfer . . . . . . . . . . . 4.4 13.2 +200.0
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 1.8 9.6 +433.3
Manuals Electronic mail . . . . . . . . . . . . . . . . . . . 0.9 9.6 + 966.6
Electronic bulletin board . . . . . . . . . . 0.9 5.3 +488.9
Electronic data transfer . . . . . . . . . . . 3.5 11.4 +225.7
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 5.3 14.0 + 164.2
Compact optical disk . . . . . . . . . . . . . — 3.5 +
Decisions/opinions Electronic mail . . . . . . . . . . . . . . . . . . . 2.6 10.5 +303.9
Electronic bulletin board . . . . . . . . . . — 5.3 +
Electronic data transfer . . . . . . . . . . . 2.6 9.6 +269.2
Magnetic tape/disk . . . . . . . . . . . . . . . 0.9 4.4 + 388.9
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 2.6 6.1 + 134.6
Rules, regulations, directives, Electronic mail . . . . . . . . . . . . . . . . . . . 8.8 18.4 + 109,1
circulars Electronic bulletin board . . . . . . . . . . 0.9 6.1 +577.8
Electronic data transfer . . . . . . . . . . . 2.6 10.5 + 303.9
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 6.1 12.3 + 101.6
Directories/c atalogs/bibl iographies Electronic mail . . . . . . . . . . . . . . . . . . . 0.9 11.4 + 1,167.0
Electronic bulletin board . . . . . . . . . . 1.8 7.9 + 339.0
Electronic data transfer . . . . . . . . . . . 8.8 18.4 + 109.1
Magnetic tape/disk . . . . . . . . . . . . . . . 8.8 11.4 +29.5
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 5.3 14.9 + 181.1
Videodisk . . . . . . . . . . . . . . . . . . . . . . . 0.9 3.5 +288.9
Compact optical disk . . . . . . . . . . . . . — 7.9 +
Statistical data Electronic mail . . . . . . . . . . . . . . . . . . . 7.9 20.2 + 155.7
Electronic bulletin board . . . . . . . . . . 3.5 12.3 +251 ,4
Electronic data transfer . . . . . . . . . . . 17.5 25.6 +46.3
Magnetic tape/disk . . . . . . . . . . . . . . . 31.6 34.2 +8.2
Floppy disk. . . . . . . . . . . . . . . . . . . . . . 19.3 31.6 +63.7
Com~act orMical disk . . . . . . . . . . . . . . 2.6 +
%alculated by adding the percentage of agencies now using the format indicated to the number of agencies who expect to use the format within the next 3 years.
Assumes that agencies currently using a format will continue to do so.
SOURCE: GAO Survey of Federal agencies, 1987.

GPO already provides multiple options for GPO has scanners that can read most of the
electronic input, as noted earlier, including popular typewriter and word processor fonts
scanned input, floppy disks, magnetic tapes, and convert the material from alphanumeric
dial-up telephone lines, and fiber optic lines. characters to electronic form. However, the
Some of these options could be refined and/or scanned copy must be very clear and legible
expanded. in order to obtain a low error rate and, in any
91

event, must be coded as well as scanned in or- tive branch agencies. Magnetic tape represents
der to produce a database for input to the GPO a high speed, high volume, low cost per page way
composition system. As a result, the cost of to transmit material from originating agencies
scanned input is higher than other purely elec- to GPO for composition, typesetting, and print-
tronic input modes but, when properly sel- ing. Magnetic tapes can be provided to GPO
ected, can be much lower than rekeyboarding. in any of three formats: database tapes, direct
drive tapes, and data tapes.
Scannable material is still a small percent-
age (about one-fifth) of all material in manu- Database tapes are produced by the originat-
script form received by GPO. As long as man- ing agency (or an agency contractor) using
uscript copy is submitted, the more that is GPO’s structured database specifications.
scannable, the lower the costs. GPO could in- These tapes require no code conversion and
tensify efforts to advise agency customers of serve as input to the GPO composition sys-
the scanner option and the typeface and for- tem. GPO has been processing database tapes
mat requirements. GPO could also aggressively for nearly twenty years. The preparation of
evaluate state-of-the-art scanner technology in camera ready copy at GPO from database
order to increase the range of typefaces and fonts tapes is inexpensive, priced at $1.85 per page.
that can be scanned, and to simplify recoding The preparation of film negatives from data
to the extent feasible. Also, agencies which have tapes costs $3.15 per page. Direct drive tapes
their own scanners could be encouraged to do are produced by the originating agency (or an
the scanning themselves and submit materials agency contractor) using not only GPO’s struc-
to GPO in floppy disk or other direct electronic tured database specifications, but also GPO’S
format. However, in the final analysis, scan- electronic composition codes (with type face
ners are much slower and more error-prone and page format already specified). Direct
than direct electronic formats. drive tapes provide direct input to GPO’s pho-
totypesetters, producing camera ready copy
GPO has a floppy disk reader that is capa- or film negatives. The preparation of camera
ble of reading over 100 different disk formats ready copy or film negatives from direct drive
from a wide range of word processing systems. tapes is $0.30 cheaper per page than from data-
Agencies can submit their material on floppy base tapes, at $1.55 and $2.85 per page respec-
disks, and the word processing codes for type- tively, and is a low-cost way of providing ma-
face and format used on these disks can be con- terials to GPO. The use of both database and
verted to GPO’s structured database codes. direct drive tapes has increased in recent years,
Very few agencies currently make use of this as indicated in Table 4-25.
option. GPO could encourage greater agency use
of floppy disk input, especially as a substitute These two forms of magnetic tape input are
for manuscript submission, which requires com- likely to continue at or above present levels
plete rekeyboarding. GPO could also survey the for the forseeable future, so long as traditional
Federal agencies to ascertain the types and dis- ink-on-paper output formats are needed and
tribution of word processing systems being GPO traditional printing services remain com-
used, and could add capability to existing GPO petitive. Some agencies have the capability to
equipment to convert other kinds of disk for- produce magnetic tapes, but do not have the
mats used by agencies. Floppy disk conversion
does require quality control on the part of the
Table 4-25.—GPO Pages Produced from Database and
agencies to insure consistently error-free cod-
Direct Drive Magnetic Tapes, Fiscal Years 1983-87
ing. Floppy disk input is generally best suited
for shorter documents, cheaper than scanned Database tapes Direct drive tapes
input, but more expensive than magnetic tape Fiscal year total pages total pages
input. 1983 . . . . . . . . . . . . . . . . 392,162 350,723
1984 . . . . . . . . . . . . . . . . 654,606 859,497
Magnetic tape input is the dominant mode 1985 . . . . . . . . . . . . . . . . 769,791 781,398
1986 . . . . . . . . . . . . . . . . 926,445 724,889
used by executive agencies, and is used to a 1987 . . . . . . . . . . . . . . . . 807,507 838,545
lesser, but still significant, extent by legisla- SOURCE U S Government Pr!ntlng Off Ice, 1988
92

expertise or desire to code the tapes to GPO Table 4-26.—GPO Dial-Up Electronic Transmission
specifications. In these cases, GPO does not Customers, January ,388
have to rekeyboard the substantive material, Executive branch
but does have to convert from agency codes Department of Commerce
to GPO’s structured database standards and Department of Housing and Urban Development
Department of Transportation, Federal Aviation
insert composition codes. These tapes are han- Administration
dled on a case-by-case basis similar to the Department of the Treasury, Customs Bureau
floppy disks. Veterans Administration
Office of Management and Budget
A final means of input to be discussed here Office of the Federal Register, Code of Federal
Regulations
is electronic transmission via dial-up telephone
Legislative branch
lines, fiber optic lines, satellite private lines, GPO—Daily Congressional Record, Record Index, Bills
and the like. Conceptually, the originating Library of Congress
agency keyboards the data on its own micro- House of Representatives
computer or other terminal, electronically Committee on Banking, Hearing
Committee on the Budget, Hearing
transmits the data via a telecommunications Committee on Energy and Commerce, Hearing,
link to GPO for composition, and electronically Committee Print
receives the proof pages back from GPO via Committee on Foreign Affairs, Committee Prints
Committee on the Judiciary, Hearings
the telecommunications link for printout on Committee on Small Business, Hearings
a laser printer. Corrections can be made on the Committee on Veterans Affairs, Committee Print,
proof copy and electronically transmitted (or Hearings, Report, Title 38 (U.S. Code)
Sergeant At Arms, Notice
mailed) back to GPO, where the final pages are Office of Legislative Counsel
produced on GPO’s phototypesetters. As of Office of the Clerk, Calendars, Lists, Stationary,
January 15, 1988, the organizations listed in Directory
Table 4-26 were using dial-up transmission for U.S. Senate
Committee on Veterans Affairs, Hearing
input and proofing of various publications. Office of the Secretary, Document, Book, Senate
Journal
While electronic transmission represents, at Office of Legislative Counsel
present, a small portion of total input to GPO, SOURCE: U.S. Government Printing Office, 1988.
this use for electronic input is likely to grow
significantly, especially if proven to be cost ef-
fective. The experience with the fiber optic (6,158 pages at a total cost of $45,592). The
transmission between the Senate Office of U.S. Code is more complex and contains more
Legislative Counsel and GPO is illustrative. characters per page than a bill, and bills usu-
The Legislative Counsel uses a fiber optic link ally go through several revisions. If one as-
to transmit draft bills to GPO, where they are sumes three revision cycles for bills, with 4 bill
typeset and transmitted back to the Legisla- pages equivalent to one U.S. Code page, the
tive Counsel’s Office and proof copies are costs are similar.
pMted out on laser printers. During fiscal year
1987, 163,893 pages were transmitted in this Structured Database Standard
fashion at a total GPO billable cost of $75,350, As mentioned in chapter 3, a structured data-
or about $0.46 per page. This appears to be base standard is a key aspect of overall stand-
a competitive price, although it presumably ards development for electronic publishing and
does not reflect any capital costs (such as the dissemination. GPO has developed and imple-
fiber optic link or laser printers) and does not mented what it calls a logically structured full-
include the GPO cost of printing copies of the text database standard or specification, or sim-
bills in final form. Also, bills are very straight ply a structured database standard for short.
forward in format. While not strictly compara- All this really means is that the database (con-
ble, the average GPO per page composition taining textual, tabular, and/or numerical in-
cost for the U.S. Code using magnetic tape in- formation) contains no coding unique to a spe
put was about $7.40 per page in fiscal year 1986 cific word processing or typesetting system.
93

Data elements are tagged with an identifier Table 4-27.—Departmental Applications of GPO
that can be used to control the format of a par- Structured Full Text Database Standard,
as of November 1987
ticular document. Users who agree on and im- — —
plement a common set of structured database Number of publications
standards are able to change the database eas- or publication series
Depart ment using GPO standard
ily and cost-effectively from one location to
Agriculture . . . . . . . . . . . . . . . . . . . . . . . . . 12
another, one format to another, and one revi- Air Force . . . . . . . . . . . . . . . . . . . . . . . . . . 4
sion to a later revision or edition. Army . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Commerce . . . . . . . . . . . . . . . . . . . . . . . . . 38
GPO uses its own structured database Defense (other) . . . . . . . . . . . . . . . . . . . . 13
standard for the vast majority of materials Energy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
HHS a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
composed and produced at the GPO main Interior . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
plant. However, the GPO standard is not, at Justice . . . . . . . . . . . . . . . . ... , . . . . . . . . 3
present, generally accepted by private indus- Labor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Navy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
try and significant parts of the Federal Gov- Transportation . . . . . . . . . . . . . . . . . . . . . . 11
ernment. For example, DoD is committed to Treasury . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
the Standard Generalized Markup Language State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
(SGML) approach, which is similar to GPO’s FIH lth and HUrnan sewlces
ea

approach, but still a distinctly different stand- SOURCE U.S Government Prfntlng Off Ice, 1988
ard. Many of the desktop and high-end elec-
tronic publishing systems located in Federal
agencies use display oriented standards (what
you see on the screen is what you get in the tail earlier, the executive branch has made and
document) and/or structured database ap- continues to make a major investment in elec-
proaches different from GPO’S. tronic publishing and related technologies,
typically as part of agency automation pro-
GPO indicates that software could be writ- grams. While implementation varies widely,
ten to convert SGML (or presumably any other electronic publishing is conceptually viewed
markup language) automatically into GPO’s as part of agency information resources nlan-
logically structured full text database ap- agement (IRM), and staffing, budgeting, and
proach. Also, GPO is prototyping a desktop the like are evolving within the IRM frame-
microcomputer-based version of its structured work. Many agencies are committed to a tran-
database software. sition from paper to electronic-based opera-
In sum, as noted in Table 4-27 there is signifi- tions, although the transition is likely to take
cant use of GPO’s database standard. But this many years.
percentage of use is only a small fraction of all For the executive branch, several roles for
government publications. This suggests a sig- GPO are emerging. First, GPO can continue
nificant, unrealized opportunity to apply GPO’s to provide traditional printing services either
or some other approach as a broadly accepted at the well-equipped main plant or via outside
and acceptable government wide database procurement. Second, GPO can continue to im-
standard. prove cost-effectiveness at the input and
prepress end of the printing process by en-
Electronic Publishing Support couraging electronic submissions, already at
Under the electronic GPO-decentralized al- very high levels, and dial-up composition serv-
ternative, GPO would need to develop an over- ices, where appropriate. Third, GPO can en-
all electronic publishing strategy that lever- courage adoption of governmentwide struc-
ages GPO strengths to meet changing needs tured database standards as discussed above.
of the Federal agencies. This strategy is likely Fourth, GPO can provide or facilitate mecha-
to differ for the executive and legislative nisms for training and education about elec-
branches of government. As presented in de- tronic publishing.
.. . . .

94

On the one hand, GPO’s greatest assets are manuals, directives, and technical documen-
its traditional printing facilities and labor tation on electronic media for filing, revising,
force, coupled with a substantially automated updating, and disseminating.
prepress capability, including electronic pho-
GPO’s high dependence on printing for de-
tocomposition and typesetting. GPO is well fense customers is concentrated at several of
suited for agency work requiring typeset qual- the regional procurement and printing facil-
ity ink-on-paper output, for large documents ities, as shown in Table 4-29. In the extreme
and long press runs, and for a variety of spe-
case that most traditional ink-on-paper defense
cialty jobs. GPO’s structured database stand-
work was phased out (over a period of several
ard, or some variation thereof, is well suited to many years), only the GPO main printing
to provide cost-effective linkages between doc- plant, rapid response center, Chicago, Denver,
ument origination, revision, and multi-format New York, and San Francisco regional print-
dissemination, regardless of who is doing the ing plants, and Denver and Seattle regional
disseminating. On the other hand, many of the procurement offices would be substantially un-
executive agencies are committed to acquir-
affected. All other offices could lose between
ing and implementing their own electronic pub-
lishing and dissemination capability, largely
as part of agency automation programs in
which GPO has little or no involvement. Some Table 4.29.—Distribution of GPO Defense Customers
agencies, and especially the defense and regu- by Procurement and Printing Offices, Fiscal Year 1986
latory agencies, are determined to reduce their
Total Defense Agency
dependence on paper drastically within the billings, fiscal year 1986a
next few years. Dollars in Percent of
mil I ions office total b
The plans and activities of defense agencies
Procurement offices
are particularly important, since, as shown in Boston Regional . . . . . . . . . 3.3 69.9
Table 4-28, the Army, Navy, and Air Force to- Philadelphia Regional . . . . . 27.0 84.8
gether account for about $250 million in GPO New York Regional . . . . . . . 7.3 58.2
Hampton Regional . . . . . . . 19.5 89.7
billings or roughly one-third of total GPO bill- Atlanta Regional . . . . . . . . . 23.7 71.1
ings. The Navy, for example, has established Chicago Regional . . . . . . . . 21.8 73.6
the “paperless ship” as a prime directive. All Columbus Regional. . . . . . . 8.3 68.3
St. Louis Regional. . . . . . . . 12.0 70.4
military services are hoping to place most Dallas Regional . . . . . . . . . . 13.7 61.4
Denver Regional . . . . . . . . . 3.5 23,7
Los Angeles Regional . . . . 7.1 77.0
Table 4-28.—Ten Largest GPO Printing Customers, San Francisco Regional . . . 7.1 52<7
Seattle Regional . . . . . . . . . 2.7 26,4
Fiscal Year 1986 San Antonio Satellite . . . . . 2.6 95.6
Charleston Satellite . . . . . . 1.1 85,2
Fiscal year 1986 billings San Diego Satellite . . . . . . . 0.2 99.4
Percent of Oklahoma City Satellite . . . 0.07 100.0
Dollars in fiscal year Rapid Response Center . . . 1.2 9.7
Customer mill ions 1986 total a
Printing Offices
Army . . . . . . . . . . . . . . . . . . . 134.7 18.3 Chicago Regional . . . . . . . . 0.02 1.0
Navy. . . . . . . . . . . . . . . . . . . . 74.6 10.1 Denver Regional . . . . . . . . . 0.1 4.8
Congress . . . . . . . . . . . . . . . 68.0 9.2 New York Regional . . . . . . . 0.5 34.3
Treasury . . . . . . . . . . . . . . . . 65.6 8.9 San Francisco Regional . . . 0.1 16.8
Postal Service . . . . . . . . . . . 53.0 Seattle Regional . . . . . . . . . 0.2 59.6
HHS b . . . . . . . . . . . . . . . . . . . 49.8 ::: Rapid Response Center . . . 1.4 15.5
Air Force. . . . . . . . . . . . . . . . 48.0 6.5 Main Piant
GSA . . . . . . . . . . . . . . . . . . . . 23.0 3.1 Procurement. . . . . . . . . . . . . 63.6 19.3
Agriculture . . . . . . . . . . . . . . 21.8 3.0 Printing . . . . . . . . . . . . . . . . . 17.2 9.3
Energy . . . . . . . . . . . . . . . . . . 19.3 2.6 a For Army, Navy, Air Force, Defense Logistics Agency, and Other Defense
aperCent of f~~Cal year IgSS GPo total revenues of $737 millionnet of wblica. Agencies
tions sales, appropriations, and reimbursements bDefense agency billings calculated as a percentage of total billings fOr each
bHealth and Human services. Office
SOURCE U.S Government Prlntlng Office, 19SS SOURCE: U S. Government Printing Office, 19SS
-.. ... —-. —.

95

half and over 90 percent of current billings, would benefit from staying abreast or, perhaps,
all other things being equal. ahead of agency applications and thereby be
in a much better position to identify opportu-
Clearly, then, an important part of GPO’s
nities to meet agency needs. In a decentralized
electronic publishing strategy would logically
and competitive electronic environment which
be a very careful analysis of how defense auto-
increasingly characterizes the Federal Govern-
mation activities are likely to affect the de-
ment, GPO will have to be imovative in match-
mand for traditional GPO printing services
ing its expertise with agency needs. Agency
over what time frame, and what are the lever- needs will vary over a wide spectrum and will
aged opportunities for GPO with respect to the
change over time at an increasingly rapid pace.
emerging defense electronic information infra-
Some agencies will look to GPO for a wide
structure.
range of electronic publishing services, assum-
Such an analysis would require, first, that ing such services are available, while others
GPO obtain basic information about defense will be completely independent of GPO, regard-
agency automation plans. Up to now, GPO has less of what GPO offers.
directly participated in only a handful of de-
At this point in time, it is not possible to
fense automation programs, including the
determine with precision the extent of agency
Army Programs 600-S (terminated before con-
needs for GPO electronic publishing services.
tract award) and 400-S (contract awarded in
In 1986, GPO conducted its own survey of
1984) and the Air Force Program 50-S (con-
agency needs. The results (based on 175 of 850
tract awarded in 1988). However, these three
questionnaires returned—a 20 percent re-
electronic publishing programs represent only
sponse rate) strongly indicate that almost all
a fraction of all relevant DoD activities. For
agencies will be using electronic publishing
example, GPO did not participate in the
within 5 years, especially for reference mate-
Navy’s “Printing on Demand System” de-
rials, technical documentation, and periodicals,
signed to produce 15,000 15-page documents
but the role of GPO is much less clear. Roughly
per day on an electronic printing-on-demand one-fifth to one-quarter of the respondents in-
basis. dicated a near-term preference for GPO auto-
In addition to keeping more fully informed matic composition, computer-aided page mak-
on agency automation activities (both military eup, and typeset quality output, increasing to
and civilian), GPO could establish an electronic about one-third of the respondents in 5 years.
publishing laboratory and innovation center This compares to the roughly two-thirds of the
for both GPO and agency personnel. GPO al- respondents who indicated an overall intent
ready has taken some action along these lines to use these electronic publishing services in
with respect to establishment of the “dial-up 5 years. This suggests that perhaps up to about
composition service’ now available. This serv- one-half of these types of agency electronic
ice permits agencies to originate material from publishing work might be done by GPO, and
agency microcomputers, transmit the material the other half by the agencies themselves (or
over telephone or fiber optic lines to GPO for by agency contractors). The results suggest
typesetting, and receive the typeset material a relatively smaller role for GPO with respect
via transmission back to the agency for print- to text input and editing, electronic display,
ing of proof copies on agency laser printers. data telecommunicating, and computer gen-
This dial-up service uses GPO’s logically-struc- erated graphics, although the overall use of
tured database, and GPO provides both a dem- these GPO services was still projected to grow
onstration room and a training program. significantly.
This concept could be expanded to a much The response rate to this 1986 survey was
wider range of electronic publishing applica- low, and both agency and GPO activities–as
tions, including high-end and optical disk con- well as the underlying technologies-have
figurations. From a strategic perspective, GPO changed markedly since then. Clearly, a new
96

survey is needed and hopefully with a much vestment in the technical infrastructure that
higher response rate. Some highlights from the makes decentralized executive branch activi-
1986 survey are shown in Table 4-30. ties a reality. Second, the legislative branch
Several of the agency respondents indicated generally does not have a large number of staff
in written comments that use of GPO elec- already trained in electronic publishing. Third,
tronic publishing services would depend in GPO is in the legislative branch, so separation
large part on whether such services were cost- of powers concerns do not apply. Fourth, GPO
effective compared to in-house costs or com- already has a central role in many legislative
mercial rates. In essence, GPO is competing branch publishing activities. And fifth, many
for agency electronic publishing business of GPO’s own pilot projects involve the legis-
lative branch, such as the fiber optic links be-
against agency inhouse, contractor, and com-
mercial vendor alternatives. GPO is already tween GPO and the Senate Office of Legisla-
moving to provide more electronic publishing tive Counsel, House Office of Legislative
options, but the pace is still much slower than Counsel, and House Information Systems Of-
the rates of change in technology, agency activ- fice (HIS).
ities, and user needs.
For all of these reasons, GPO could develop
While GPO training and innovation activi- plans for an expanded role with respect to the
ties are relevant to all branches of government, legislative branch. These plans could include the
the role of GPO with respect to legislative GPO provided capability for congressional com-
branch electronic publishing could be differ- mittees and offices to search, retrieve, and print-
ent in several key ways. First, the legislative on-demand key governmental process documents
branch has not yet made the major capital in- such as the Congressional Record, Record Index,

Table 4-30.—Federal Agency Electronic Publishing Activities and Plans,


as of 1986 in Percent of Agencies Responding

Types of documents for which electronic


publishing is/will be used Currently In 5 Years
Reference Materials . . . . . . . . . . . . . . . . . . . . . . 35’/0 57 ”/0
Technical documentation . . . . . . . . . . . . . . . . . . 37 56
Periodicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 48
Throwaway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 36
Catalogs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 31
Legal documentation . . . . . . . . . . . . . . . . . . . . . 10 20
Electronic publishing services that
are/will be used Currently Next year In 5 years
Text input and editing (microcomputers,
word processors) . . . . . . . . . . . . . . . . . . . . . . . 75’/0 820/o 87°\o
Automatic composition (software/systems) . . 25 40 56
Computer generated graphics . . . . . . . . . . . . . . 42 59 77
Computer-aided page makeup, . . . . . . . . . . . . . 17 34 61
Typeset quality output . . . . . . . . . . . . . . . . . . . . 34 45 57
Electronic display . . . . . . . . . . . . . . . . . . . . . . . . 33 43 58
Data telecommunicating . . . . . . . . . . . . . . . . . . 43 54 68
GPO electronic publishing services
that will be used Next year In 5 years
Text input and editing . . . . . . . . . . . . . . . . . . . . 90/0 140/0
Automatic composition . . . . . . . . . . . . . . . . . . . 19 28
Computer generated graphics . . . . . . . . . . . . . . 14 20
Computer-aided page markup . . . . . . . . . . . . . . 22 30
Typeset quality output . . . . . . . . . . . . . . . . . . . . 29 38
Electronic display . . . . . . . . . . . . . . . . . . . . . . . . 11 16
Data telecommunicating . . . . . . . . . . . . . . . . . . 13 18
SOURCE U.S Government Printing Off Ice, 1988.
97

Federal Register and Code of Federal Regula- GPO planning for an expanded offering of
tions, all of which are already in an electronic electronic formats would need to take into ac-
structured database format. This concept could count questions of demand, economies of scale,
be extended to include the capability to search, cost, private sector competition, and market-
retrieve, and print-on-demand selected commit- ing. The results of the GAO surveys of Fed-
tee prints, reports, and hearings, although the eral information users clearly indicated a grow-
degree of difficulty would be considerably ing demand for electronic formats, as discussed
greater since little of this material is currently earlier. As part of these surveys, GAO also
in structured database format. This problem asked respondents to estimate the usefulness
could be gradually eliminated overtime if more of a variet y of online and offline Federal infor-
committees utilized GPO electronic database mation formats. The depository library com-
and photocomposition capability, either on a munity indicated the strongest positive re-
dial-up or floppy disk basis. These alternatives sponse, with a clear majority of respondents
will become more widely possible as the pene- finding the illustrative items to be useful.
tration of microcomputers on Capitol Hill con- These results (based on responses from 318 de-
tinues. Any detailed planning along these lines pository libraries, out of a sample of 451) are
would need to take into account the inevita- summarized in Table 4-31 and the number of
ble increase in desktop publishing (software libraries rating each item as moderately use-
has decreased to the $600 per unit range) and ful, useful, or greatly useful out of the total
the complementary roles of GPO, the Library respondents for that item. The remaining per-
of Congress, HIS, and the Senate Computer centages (not shown in Table 4-31) include
Center. those libraries rating the item as somewhat
useful or having little or no use. The respond-
ents were asked to reply irrespective of how
Dissemination of Electronic Formats the formats might be provided (e.g., by the Fed-
Under the “Electronic GPO-Decentralized’ eral Government, commercial vendors, and/or
alternative, SupDocs would select agency elec- not-for-profit organizations).
tronic format information products for inclu- The depository library results suggest very
sion in the GPO sales program, presumably strong demand for the Congressional Record,
based on an evaluation of sales potential. Agen- calendars and bill status, the Federal Reg”s-
cies could sell such products as well, but could ter, an index to Federal information, and an
also opt, at agency discretion, to use SupDocs integrated database in both online and offline
as their sales outlet. electronic formats. Demand for agency press
releases and reports is weaker. In terms of in-
At the present time, SupDocs includes only tensity of demand, as measured by the percent-
a few dozen magnetic tape products in the sales age of libraries rating these items as greatly
program. These tapes are sold at the usual cost useful, the results indicate the highest rank-
plus 50 percent (in accordance with Title 44 ings for the index and integrated database (on-
of the U.S. Code) and include, for example, the line and offline) followed by the Reg”ster (on-
Congressional Record, U.S. Code, Code of Fed- line and offline), committee calendar and bill
eral Regulations, Federal Reg”ster, Statistical status (online), and Record (online and offline).
Abstract of the U. S., Budget of the United
States, Weekly Compilations of Presidential The information needs of depository libraries
Documents, and U.S. Government Manual. At could, of course, be met to some extent through
the moment, these magnetic tape products are DLP, as discussed in detail in chapters 6 and
sold primarily to commercial information ven- 7. However, the depository libraries serve as
dors–such as Mead, WE STLAW, DIALOG, a good indicator of demand among library and
Legi-Slate, Congressional Quarterly, and BRS information science professionals and those
—which repackage or enhance and resell the groups in American society that are the most
information. information-intensive.
98

Table 4-31 .—Depository Library Demand for Federal Information Electronic Formats

Percent of libraries responding


moderately to greatly useful
Online Offline CD-ROM
Item immediate access issued monthly
Congressional Record. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 74
Committee Calendar and Bill Status . . . . . . . . . . . . . . . . 70 60
Federal Register. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 80
Federal Agency Press Releases . . . . . . . . . . . . . . . . . . . . 46 40
Agency Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 62
Comprehensive Index to Federal Information . . . . . . . . . 94 90
Integrated Database of Key Federal
Statistical Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 88
SOURCE GAO Survey of Federal Information Users, 1988

GAO also surveyed scientific and technical associations reported having access to CD-
associations and general associations (trade, ROM readers compared to the libraries, as in-
professional, consumer, etc.). The results dicated in Table 4-33.
(based on 133 responses from a sample of 250
Clearly, depository libraries have better ac-
scientific and technical associations, and 134
cess to all categories of technology listed ex-
out of 350 general associations) are similar to
cept mainframe computers. Scientific and tech-
those of the depository libraries, but with a nical associations have generally better access
considerably lower level of intensity. In other than the general associations, with the excep-
words, while the relative preferences for vari- tion of microfiche readers, CD-ROM readers,
ous types of electronic information formats
and videodisk players, where the groups of
were roughly the same, the overall percentages
associations are about equal.
of respondents rating the items as moderately
to greatly useful were about half to three- The implications for SupDocs are several.
quarters that of the libraries for online access, First, there does appear to be an already sig-
and about one-third to one-half for offline ac- nificant demand for electronic formats, but,
cess. The survey results for the associations second, this demand at the moment is some-
are highlighted in Table 4-32. what ahead of the actual technical capability
of users, especially with respect to CD-ROM.
All categories of respondents indicated that
Taking all survey groups together, online de-
the index and integrated database would be
mand ranges from 34 to 94 percent of respond-
the most useful among the items included in
ents while microcomputer with modem capa-
the survey. Unlike the depository libraries,
bility ranges from 54 to 83 percent. Offline
which indicated little difference in usefulness CD-ROM demand ranges from 22 to 90 per-
of online versus offline formats, the associa-
cent, but CD-ROM reader capability ranges
tions showed a clear preference for online elec- from only 6 to 41 percent. However, third, con-
tronic formats. This may reflect, in part, differ-
tinually declining equipment costs mean that
ences in the nature of demand. For example,
the gap between user demand and technical
researchers using libraries may have a less ur-
capability is likely to close rather rapidly.
gent need for some types of Federal informa-
Microcomputers cost $1,500 or less, modems
tion and, therefore, might find monthly CD-
about $300, and CD-ROM readers about $700.
ROMS to be adequate. Many associations may
be primarily concerned with only the latest, For types of information where a demand
up -to the minute information that necessitates has been established, SupDocs would need to
online access. It is also probable that the asso- determine if including a particular item in the
ciations are less familiar with CD-ROM tech- sales program would be cost-effective and com-
nology than the libraries. Indeed, relatively few petitive relative to any other alternatives that
.

99

Table 4-32.—Scientific, Technical, and General Association Demand for Federal Information Electronic Formats

Percent of associations responding moderately to greatly useful


Scientific and technical
associations General associations
Online Offline CD-ROM Online Offline CD-ROM
Item immediate access issued monthly immediate access issued monthly
Congressional Record . . . . . . . . . . . . . . . . . . . . 34 ”/0 220/0 530/0 390/0
Committee Calendar and Bill Status . . . . . . . . 39 23 54 36
Federal Register . . . . . . . . . . . . . . . . . . . . . . . . . 40 27 55 41
Federal Agency Press Releases . . . . . . . . . . . . 52 30 54 36
Agency Reports . . . . . . . . . . . . . . . . . . . . . . . . . . 54 33 53 40
Comprehensive Index to Federal
Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 43 60 43
Integrated Database of Key Federal
Statistical Series . . . . . . . . . . . . . . . . . . . . . . . 63 42 63 43
SOURCE GAO Survey of Federal Information Users, 1988

Table 4-33.—Library and Association Access to


Information Dissemination Technology

Percent of libraries or
associations responding
Scientific
Depository and technical General
Technology libraries associations associations
Microcomputer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 700/0 640/o 51 “/0
Microcomputer with modem (for online access) . . 83 65 54
Microfiche reader. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 24 22
Microfiche reader with printer. . . . . . . . . . . . . . . . . . 96 18 12
CD-ROM reader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 6 6
Videodisk player . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 11 12
Mainframe computer (for tape/disk access) . . . . . . 37 36 27
SOURCE. GAO Survey of Federal Information Users, 1988

may be available to users. For many types of product. Some items, such as a government-
Federal information, individual agencies and wide information index, could be developed by
private sector vendors might decide to mar- SupDocs and/or NTIS, have an apparently
ket electronic formats. In other words, Sup- broad demand, and could be sold in both on-
Docs would be operating in a more competi- line and CD-ROM formats. SupDocs would
tive environment than has traditionally been need to determine if electronic format products
the case with respect to paper formats. Thus, could be produced and sold at an acceptable
for example, at present the Bureau of the price. For example, most depository library re-
Census sells paper formats via the SupDocs spondents to the GAO survey indicated that
sales program, but sells magnetic tapes and they would be willing to pay no more than $49
floppy disks itself and also offers online elec- per hour online and $199 per CD-ROM for ac-
tronic bulletin board services. Both the Bureau cess to a governmentwide index. However,
of the Census and private vendors are devel- most of the associations responding indicated
oping CD-ROM products, and some Census that they would be willing to pay no more than
CD-ROM products will be disseminated as part $24 per hour online and $19 per CD-ROM. OTA
of DLP. has not done a detailed analysis of this data,
but $24 per hour online is in line with non-profit
The approach taken by SupDocs in decid- rates for similar information products. And $19
ing what to include in the sales program could (or less) per CD-ROM is realistic at production
vary depending on the particular information volumes of over 1,000 or so disks. If the index
100

on CD-ROM were distributed to depository Perhaps the most difficult format for Sup-
libraries, this would guarantee a base volume Docs could be online. Whereas SupDocs could
of about 1,400 disks. Initial demand for a CD- ride the agency orders for CD-ROMs, magnetic
ROM index product could easily be in the sev- tapes, and floppy disks, regardless of where
eral thousands, based on the GAO survey re- and by whom the copies were produced, it is
sponse. Again, since the index information hard to conceptualize riding an online data-
would not be copyrightable, and assuming the base. It seems unlikely that, as a general rule,
electronic master tape (or the equivalent) would agencies and SupDocs would be offering the
be available for purchase, private companies same online databases. SupDocs could offer
could put the index up as a file on DIALOG agency online databases at agency discretion,
and similar value-added database services, and or could offer a gateway to agency databases.
even could add value and sell an enhanced CD- Also, SupDocs could contract with a private
ROM product. commercial (or non-profit) gateway service.
Further, private gateway or value-added data-
In general, CD-ROM products can be pro- base companies could contract directly with
duced at low unit costs at volumes over 500 individual agencies and/or purchase the mag-
to 1,000 disks. Thus, there would likely be a netic tapes, as some do today.
growing number of opportunities for SupDocs
to “ride” the order for production of agency On the other hand, SupDocs could serve as
CD-ROMs, where a sufficient market exists, the primary Federal outlet for online access
just as SupDocs now rides the agency orders to key governmental process information items
for printed ink-on-paper products. Presuma- such as the Congtessiona.1 Record and Federal
bly, the mastering and duplications of CD- Re~”ster. These kinds of items are all well
ROMS would be contracted out to the private suited to online format because the informa-
sector, by either the agency or GPO, at least tion is frequently time sensitive and of selec-
until such time that inhouse government ca- tive interest. That is, many users are not in-
pability might be more cost-effective. terested in reading these documents cover to
cover at their leisure, but, instead, want to
As for other formats, the market for mag- quickly search for and retrieve information on
netic tapes is probably not going to be large selected topics of interest. The GAO survey
in the forseeable future, due to the need to have results suggest that there would be broad de-
a mainframe or minicomputer and related mand for these items if priced below $24 per
peripheral equipment. Major customers are
hour. Since items such as the Record and Reg-
likely to continue to be the value-added ven-
ister are bought by vendors in magnetic tape
dors and scientific or research organizations.
format from SupDocs and then put online and
Government experience to date (at GPO and
sold at a significant mark-up, it seems plausi-
various agencies) is that sales in the hundreds
ble that SupDocs could itself offer these items
of copies per year are considered good. Simi-
online at a competitive price. SupDocs could,
larly, sales of floppy disks to date by NTIS
of course, itself contract with a private gate-
and various agencies have been minimal.
way or database vendor. SupDocs offerings
Floppy disks can be produced at only $1 to $5
would not necessarily have any significant im-
dollars per unit, compared to about $100 to
pact on private services, since the markets
$200 for magnetic tapes (depending on bit den- served may be quite different. Again, detailed
sity). Also, floppy disks can run on the increas-
feasibility and marketing studies would be
ingly commonplace microcomputers. Thus, the
needed.
potential market for floppy disks would appear
to be large compared to magnetic tapes. How- Overall, the development of a rational and
ever, detailed market analyses are needed to workable plan for SupDocs sales of electronic
establish reliable estimates. formats would require close consultation and
101

coordination with mission agencies and espe- Table 4-34.—GPO Union Bargaining Units,
cially those agencies that already have clear- as of April 1987
inghouse or gateway functions for electronic Number of
formats, such as NTIS and NLM. NTIS cur- Bargaining unit employees
rently serves as a clearinghouse for some American Federation of Government
agency floppy disk and magnetic tape prod- Employees (AFGE) Local 2876/Printing
Crafts Joint Council (Main Plant White
ucts, and NLM currently offers several agency Collar Workers) . . . . . . . . . . . . . . . . . . . . ., . . 1,327
online database services. For paper formats, Washington Federal Printing Workers’ Union
SupDocs has included in the sales program pri- (Member of GCIU see below) Local 713-S
marily items judged to have significant de- (Printing plant workers) . . . . . . . . . . . . . . . . 1,020
mand, given the economics of traditional print- Columbia Typographical Union Local 101
(2 units, Composers and Rapid Response
ing which penalizes small press runs and given Center). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 672
the need to spread overhead, processing, and Graphic Communications International Union
marketing costs over as large a sales volume (GCIU) Local 4-B (2 units, Bookbinders and
as possible. However, somle electronic formats and Journeymen Bindery) . . . . . . . . . . . ... 247
could be economically viable at much lower Washington Printing and Graphic
Communications Union (member of GCIU)
sales volumes. To the extent SupDocs might Local 1-C (2 units, Pressmen and Masonry
seek to include low demand and perhaps even Workers) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
printing-on-demand items in the sales pro- GCIU Local 285 (Offset Strippers) . . . . . . . . . . 138
gram, then SupDocs would be taking on NTIS- Washington Government Photo Offset Union
like functions. This would intensify the need (member of GCIU) Local 538-C (Offset
Platemaker Strippers) . . . . . . . . . . . . . . . . . . . 69
to consider SupDocs-NTIS relationships, as
AFGE Local 3392 (Pueblo Distribution
will be discussed in chapter 5 and 12 in more Center). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
detail. International Brotherhood of Electrical
Workers (IBEW) Local 121 (Electricians and
Sanitary Engineers) . . . . . . . . . . . . . . . . . . . . 65
Staffing AFGE Local 2738 (Police) ... , . . . . . . . . . . . . . 62
GPO faces two major challenges with respect AFGE Local 1248 (Denver Printing Plant) . . . . 42
to staffing: retaining the necessary skilled la- International Association of Machinists
Local 2135 (Machinists) . . . . . . . . . . . . . . . . . 36
bor force to maintain traditional printing serv- AFGE Local 1292 (Chicago Printing Plant) . . . 27
ices at a level commensurate with demand, and
Sheetmetal Workers’ International Union
obtaining personnel with the new skills needed Local 100 (Sheet Metal Workers and Pipe
to implement GPO’s future role in electronic Fitters) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
publishing and electronic information dissem- United Brotherhood of Carpenters and
ination, however that role may be defined. Joiners of American, Local 2456
(Carpenters) ... , . . . . . . . . . . . . . . . . . . . . . . . 17
AFGE Local 2618 (New York Printing Plant). 14
As noted earlier, GPO has a relatively old Brotherhood of Painters and Allied Trades,
labor force, with about thirteen percent of all Local 1632 (Painters). . . . . . . . . . . . . . . . . . . . 9
current employees eligible for retirement (and Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4,065
aNumber of employees represented by their Union, nOt al I emplOyees represented
up to 25+ percent in some key areas). With are union members
a natural attrition rate of 5-10 percent (retire- SOURCE U S Government Prlntlng Off Ice, 1988
ments and quits), GPO has considerable flexi-
bility to reshape the labor force to match fu-
ture needs. About 80 percent of the GPO labor Collective bargaining has been able to accomm-
force is unionized and works under collective odate major changes in the size and job struc-
bargaining agreements. The twenty union bar- ture of the GPO labor force over the past fifteen
gaining units and the approximate number of years, responding in large part to technological
employees in each are listed in Table 4-34. change in composition, prepress, and press tech-
102

nology. Collective bargaining should be able to off set presses for the printing of the Cozzgres-
accommodate future changes, so long as labor sional Record and Federal Register at an esti-
and management work closely together and bar- mated cost of about $10.5 million for the two
gain in good faith. units. OTA’S independent printing consultant
endorsed this capital investment on the
At the moment, one possible impediment to grounds of improved efficiency and produc-
successful labor-management relations, as tivity. However, GPO’s rationale for this in-
pointed out by OTA’S independent labor con- vestment presumes that traditional printing
sultant, is the absence of a clear strategic vi- of the Record and Re~”ster will continue for
sion of GPO’s future role. The lack of a clear at least 10 years substantially unchanged from
vision not only contributes to employee uncer- today. As noted earlier, the Record and Reg-
tainty, but makes staffing decisions difficult. ister are well suited to online and offline elec-
It seems likely that, absent major changes in tronic formats for which there is growing de-
GPO’s traditional printing role as discussed mand. Should Congress decide to make these
earlier, GPO should be able to continue its pol- publications available online and through CD-
icy of no involuntary reductions in force (gov- ROMS issued periodically to the legislative
erned by a May 1982 resolution of the Joint branch and depository libraries and on a sales
Commi ttee on Printing). Any reductions in the basis via SupDocs, then it is conceivable that
traditional labor force should be able to be han- the paper format versions of the Record and
dled through retirements and reassignments. Reg”ster could be reduced significantly in a few
It also seems likely that, if GPO pursues a sig- years. Indeed, the volume of paper copies could
nificant role in electronic publishing and dis- be reduced to the point where the large web
semination, GPO would need to bring in new offset presses would no longer be cost-effective.
skills from outside. While some existing per- For example, even if paper copies were still pro-
sonnel undoubtedly could be retrained for new vided to every Member, committee, and office
jobs involving electronic processes and for- of Congress, every depository library, mem-
mats, as has been done in the past, some new bers of the press, high-level executive officials,
personnel with advanced engineering, techni- and Federal and State archival agencies, the
cal, and marketing education and experience required press run of several thousand copies
would be required. The exact skills mix of could be uneconomical for the large presses.
retrained personnel and new hires cannot be
determined in the absence of an overall stra- GPO notes that a change of this magnitude
tegic plan. would conflict with current provisions of Ti-
tle 44 that require the printing and distribu-
Capital Investment tion of specified numbers of the Record and
Another important element of GPO’s over- Register. However, electronic formats could
all strategic plan would be capital investment be provided first as a complement to paper and
alternatives. Again, a detailed capital invest- eventually, after a transition period, as a sub-
ment program would require a well developed stitute, and Congress could amend Title 44 if
strategic plan. Short of that, it would seem pru- necessary. GPO also notes that any excess ca-
dent for GPO to reevaluate carefully its capi- pacity on the two new web offset presses could
tal investment plans in light of possible ad- be used to absorb workloads from other, older
justments to traditional printing services and equipment, and to facilitate a gradual phaseout
possible new initiatives in electronic publish- of some of that equipment. In addition, the new
ing and dissemination. presses would be less labor intensive and would
be technologically up-to-date. GPO and Con-
As discussed earlier, GPO has already sub- gress need to carefully evaluate whether, even
stantially updated its main plant press and if electronic formats are encouraged or re-
bindery equipment. The major outstanding quired, the remaining volume of paper copies
item is the pending purchase of two new web is sufficient to justify use of the large web off-
103

set presses or, if not, whether the other advan- Table 4-35.—Selected GPO Electronic Composition
tages noted above would by themselves be Equipment, a Fiscal Year 1987
compelling. Number
Equipment of units Year(s) acquired
Future capital investment in the prepress, Personal computers . . . . . . . . . . 12 1984-1987
press, and bindery areas should also reflect any Video display terminals ., . . . . . 160 1978-1987
decisions on changing the work load distribu- Multi processor control
system . . . . . . . . . . . . . . . 4 1983-1987
tion at the GPO main plant. The main plant PDP 11-44 minicomputer . . . . . . 2 1981,1985
carries out a much more diverse range of print- Floppy disk reader . . . . . . . . . . . 1 1985
ing work than almost all private printing com- Text editing systems (ATEX) . . . 16 1978-1987
Text editing system
panies. GPO could consider some greater de- (Videocomp) . . . . ... . . . . . . 1 1986
gree of specialization in order to help reduce Photocomposers (Videocomp,
indirect labor and overhead costs. (The fiscal Comp 80). . . . . . . . . . . . . . . . . . 4 1976-1981
.-
aEXcludes Electronic Job Sect Ion
year 1986 cost allocation for the main plant
SOURCE U S Government Prlntlng Off Ice, 1988
production department was about 31 percent
direct labor, 34 percent section burden [indirect
labor, materials, etc.], 28 percent overhead solete. The Compugraphic system is used for
[general management and staff, utilities, rent, fully processing about 60 percent of the work
etc.], 2 percent depreciation, and 5 percent done in this section, and is used for partially
other [supplies, maintenance, etc.]). Presuma- processing (in conjunction with ATEX or other
bly future capital investments would be made systems) another 35 percent of the work. While
primarily in those areas designated as GPO performance of the Compugraphic appears to
specialities. Also, the shift to electronic for- be satisfactory, the now 3-year old system is
mats for the Record and Re~”ster could fur- obviously not state-of-the-art.
ther reduce indirect labor and general overhead
since the overnight main plant operations In essence, GPO’s capital investment strat-
could be scaled back although not eliminated, egy depends in part on whether GPO strives for
due to the continuing need for input to and cre- (or is directed to take) a leaders~p role in elec-
ation of the online databases by the next day. tronic publishing technology. An effective leader-
ship role probably requhes a heatier investment
With respect to composition equipment, in state-of-the-art technology, partly to learn
OTA’S independent printing consultant con- about the technology for GPO’s own purposes
cluded that GPO’s current equipment is strongly but, equally important, to also at least stay
competitive with private industry. GPO uses an abreast of the mission agencies, some of which,
ATEX minicomputer-based text editing sys- at this point in time, are well ahead of GPO. For
tem and Videocomp and Comp80 phototype- example, GPO has no significant activity
setters. As with press and bindery, the com- underway in optical disk or compact disk tech-
position equipment has been substantially nologies and expert information retrieval sys-
updated over the past decade, as highlighted tems, and is behind the state-of-the-art in high-
in Table 4-35. end electronic publishing work stations and
software, all of which are under active testing
One area where GPO is not competitive is or actually being implemented by various
high-end electronic publishing equipment. The
agencies.
ATEX system is designed to handle large text
files and is not well suited for smaller and With respect to the provision of online data-
specialty jobs involving complex layouts, bases, GPO would need to decide whether ex-
graphics, and the like. To help meet this need, isting computer capability would be adequate
GPO established an Electronic Job Section and, if not, whether to purchase or lease acicii-
equipped with Compugraphic and Bedford tional capability or whether to, at least ini-
electronic publishing systems, among other tially, utilize the services of private sector
equipment. The Bedford system, although two value-added gateway carriers and database
years old, is rarely used and is essentially ob- providers. For example, if SupDocs decided to
sell the Congressional Record online, the Rec- available via the FTS-2000, when implemented,
ord could be established as a file on The Source, and could make use of advanced satellite and
CompuServe, Easylink, and/or DIALOG. This fiber optic transmission technologies embedded
would minimize GPO’s capital investment re- in FTS-2000 and various commercial telecom-
quirements until experience with actual de- munication systems.
mand levels and patterns could be analyzed.
In an era of constrained resources, GPO may
Alternatively, or in addition, the online Rec-
have to make some difficult choices between
ord could be set up as a file on NLM’s MED- investment in traditional versus electronic
LARS, on the gateway system operated by the
publishing technology, and between capital in-
Defense Technical Information Center, andlor
vestment versus the training and recruitment
on the library community’s various networks.
of personnel to apply the technology. These
There are numerous possibilities, especially for
decisions are best made within an overall stra-
key governmental process information such as
tegic framework.
the Record and Reg”ster. Eventually, SupDocs
online information products could be made

Chapter 5

An Electronic National Technical


Information Service and
NTIS/Superintendent of
Documents Cooperation

- -

Clockwise from top left: NTIS staff searching the NTIS database for a customer; NTIS staff “blowing back” from microfiche
to produce a paper copy of a technicalreport; NTIS staff reproducing additional shelf stock; and NTIS staff pulling an
archive document from the NTIS collection (photo credits: National Technical Information Service).
.

CONTENTS
Page
summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................107
Role and Current Status of NTIS. . . . . . . . . . . . . . . . . . . . . . .............108
Opportunities and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ....111
Trends in Demand and Revenues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...111
Possible New Initiatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........116
NTIS/SupDocs Cooperation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , .........119
Differences and Similarities . . . . . . . . . . . . . . . . ......................119
Disadvantages and Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....121

Tables
Table Page
5-1. Source of NTIS Reports, Fiscal Year 1987 . . . . . . . . . . . . . . . . . . . . . .108
5-2, Distribution of NTIS Sales, Fiscal Year 1987 . . . . . . . . . . . . . . . . . . ..1o9
5-34 NTIS Revenues and Costs, Fiscal Years 1980-87 . ................112
5-4. Demand for Selected Major NTIS Products, Fiscal Years 1980-87. ..112
5-5. Demand for Selected Minor NTIS Products, Fiscal Years 1983-87. ..113
5-6. Average Per Unit Prices for Selected Major NTIS Products
Compared to Inflation Rate, Fiscal Years 1980-87 . ...............113
5“7. NTIS Revenues by Product Group, Fiscal Years 1986-87 . .........113
5-8, Federal Agency Evaluation of the Cost of NTIS Reports, Paper and
Microfiche Formats, 114 Agency Components Responding . ........114
5-9. Trend in New Titles Received by NTIS, Fiscal Years 1983-87 .. ...,114
5“10. Federal Civilian Agency Dissemination of Scientific and
Technical Informa~ion ; . . . . . . . . . . . . . . .........................115
5-11. Scientific and Technical Association Use of Federal Information
Sources, Rank Order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........115
5-12. Demand for NTIS Documents Announced in Calendar Year 1986 ...116
5-13.AgeofNTIS Documents Sold in Calendar Year 1986 . ...........,116
5-14. SupDocs Sales and Distribution Activity, Fiscal Years 1981-8’7 .. ...122
5-15. Conventional Printing Functions Affected by Length of Press Run. ,123
5-16. Estimated Page Length and Content of Government Documents. . ..123
Chapter 5
An Electronic National Technical
Information Service and
NTIS/Superintendent of
Documents Cooperation
SUMMARY
This chapter discusses the current status of diversity and timeliness of NTIS (and related
and future prospects for the National Techni- private vendor) offerings, increase the ability
cal Information Service (NTIS), and opportu- of NTIS (and private vendors) to match infor-
nities for cooperation between NTIS and mation products with potential users, and re-
GPO’s Superintendent of Documents (Sup- duce the cost of NTIS products. An electronic
Docs). The debate over the privatization of NTIS should be better able to serve all users,
NTIS is discussed in chapters 11 and 12. The but especially small and medium businesses
discussion in this chapter assumes that this and individual researchers.
debate will be resolved by Congress in favor
of retaining NTIS within the Federal Govern- NTIS/SupDocs cooperation could create new
ment— as a separate agency or government cor- opportunities for improvements in the index-
poration within the Department of Commerce ing, marketing, and international exchange of
or consolidated with SupDocs or even with the Federal information. NTIS/SupDocs cooper-
Library of Congress. The major challenge fac- ation could be synergistic with respect to im-
ing Congress is defining a viable role for NTIS plementing an electronic document system
in the future. A variety of indicators strongly that would meet NTIS needs plus a broaden-
suggest that the current role may not be sus- ing of the SupDocs product line to include
tainable absent some significant changes. selected low demand items. The NTIS/Sup-
Docs combined low-demand sales volume could
NTIS operates in a highly constrained envi- help justify investment in the necessary equip-
ronment, characterized by ambivalent support ment, which could be funded out of the GPO
from the executive branch, limited financial revolving fund and/or NTIS retained earnings
resources, mixed support from the information
(if authorized) and charged back as deprecia-
industry, limited technical resources, and a dif-
tion. NTIS/SupDocs cooperative initiatives
ficult product mix (many low volume items). would need to be sensitive to concerns about
In addition, the basic demand for NTIS prod-
separation of powers between the executive
ucts appears to be significantly eroding. Most
and legislative branches, and about the
NTIS users and client agencies believe in
strengthening of government wide dissemina-
the NTIS concept and seek to find ways to
tion mechanisms at the possible expense of de-
strengthen NTIS or at least the core NTIS
centralized agency activities.
functions as a continuing element of the Fed-
eral Government. Regardless of the ultimate institutional struc-
NTIS appears to be ideally suited for the im- ture, there are significant opportunities for im-
plementation of an electronic document system provement in both NTIS and SupDocs product
(with multi-format output—paper, microfiche, or line analyses, development, and marketing.
electronic), regardless of organizational location. Strengthened cooperation between NTIS and
NTIS could use aversion of the Defense Tech- SupDocs would not only help identify mutually
nical Information Center (DTIC) system as a advantageous joint activities, but would seem
prototype. An electronic document system almost mandatory to the extent that both agen-
could help revitalize NTIS if coupled with im- cies pursue sales of electronic format products
proved agency participation. Overall, an elec- and that SupDocs enters the low-demand
tronic NTIS should be able to increase the market.
107
108

ROLE AND CURRENT STATUS OF NTIS


The primary role of NTIS is to serve as a Table 5“1 .—Source of NTIS Reports, Fiscal Year 1987
central governmentwide source of scientific
Agency Percent of total
and technical reports describing research per-
Department of Defense . . . . . . . . . . . . . . 23
formed by Federal agencies, contractors, and Department of Energy . . . . . . . . . . . . . . . 23
grantees. NTIS depends on the voluntary sub- NASA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
mission of these reports by the Federal agen- All other Federal agencies . . . . . . . . . . . 20
Non-Federal agencies . . . . . . . . . . . . . . .
cies. NTIS maintains a permanent archive of Foreign countries . . . . . . . . . . . . . . . . . . . 2;
these reports, establishes bibliographic control Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
over these materials, prepares various index SOURCE: National Technical Information Service, 1988
and abstract materials, and sells copies of the
reports. In recent years, NTIS has increased
its collection to include additional reports pre- Scientific and Technical Information Facility
pared by state and local governments and by (STIF)). The percentage distribution is shown
foreign government research organizations, in Table 5-1.
and to include Federal databases and software.
NTIS operates under several constraints.
In addition to the basic archival and clearing- One is the variable and limited funding com-
house functions, NTIS is responsible for: mitment of the government to NTIS. NTIS
began in 1945 as the Publication Board. The
the Federal Research in Progress (FEDRIP)
Publication Board was established by Execu-
Program that provides information
tive Order 9568, which charged the Board with
describing on-going Federally funded re-
reviewing all government-generated scientific
search projects;
and technical documents and determining
the Center for the Utilization of Federal
Technology (CUFT); what could be released to the public. Execu-
tive Order 9604 expanded the Board’s respon-
acquisition and licensing of government-
sibilities to include scientific and technical doc-
owned patents;
uments captured from the enemy during and
provision of production and billing/collec-
at the end of World War II. The Board’s ob-
tion services for information dissemina-
jectives were to organize declassified informa-
tion activities of other Federal agencies;
and tion so as to permit researchers, and especially
industry, fast and easy access to information,
provision and processing of FOIA re-
and to notify the public ‘and industry about
quests for agency materi~s placed on file
what was available. The intent was to promote
at NTIS.
economic growth and development through the
This discussion focuses primarily on the NTIS rapid dissemination of scientific and techni-
archival, clearinghouse, and dissemination cal information.
functions.
Since established, questions have been raised
As of fiscal year 1987, the NTIS archive in- concerning the appropriate functions and fund-
cluded close to 2 million reports and over 2,500 ing for NTIS. The Publications Board became
data and software files. About 60,000 to 70,000 part of the Office of Declassification and Tech-
new items are added each year. About half of nical Services (OTS) in late 1945, the Office
the NTIS reports originate from just three of Technical Services in 1946, the OTS Clear-
agencies: the Department of Defense (Defense inghouse in 1950, the Clearinghouse for Fed-
Technical Information Center (DTIC)), the De- eral Scientific and Technical Information in
partment of Energy (DOE) (primarily the DOE 1964, and NTIS in 1970. The history of NTIS
Office of Scientific and Technical Information has reflected uncertainty on the part of the
(OSTI)), and National Aeronautics and Space Federal Government as to the appropriate Fed-
Administration (NASA) (primarily the NASA eral commitment to a central clearinghouse for
109

dissemination of scientific and technical infor- for direct dissemination of reports to agency
mation, the role of the clearinghouse vis-a-vis personnel and contractors. NTIS then makes
the Federal science agencies and the private secondary distribution to the business commu-
sector, and the appropriate pricing of such nity and general public. In addition, the GPO
clearinghouse services. SupDocs includes some scientific and techni-
cal reports in the SupDocs sales program.
In general, representatives of the scientific NTIS includes some GPO titles in the NTIS
and technical community believe that the cost clearinghouse. However, the overlap is thought
of such clearinghouse services is a very small to be small, since SupDocs selects titles based
price to pay compared to the substantial Fed- on significant market potential (projected sales
eral investment in research and development. of several hundreds to thousands of copies),
To place this in perspective, the fiscal year 1987 whereas almost all NTIS includes titles are in-
NTIS revenues and costs were about $22 mil- cluded regardless of demand, which is gener-
lion each (breakeven operation) compared to ally very small (an average sales of 10 copies
the fiscal year 1987 Federal research and de- per title). In sum, NTIS must achieve break-
velopment budget of about $59 billion. Exclud- even operations working with a substantially
ing defense R&D, the NTIS operating budget incomplete collection of reports that sell very
of $22 million represents about one one- few copies on the average. This is a difficult
thousandth of the civilian R&D budget ($21.5 challenge.
billion in fiscal year 1987). Advocates of a
strong Federal role in dissemination of scien- A third major constraint is a complex rela-
tific and technical information argue that the tionship with the private sector and the infor-
level of Federal support is far too small. Others mation industry in particular. On the one hand,
believe that, while a Federal role is needed, it NTIS was established to help serve the scien-
should be limited in terms of functions and tific and technical information needs of busi-
budget. ness and industry. NTIS estimates that about
75 percent of its business customers are small
The result is that NTIS receives no appro- firms, with major corporations accounting for
priated funds for its basic archival and the other 25 percent of business customers.
clearinghouse functions, with costs covered by Overall, the U.S. business community accounts
sales of documents and services. NTIS does for about two-thirds of NTIS sales, as shown
not have a working capital revolving fund. As in Table 5-2.
a consequence, since any net revenues must
be returned to the U.S. Treasury, it has proven The information industry appears to be gen-
difficult for NTIS to obtain up-to-date equip- erally comfortable with the NTIS archival
ment—especially modern information tech- function and clearinghouse activities with re-
nology. spect to dissemination of paper and microfiche
copies. However, NTIS initiatives with respect
A second major constraint is that NTIS has
a voluntary relationship with the source agen- Table 5.2.— Distribution of NTIS Sales,
cies and cannot require agencies to submit ma- Fiscal Year 1987
terials. NTIS estimates that more than one-
third of Federal scientific and technical reports Percent of
NTIS sales,
are never submitted. There is also concern that Customer fiscal year 1987
agencies may delay submission of key reports U.S. business . . . . . . . . . . . . . . . . . . . . . . . . . 64
and/or submit primarily reports with less per- Foreign (business and government) . . . . . . 20
ceived interest or demand. NTIS functions, for U.S. Federal and State government . . . . . . 6
Academic researchers/institutions and
the most part, as a secondary distributor of public libraries . . . . . . . . . . . . . . . . . . . . . . 6
Federal scientific and technical information. General public . . . . . . . . . . . . . . . . . . . . . . . . 4
The key Federal science agencies, such as DoD, Total . . . . . . . . . . . . . . . . . . . . ... . . . . . . . 100
DOE, and NASA, have their own mechanisms SOURCE. National Techn!cal Information Servtce, 1988
110

to direct electronic dissemination of documents most NTIS users and client agencies believe in
and derived products (e.g., indices, abstracts, the NTIS concept and seek to find ways to
searches) are perceived by NTIS as meeting strengthen NTIS or at least the core NTIS func-
resistance from the information industry. The tions as a continuing element of the Federal Gov-
record of debate over NTIS privatization sug- ernment.
gests a basis for this concern, although the The record of the NTIS privatization debate
views of industry are varied and complex. As provides ample evidence of support for the
a matter of practice, NTIS depends on the pri- NTIS concept. For example, in response to an
vate sector for dissemination of online prod- April 1986 request for public comment, ’
ucts (such as the NTIS Bibliographic Database NTIS received 138 written responses from ex-
available via DIALOG, BRS, and the like). ecutive agencies, the legislative branch, the in-
NTIS estimates that private sector revenues formation industry, and individuals or orga-
derived from adding value to or re-marketing nizations that used NTIS. 2 Review of the
NTIS products totals about $11-12 million an- responses, by NTIS3 and OTA, indicates
nually. NTIS reliance on the private sector for
that:
electronic and/or enhanced dissemination has
had the perhaps unintended effect of discouragi- ● The NTIS user community overwhelrning
ng NTIS from aggressively pursuing how elec- opposed privatization, supported NTIS as
tronic technology might improve even the a government entity, and testified to the
NTIS core archival and clearinghouse func- importance of scientific and technical in-
tions. Improvements here could benefit both formation available from NTIS. The user
NTIS customers served directly by NTIS and community was heavily represented by li-
those served indirectly via private sector brary associations and individual univer-
vendors. sity, public, and technical libraries.
● The Federal agencies which supply the
There is also an equity issue involved con-
source documents to NTIS opposed
cerning access to the online NTIS Biblio- privatization, cited numerous problems
graphic Database and other online products. that could or would result if NTIS were
Some customers, and especially small busi- not a government entity, and emphasized
nesses, independent researchers, and the gen- their reliance on NTIS clearinghouse and
eral public, may not be able to afford the com- archival functions.
mercial rates (which can typically range from ● The information industry and individual
$50 per online hour and up). Since online company representatives overwhelming
searching of the NTIS database offers signifi- opposed privatization of the core NTIS
cant advantages, these customers could be dis- functions on the grounds that these func-
advantaged in terms of their ability to effec- tions were not appropriate for the private
tively access and retrieve Federally funded
sector and/or would create unfair competi-
scientific and technical reports.
tive conditions. However, industry and
In sum, NTIS presently operates in a con- company representatives strongly favored
strained environment, characterized by ambiva- privatization of various dissemination and
lent support from the government, limited finan- value-added functions.
cial resources (no public appropriation for the
‘U.S. Department of Commerce, “Study of Alternatives for
core clearinghouse and archival operations), Privatizing the National Technical Information Service”, No-
mixed support from the information industry, tice and request for public comment, Federal Register, vol. M,
limited technical resources, and a difficult prod- No, 81, Apr. 28, 1986, pp. 15868-15870.
2U s. Department
. of Commerce, NTIS Privatization StUdY Re-
uct mix (many low volume items). In addition, sponses to April 28, 1986 Federal Register Notice Request for
the basic demand for NTIS products appears to Pub]ic Comment, PB86-21 1240, National Technical Information
be significantly eroding, thereby placing in con- Service, Springfield, Virginia, June 1986.
3U s. ,Department of Commerce, National Technical Informa-
siderable jeopardy the overall viability of NTIS tion Service, “Analysis of Comments to Federal Register No-
as it is presently operating. At the same time, tice”, prepared by NTIS staff, 1986,
.

111

These general positions were reaffirmed at In a March 3, 1988 letter, the Chairman of
congressional hearings held in July 1987 and the Subcommittee on Science, Research, Tech-
February 1988 by the House Committee on Sci- nology, and Space sought the views of the Sec-
ence, Space, and Technology, Subcommittee retary of Commerce on legislation to establish
on Science, Research, and Technology. At the a National Technical Information Corp. The
latter hearing, the Subcommittee chairman re- Chairman advised that “[i]t is clear tome that
leased a letter from the Information Industry privatization of NTIS will not occur in the near
Association stating its position that “the Ad- future . . . To engage in a protracted NTIS
ministration’s proposal to privatize NTIS is privatization argument would be less than
not in the public interest in that it will ulti- productive. I would much prefer to look beyond
mately reduce the availability of Federally the privatization controversy to implement-
funded scientific and technical information. ing everyone’s underlying goal of transform-
ing NTIS into a modern, low-cost deliverer of
Overall, the years-long debate over privati-
scientific and technical documents.”~ Con-
zation of NTIS has further constrained the abil-
gressional and agency officials, as well as pub-
ity of NTIS to take initiatives, and has diverted
lic witnesses, have concluded that the drive
substantial NTIS and Department of Commerce
for NTIS privatization was not based on a bal-
resources (primarily staff time and attention).
anced analysis and finding of clear net bene-
Members of Congress and public witnesses have
fits, and furthermore that the Administration
criticized the Administration for prolonging the
did not have the capacity to successfully im-
debate when congressional sentiment against plement the NTIS privatization plans, even if
privatization is clear. In a February 23, 1988
thought to be desirable.~ (For further discus-
letter, the Chairman and Ranking Minority
sion, see chs. 11 and 12 on policy issues and
Members of the House Committee on Science, implications.)
Space, and Technology and Senate Commit-
tee on Commerce, Science, and Transportation
wrote the Secretary of Commerce and re-
quested delay in the privatization of NTIS
activities until Congress completed legislative 6Letter to Hon. C. William Verity, Secretary of Commerce,
from Hon. Doug Walgren, Chairman, Subcommittee on Scien(:e,
actions Research, and Technology, House Committee on Science, Space,
and Technology, Mar. 3, 1988.
~l~~tt~~ tO Hon. Doug wal~ren, Chairman, Subcommittee ‘n ‘See U.S. Congress, House Committee on Science, Space, and
Science, Research, and Technology, House Committee on Sci- Technology, Subcommittee on Science, Research, and Technol-
ence, Space, and Technology, from Kenneth E. Allen, Senior ogy, National Technical Information Serl’ice, Hearing, IC)Oth Con-
Vice President, Information Industry Association, Feb. 12, 1988. gress, 2nd Session, U.S. Government Printing Office, Washing-
5Letter to Hon. C. William Verity, Secretary of Commerce, ton, D. C., Feb. 24, 1988, Also see A.S. Levine, “Legal Financial
from Hon. Ernest F. Hollings, John C. Danforth, Robert A. Roe, Woes Hamper NTIS P]an ”, Federal Computer Week, May 2, 1988,
and Manuel Lujan, Jr., U.S. Congress, Feb. 23, 1988. PP. 15-16.

OPPORTUNITIES AND CHALLENGES


The major challenge facing NTIS is defin- major NTIS products and services. As shown
ing a viable role for its future. A variety of in- in Table 5-3, NTIS appears to have had a
dicators strongly suggest that the current role healthy total revenue and cost performance in
may not be sustainable absent some signifi- recent years, with net revenues realized in fis-
cant changes. cal years 1981, 1983, 1984, 1986, and 1987, and
net losses experienced in fiscal years 1982 and
Trends in Demand and Revenues 1985. For the entire eight year period, NTIS
The starting point for this analysis is the his- realized net revenues of $4.6 million or about
torical trend in demand for (and sales of) the 2.8 percent of total sales.
112

Table 5.3.—NTIS Revenues and Costs, Fiscal Years 1980.87

Total revenues Total costs Net revenues or loss


Fiscal year $ millions $ millions $ millions Percent
1980 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.6 17.8 0.8 4.3
1981 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.3 18.6 12.7
1982. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19.4 19.8 (:::)a (2.1)
1983 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.4 20.4 1.0 4.7
1984 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.7 20.4 1.5
1985. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.3 22.1 (N) (3.8)
1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22.4 21.6 0.8 3.6
1987 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22.3 22.1 0.2 0,9
Totals . . . . . . . . . . . . . . . . . . . . . . . . . . . 167.4 162.8 4.6 2.8
aParentheses Indicate net loss
SOURCE National Technical information Service, 1988

However, a detailed analysis by major NTIS An obvious question is how could NTIS main-
product line reveals a much different picture. tain a breakeven operation with slightly in-
Sales of all major NTIS products have declined creased revenues (in current dollars) over the
markedly since 1980, inmost cases by about fiscal year 1980-87 period, given the large re-
50 percent. For example, sales of paper copies duction in product sales? Part of the answer
dropped from752,000copiesin fiscal year1980 is that NTIS per unit prices increased signifi-
to 393,000 copies in fiscal year 1987. Sales of cantly over this same period of time, and with
microfiche copies dedined from 155,000 copies net price increases that typically equalled or
in fiscal year 1980 t0 67,000 in fiscal year1987. exceeded the rate of inflation. For example,
This pattern is repeated throughout the NTIS while paper copy sales decreased by about 50
product line, as shown in Table 5-4. percent, the average per unit price for paper
copies increased by 70 percent over the 1980-
Overall, sales of the abovesevenmajor NTIS 87 period while inflation averaged 45 percent.
products collectively declined from about3.69 Thus, in the case of paper copies, net revenues
million units (copies or subscriptions) in fiscal actually increased despite the drop in demand.
year 1980 t oabout 1.82 million units in fiscal This general pattern holds for all of the major
year 1987, a net declineof51 percent. Are- NTIS products, as illustrated in Table 5-6.
view of all other NTIS products indicated that
sales increased only for CUFT publications, In addition to maintaininggrevenues through
data tapes, and data diskettes, but these items
increased prices despite declining demand,
account for a small percentage of total NTIS
NTIS augmented sales revenues through:
sales. Trends in these three items and for soft-
ware tapes and catalogs (which declined) are s services to other agencies (such as order
shown in Table 5-5. billing and processing),

Table 5-4.—Demand for Selected Major NTIS Products, Fiscal Years 1980-87

Demand by fiscal year Net change


Product 1980 1981 1982 1983 1984 1985 1986 1987 Number Percent
On th~:;ands of copies)
Paper copy demand . . . . . . . 752 676 550 493 457 451 393 –359 –48
Microfiche copy demand 155 154 134 120 121 101 85 67 –88 –57
(in millions of copies)
Selected research in
microfiche (SRIM) . . . . . . . . 2.72 2.74 2.48 2.37 2,34 1.94 1.78 1.33 – 1.39 –51
(in thousands of subscriptions)
Government research
announcements and index 2.22 2.01 1.85 1.61 1.49 1.38 1.25 1,15 – 1,97 –48
Annual index ... . . . . . . . . . 0.91 0.84 0.96 0.82 0,73 0.63 0.61 0.50 –0.41 –41
Abstract newsletters . . . . . . 16.0 14.0 12.5 12,2 11,0 10.4 8.6 6.8 –9.2 –58
Published searches, ... .33.9 41.0 32.7 28.9 27.7 31.0 26.8 21.2 – 12.7 –38
SOURCE National Techn!cal Information Service, 1968.
113

Table 5-5.— Demand for Selected Minor NTIS Products, Fiscal Years 1983-87

Demand by fiscal year (number of copies) Net change


Product 1983 1984 1985 1986 1987 Number Percent
CUFT publications . . . . . . . . . . . . . . . . . . . . . . . . . . NA 4,227 5,412 6,577 5,552 + 1,325 +31
Software tapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 524 586 537 638 380 – 144 –28
Data tapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,405 1,783 2,174 2,493 2,503 +1,098 +78
Data diskettes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NA NA 100 179 338 +238 +238
Software catalog . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2,064 648 3,486 1,622 969 –1,095 –53
NOTE NA=not available
SOURCE National Technical Information Service, 1988

Table 5-6.—Average Per Unit Prices for Selected Major NTIS Products Compared to Inflation Rate,
Fiscal Years 1980.87

Fiscal year 1980 Fiscal year 1987 Net change Inflation


Product per unit price per unit price Dollars Percent rate
Average price per copy
Paper copies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $16.50 $27.87 $11.37 +690/o +450/0
Microfiche copies . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.50 6.50 3.00 +86 +45
Selected research in microfiche. . . . . . . . . . . . . . . 0,85 1.25 0.40 +47 +45
Average price per subscription
Government research announcements
and index . . . . . . . . . . . . . . . . . . . . . . . . . $275 $379 $104 +38V0 +450/0
Annual index ..., . . . . . . . . . . . . . . . . . . . . . . . . . . 375 479 104 +28 +45
Abstract newsletters . . . . . . . . . . . . . . . . . . . . . . . . 62 89 27 +44 +45
Published searches . . . . . . . . . . . . . . . . . . . . . . . . . 30 45 15 +50 +45
SOURCE National Technical Information Service, 1988

● sales ofcomputer-relatedproducts (which The comparison between fiscal year 1986 and
haveahigh averageperunit price, about~ fiscal year 1987 revenue data shows how NTIS
$67 per unit comparedto $28 for paper has offset revenue decreases in full text reports
and$6.50 for microfiche, asoffiscalyear and subscription, bibliographic, and announce-
1987~ and ment products with revenue increases in com-
● NTIS brokerage fees on sales of other puter products and services to other agencies.
agency materials.
Whether and how long NTIS can be expected
Infiscalyear 1987 these three itemstogether to remain viable operating on this basis re-
accounted foroveraquarter oftotalNTlSrev- quires examination. One risk is that continued
enues. The fiscal year 1986 and fiscal year 1987 reductions in the sales volume of reports, sub-
revenue breakouts are shown in Table 5-7. scriptions, bibliographies, and the like could

Table 5-7.—NTIS Revenues by Product Group, Fiscal Years 1986-87


(in thousands of dollars)

Fiscal year 1986 Fiscal year 1987 Net change


Product group Dollars Percent Dollars Percent Dollars Percent
Full text reports . . . . . . . . . . . . . . . . . . . . . . . 11,195 50.0 10,403 46.6 – 792 –7.1
Subscription, bibliographic, and
announcement products . . . . . . . . . . . . . . 6,100 27.3 5,429 24.3 –671 – 11.0
Computer products . . . . . . . . . . . . . . . . . . . . 1,416 6.3 2,167 9.7 + 751 +53.0
Services to other agencies . . . . . . . . . . . . . . 1,811 8.1 2,451 11.0 +640 +35.0
Patent licensing fees. . . . . . . . . . . . . . . . . . . 617 2.8 575 2.6 –42 –6.8
Brokerage fees . . . . . . . . . . . . . . . . . . . . . . . . 1,095 4.9 1,220 5.5 + 125 + 11.4
Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 0.7 102 0.5 –45 –30.6
Totals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22,381 100.1 22.347 100.2
SOURCE National Technical Information Service, 1988
114

necessitate further price increases, which by NTIS and the library and information sci-
could, in turn, further reduce sales, and so on. ence community.
The elasticity of demand for NTIS products
Other contributing factors include declining
is not precisely known. NTIS staff believe that
agency participation in the NTIS program and
increasing price is one significant factor con- limited customer awareness of NTIS products.
tributing to the decline in sales. For example,
With respect to the former, NTIS reports that
various library officials observe that rising
the number of new titles provided to NTIS by
NTIS prices have been a major factor contrib-
Federal agencies has declined by about 20 per-
uting to reductions in NTIS subscriptions,
cent over the fiscal year 1983-87 period, as
especially as libraries are faced with increas-
shown in Table 5-9. Assuming that NTIS was
ingly tight budgets. This view is supported to not receiving one-third of relevant agency ma-
some extent by results of the GAO survey of
terials in fiscal year 1983 (NTIS estimate), the
Federal agencies. Of the 114 civilian depart-
fugitive document percentage would now be
mental components responding, 40 agency
up to about 47 percent. In effect, this trend
components evaluated the cost of NTIS may compromise both the perceived and real
reports in paper format and 27 evaluated the
utility of the NTIS archive. NTIS may now
cost of microfiche format. Two-thirds of the be receiving only about one-half of relevant
agencies evaluated the costs for paper as high
agency documents.
or very high, while two-thirds evaluated mi-
crofiche cost as moderate or low, as indicated This conclusion is qualitatively consistent
in Table 5-8. with the results of the GAO survey of Federal
agencies. Of the 72 civilian departmental
NTIS staff believe that online searching of
agency components disseminating scientific
the NTIS bibliographic database may also be and technical information, only onehalf of the
contributing to a reduction in the number of agencies responding use NTIS. Agencies ap-
requests for reports. The NTIS index products pear to rely primarily on themselves for dis-
offered directly in paper or microfiche form and semination, secondarily on GPO, NTIS, and
in electronic form via private sector vendors the Depository Library Program (DLP), and
may well be improving the efficiency of cus- to an even lesser extent on the private sector.
tomer searches of the NTIS archives, while at The results are presented in Table 5-10.
the same time may be undercutting sales of
NTIS documents. The effects of online search- With respect to customer awareness of NTIS,
ing on overall NTIS demand are debatable. The NTIS has an ongoing series of activities to in-
experience with other online bibliographic form potential customers of NTIS services.
databases has tended to be just the opposite; However, the results of the GAO survey of
online searching has facilitated more aware-
ness of and requests for the referenced docu-
ments. This subject warrants further research Table 5-9.–Trend in New Titles Received by NTIS,
Fiscal Years 1983.87

Titles received Estimated percent


Table 5-8.—Federal Agency Evaluation of the Cost
Percent of all relevant
of NTIS Reports, Paper and Microfiche Formats, Fiscal year Number change agency titlesa
114 Agency Components Responding
1983 . . . . . . . . . . . . 79,471 67.0
1984 . . . . . . . . . . . . 71,587 60.4
Percent of agencies using NTIS
1985 . . . . . . . . . . . . 70,211 59.3
Cost of NTIS report Paper Microfiche 1986 . . . . . . . . . . . . 69,760 58.9
Very high . . . . . . . . . . . . . 22.5 1987 . . . . . . . . . . . . 62,856 53.0
High . . . . . . . . . . . . . . . . . 45.0 2;:; Net change
Moderate . . . . . . . . . . . . . 27.5 63.0 (1983-87) . . . . . . –16,615 –2o.9 – 14.0
Low . . . . . . . . . . . . . . . . . 2.5 7.4 a&SUmOS number Of relevm! agency titles remains constantat 11 g,000 PN year
Very low . . . . . . . . . . . . . 2.5 —
SOURCE National Technical Information Service and Office of Technology As-
SOURCE: GAO Survey of Federal Agencies, 1987, sessment, 1988.
.—.— ——. —.

115

Table 5-10.—Federal Civilian Agency Dissemination Table 5.11 .—Scientific and Technical Association Use
of Scientific and Technical Information of Federal Information Sources, Rank Order

Percent of Percent of associations


Dissemination channel agencies a responding that the
Own agency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 931 Source of Federal information source is usedb
Government Printing Office . . . . . . . . . . . . . . . . 52.7 1. Trade, professional, or
National Technical Information Service. . . . . 50.0 scientific journals. . . . . . . . . . . . . . 88.9
Depository libraries . . . . . . . . . . . . . . . . . . . . . . . 50.0 2. Newsletters. . . . . . . . . . . . . . . . . . . 82,9
Private sector vendors/contractors . . . . . . . . . . 36.1 3. Newspapers . . . . . . . . . . . . . . . . . . 71.9
Consumer Information Center . . . . . . . . . . . . . 4,2 4. News magazines , . . . . . . . . . . . . . 68.9
a&ult~ expressed as percentage of agencies thatdisseminate STI that use each 5. Individual Federal Agencies . . . . . 64.1
channel 6. Radio/television . . . . . . . . . . . . . . . 53.8
SOURCE GAO Survey of Federal Agencies, 1987 and Off Ice of Technology As- 7, Congressional agency (LOC I
sessment, 1988 GAO, OTA, CBO, CRS) . . . . . . . . . 50.8
8. College/university library . . . . . . . 45.8
9. GPO mail/telephone orders . . . . . 44.4
10. Office of U.S. Senator or
Federal information users suggest that, over- Representative . . . . . . . . . . . . . . . 43.2
11, State or local government
all, NTIS plays a rather limited role relative agency . . . . . . . . . . . . . . . . . . . . . . . 42.6
to other direct and indirect sources of Federal 12. Inhouse library or information
information. Among other groups, GAO sur- center. . . . . . . . . . . . . . . . . . . . . . 39.5
13, Commercial bookstore . . . . . . . . 38.7
veyed a random sample of scientific and tech- 14. Congressional committee. . . . . . 37.7
nical associations. Based on the responses of 15. Local public library . . . . . . . . . . . . 37.5
133 associations (out of 250 sampled), NTIS 16. Commercial mail/telephone
orders. . . . . . . . . . . . . . . . . . . . . . . . 35.8
is used by about one-third. Individual Federal 17. NTIS . . . . . . . . . . . . . . . . . . . . . . . . . 30,6
agencies are used occasionally to very often, 18. Federal agency library. . . . . . . . . . 25.2
as are newspapers, news magazines, newslet- 19, Commercial online database
vendor . . . . . . . . . . . . . . . . . . . . 24.8
ters, and trade, professional, and scientific 20, GPO bookstore . . . . . . . . . . . . . . . 23.7
journals. Compared to other governmentwide 21, GPO depository library . . . . . . . . 19.8
information dissemination mechanisms, NTIS 22. Commercial information
brokers . . . . . . . . . . . . . . . . . . . . . . . 15.4
is used more often than the Consumer Infor- 23. Consumer Information Center . . 12.1
mation Center (CIC) or DLP, but less often 24. State agency library. . . . . . . . . . . . 11.2
then GPO mail/telephone orders. While asso- 25, State government library. . . . . . . . 7.8
26, Local school (Grades 1-12)
ciations are perhaps not the best indicator of library . . . . . . J . . . . . . . . . .‘. . . . . . . 1.7

NTIS customer awareness, science, environ- aFederal sources are italicized
bAgencies repo~ing that a source of Federal !nformat!on IS used Occasionally,
ment, and technology were the most frequently often, or very often.
cited categories of Federal information used SOURCE: GAO Survey of Federal Information Users 1988
by the respondents. The relative ranking of
Federal information sources for these associa-
tions is shown in Table 5-11.
documents is appreciated, especially in the li-
The relatively low use of NTIS may reflect brary community and among university re-
a combination of low demand for NTIS prod- searchers. a
ucts, low awareness of NTIS products, and/or,
as mentioned earlier, high cost of NTIS prod-
“Mark P. Haselkorn, Philip L. Bereano, Caro]yn Plumb, and
ucts. In addition, OTA’S independent consul- Patricia Tetlin, “Perspectives on Federal Dissemination of Sci-
tant on university use of scientific and techni- entific and Technical Information”, OTA cent ractor report pre-
cal information concluded that NTIS is not pared by the program in Scientific and Technical Communica.
tion, School of Engineering, University of Washington, Seattle,
viewed as a source of state-of-the-art informa- February 1988, Also see Charles R, McClure, Peter Hernon, and
tion, due to the time delays between the exis- Carj R. Purcell, Linking the U.S. IVational Technical Informa-
tence of a document and its availability via tion Sert’ice With Academic and Public Libraries, (Norwood,
NJ: Ablex Publishing, 1986); and Peter Hernon and Charles R.
NTIS. On the other hand, the role of NTIS as McClure, Federal Information Policies in the 1980s: Conflicts
a secondary source of scientific and technical and Issues, INorwood, NJ, Ablex Publishing, 1987).
116

Possible New Initiatives Table 5-13.—Age of NTIS Documents Sold


in Calender Year 1986
The nature of demand for NTIS documents
Percent of
makes NTIS highly suitable for application of Date of document Co~ies sold total
electronic publishing and printing-on-demand 1968 and prior. . . . . . . . . . . . . . . . . . 7,730 1.9
systems. As noted earlier, the average demand 1969 . . . . . . . . . . . . . . . . . . . . . . . . . . 1,412 0.3
for NTIS documents is 10 copies, and perhaps 1970 . . . . . . . . . . . . . . . . . . . . . . . . . . 1,931 0.5
one-quarter of the documents never sell a sin- 1971 . . . . . . . . . . . . . . . . . . . . . . . . . . 2,489 0.6
1972 . . . . . . . . . . . . . . . . . . . . . . . . . . 3,744 0.9
gle copy. As an illustration, for all documents 1973 . . . . . . . . . . . . . . . . . . . . . . . . . . 4,346 1.1
archived by NTIS in calendar year 1986, there 1974 . . . . . . . . . . . . . . . . . . . . . . . . . . 4,278 1.0
1975 . . . . . . . . . . . . . . . . . . . . . . . . . . 4,942 1.2
was no demand for 43 percent, only 5 percent 1976 . . . . . . . . . . . . . . . . . . . . . . . . . . 7,282 1.8
sold more than 10 copies, and only 1 percent 1977 . . . . . . . . . . . . . . . . . . . . . . . . . . 8,117 2.0
sold more than 50 copies. The detailed demand 1978 . . . . . . . . . . . . . . . . . . . . . . . . . . 8,660 2.1
distribution is shown in Table 5-12. 1979 . . . . . . . . . . . . . . . . . . . . . . . . . . 9,571 2.3
1980 . . . . . . . . . . . . . . . . . . . . . . . . . . 10,446 2.5
In addition to very low total demand for 1981 . . . . . . . . . . . . . . . . . . . . . . . . . . 14,799 3.6
1982 . . . . . . . . . . . . . . . . . . . . . . . . . . 15,615 3.8
most NTIS documents, demand for a given 1983 . . . . . . . . . . . . . . . . . . . . . . . . . . 24,318 5.9
document can be spread over many years. For 1984 . . . . . . . . . . . . . . . . . . . . . . . . . . 30,448 7.4
example, of the average sales of 10 copies per 1985 . . . . . . . . . . . . . . . . . . . . . . . . . . 72,149 17.6
1986 . . . . . . . . . . . . . . . . . . . . . . . . . . 138,431 33.7
document, only 3 copies might be sold in the Pre-announcement demanda . . . . . 2,854 0.7
first year after announcement, 2 copies in the Announcement date unknown b . . . 37,423 9! 1
second year, 4 copies spread over the third Total . . . . . . . . . . . . . . . . . . . . ., . . 410,985 100.0
through tenth years after announcement, and aPre.announcement demand Announcement date of 1987, orders were received
the last copy might be sold 11 to 15 years or in FY 1986 due to source pre-announcement, etc , and NTIS was able to fill the

more after being made available. This phenom- b~~~~~~cement Date IJnknown Announcement date IS not included In NTIS in.
ventory file
enon is known as the demand decay curve, and
SOURCE National Technical Information Service, 1988
is illustrated in Table 5-13 for NTIS documents
sold during calender year 1986.
This highlights the NTIS dilemma: low sales
volume spread over many years, but a document tribute to a major scientific or technical break-
that sells only a single copy could contribute to through. This is the primary rationale for the
significant innovations. While NTIS is able to NTIS archive, and underpins the need to main-
identify documents that are relevant to current tain NTIS documents on file indefinitely.
technical issues and research and development Fortunately, technological advances have
priorities, it is difficult to predict which docu- created several possibilities that appear to be
ments will have high demand and virtually ideally suited to the nature of NTIS demand.
impossible to predict which documents will con- First, most federally-sponsored or conducted
scientific and technical reports are created on
Table 5.12.—Demand for NTIS Documents word processing or microcomputer-based sys-
Announced in Calendar Year 1986
tems. Thus, the keystrokes are captured elec-
Number of Percent of total tronically. The electronic versions of these
Level of demand documents documents reports are typically converted to paper (or mi-
(total annual) (per demand level) announced
crofiche) format by the originating agency (or
No demand. . . . . . 28,364 43
1 copy . . . . . . . . . . 10,906 17
the agency contractor) and submitted to NTIS
2-5 copies . . . . . . . 16,853 26 as paper (or microfiche) copies. NTIS then dis-
6-10 . . . . . . . . . . . . 5,597 9 seminates copies in paper (or microfiche) in re-
11-20 . . . . . . . . . . . 2,228 3 sponse to requests. About 80 percent of the
21-50 . . . . . . . . . . . 967 1
51 + . . . . . . . . . . . . 379 1 NTIS reports are disseminated in paper for-
Totals . . . . . . . . 65,294 100 mat and the remaining 20 percent in micro-
SOURCE National Technical Information Service, 1986 fiche. Because demand is low, and typically for
117
——...—— .—

one copy at a time, the per unit costs are high production of full copies, and facilitates
—averaging $25-30 per paper copy. printing-on-demand of selected pages, since the
text of documents stored on the system could
If NTIS could receive agency reports in elec- be made available for electronic display at re-
tronic format, using compatible document de- mote terminals. DTIC plans to use a standard
scription standards, then NTIS could apply structured database approach known as Stand-
electronic publishing and printing-on-demand ardized General Markup Language (SGML) to
technology where appropriate. For example, facilitate electronic document reproduction on
NTIS could develop and implement a version
a fast turnaround basis.
of the Electronic Document System similar to
that being developed by DTIC. Documents DTIC intends to continue its present micro-
could enter NTIS either indirect electronic for- fiche-based production system in parallel with
mat (by magnetic computer tape, floppy disk, the Electronic Document System. However,
or electronic transmission, similar to GPO re- as new documents are added in electronic for-
ceipt of publications material) or by scanning mat, the use of microfiche is expected to de-
paper copies. DTIC envisions the use of high- cline substantially. DTIC estimates that the
speed, high-resolution optical scanning equip- electronic system will fill about 40 to 45 per-
ment that will compete favorably with the cur- cent of paper copy requests after one year of
rent microfiche system used by DTIC. How- operation, and about 60 to 70 percent of such
ever, direct electronic input should be less requests after 3 years of operation. Should
expensive than scanning for new input, at least NTIS implement a similar system, NTIS could
for NTIS purposes. Scanning may, however, expect comparable results, with an estimated
be the only option for converting old material. 35 percent of requests handled with electronic
DTIC also envisions using high capacity, low printing on demand after the first year, 60 per-
cost per bit digital storage systems such as cent of requests after 3 years, and perhaps 75
those using laser optical disks. High capacity percent of requests after 5 years. These esti-
WORM (Write Once, Read Manytimes) opti- mates assume that the NTIS backfiles (ar-
cal disk juke boxes should be applicable to both chival documents entered in earlier years)
DTIC and NTIS. The 12-inch disks can store would be retained in microfiche, since demand
about one gigabyte of information per side, or is so low and the cost of conversion may not
roughly 400,000 to 500,000 pages of double- be justified.
spaced typed material per side. This means
that, assuming straight digitized text only and Should NTIS implement its version of an
an average length of 200 double-spaced pages Electronic Document System, NTIS would be
per document, the roughly 70,000 new docu- positioned to offer search and retrieval capa-
ments received by NTIS each year could be bility, directly and/or via private vendors or
stored on about 14 doublesided WORM opti- other government agencies. Such software
cal disks.~j could build on the results, as they become avail-
able, of DTIC’S Artificial Intelligence/Decision
For output, DTIC plans to use high-speed, Support Laboratory. Another prototype is
high resolution electronic printing equipment “Grateful Meal, ’ bibliographic search software
for producing paper documents on demand. developed by the National Library of Medicine
Documents will be printed double-sided using for users of the MEDLARS database.
plain bond paper to reduce paper and mailing
costs. The reproduction cost should drop by An NTIS Electronic Document System, if
an order of magnitude. The fully developed properly interfaced with the source agencies,
Electronic Document System permits the re- should be able to substantially reduce the time
lag between the existence of a document and
ITOK d [)(:11 In ~!Ilt ~ i 200 j)agf;s p e r docunlellt = 1 4 million
its availability via NTIS. The time lag could
pages divided by 1 million pages perdoubl~sided 12-inch 11’ORM be further reduced if interagency procedures
disk. are strengthened so that agencies are required

-
.. “

- >. — , ‘
to make more timely submissions to NTIS. associations. Demand for paper and microfiche
Strengthening of such procedures could also is anticipated to decrease moderately over the
address the question of how to increase the next three years.
completeness of the NTIS archive. As noted
Overall, an electronic NTIS should be able
earlier, perhaps one half of the scientific and to greatly increase the diversity and timeliness
technical reports generated by or for Federal of NTIS (and related private vendor) offerings,
agencies are not submitted to NTIS. While all increase the ability of NTIS (and private ven-
the fugitive documents may not be relevant, dors) to match information products with po-
it is likely that a significant portion of these tential users, and reduce the cost of NTIS (and
are. Agencies could be required to provide more
private vendor) products. An electronic NTIS
complete submissions and/or follow a specified also should be better able to serve, especially,
set of procedures for determining what items small and medium businesses and individual
should be submitted. researchers who tend to be penalized by the
An electronic NTIS with a more complete present paper-based system that assigns a
and up-to-date archive would open up a wide premium to economies of scale. Understand-
range of possibilities for marketing and dis- ably, NTIS directs much of its marketing ef-
semination of scientific and technical informa- forts at its largest customers who generate the
tion. The following illustrative activities could most sales, but who also are typically well
be implemented by NTIS, individual Federal staffed with information specialists. An elec-
science agencies, and/or private vendor: tronic NTIS would increase the incentives and
available options to reach smaller market seg-
● CD-ROM distribution of NTIS biblio- ments from whence many innovations ulti-
graphic database on selected subjects;
mately originate. It is certainly conceivable
● CD-ROM distribution of NTIS documents
that NTIS could eventually be used by the in-
on selected subjects; dividual researcher and entrepreneur who de-
● Floppy disk distribution of individual
pend heavily on informal and collegial net-
NTIS reports; works for the sharing of scientific and technical
● Online distribution of selected NTIS doc- information. Also, NTIS would be a logical key
uments with printing-on-demand of the en- participant in the development of a govern-
tire document or selected pages at NTIS mentwide information index, for which re-
or remote locations; spondents to the GAO survey of Federal in-
● Electronic bulletin board announcement
formation users indicated strong interest. Such
of selected new NTIS documents of gen-
an index would also help improve the ability
eral interest; and
of researchers and entrepreneurs to know of
● Electronic bulletin board announcement
potentially relevant information. Finally, as
of NTIS documents on subject matter a complement to the electronic document sys-
matched to the bulletin board partic- tem and improved indexing, increased agency
ipants.
participation in the NTIS clearinghouse may
In essence, an electronic NTIS would have need to be mandated. The declining trend in
the capability to produce multi-format output the percentage of agency scientific and tech-
—paper, microfiche, offline electronic, or on- nical documents submitted is cause for con-
line electronic as appropriate, depending on the cern. While including 100 percent of agency
type of product and user needs. As noted documents in NTIS is unrealistic, some steps
earlier, the GAO survey results highlighted the could be taken to broaden the coverage and
significant anticipated increase over the next increase the timeliness of agency submissions.
three years in demand for scientific and tech- This could be accomplished through inter-
nical information in electronic formats on the agency agreements, OMB circulars, and./or, if
part of libraries and scientific and technical necessary, legislation.
119

NTIS/SUPDOCS COOPERATION
The consolidation of NTIS with GPO’s Sup- while in stock or if reprinted (usually due
Docs has been proposed by the Public Printer to strong demand);
and as part of legislation introduced in the past NTIS has 2 million document titles for
two Congresses that would establish a Gov- sale whereas the average SupDocs sales
ernment Information Office. However, this inventory is about 20,000 or about one per-
section focuses on opportunities for improved cent of the NTIS inventory;
cooperation between NTIS and SupDocs, ir- the average NTIS sales volume is about
respective of the formal institutional structure, 10 copies per title whereas the SupDocS
since the need for improvements in NTIS/Sup- average is on the order of 2,000 copies per
Docs marketing, product line analyses, and co- title;
ordination will exist regardless of the institu- NTIS retains all titles received in the
tional structure. (See chs. 11 and 12 for further NTIS archive and available for sale, while
discussion of institutional alternatives.) SUpDocS for the most part includes only
the titles judged to have significant sales
The major reasons advanced for improved potential;
NTIS/SupDocs cooperation (whether or not the NTIS annual sales volume is in the
through formal consolidation) are: efficiencies few millions range whereas the SupIXxx
in management and operations, improved co- volume is in the few tens of millions range;
ordination of Federal information dissemina- and,
tion, enhanced opportunities for use of new NTIS is considerably smaller than Sup-
technology, strengthened joint marketing pro- Docs–at yearend fiscal year 1987, NTIS
grams, reduced overlap and duplication in had 344 employees compared to 930 for
government dissemination activities, and im- SupDocs, NTIS had total revenues of
proved overall public access to Federal infor- about $22 million in fiscal year 1987 com-
mation. Possible drawbacks of or barriers to pared to about $100 million for SupDocs
improved cooperation include: some differ- (figures include reimbursed services and
ences in current missions of the NTIS and Sup- services funded through appropriations).
Docs and resultant potential problems in more
closely coordinating these functions, difficul- At first glance, these differences could ap-
ties inherent in cooperative activities of agen- pear as, collectively, a significant barrier to im-
cies from different branches of government, proved cooperation. However these differences
and reluctance on the part of some Federal could become complementary aspects of a com-
agencies to cooperate with NTIS and/or Sup- bined strateg~ for institutional sur~~iva] and
Docs, regardless of the institutional structure. growth.

There are significant similarities between


Differences and Similarities NTIS and SupDocs:
● Both must operate their sales programs
The major differences between NTIS and
on a breakeven basis, that is, there are no
SupDocs are that:
appropriations to subsidize the cost of
● NTIS is in the executive branch while Sup- sales.
Docs is in the legislative branch; ● Both must compete with private vendors,
● NTIS maintains a permanent archive of who can always reprint and resell govern-
scientific and technical documents totall- ment documents since these materials
ing close to 2 million items, while GPO cannot be copyrighted.
maintains documents in inventory only ● Both must compete to some extent with
120

Federal mission agencies, who frequently for licensing, and foreign government reports
distribute significant numbers of copies exchanged with Federal agencies and any
of documents free of charge to agency federally-generated translations thereof. The
clients, contractors, and interest groups. NTIS database is updated biweekly and, is
available online through commercial vendors.
Both NTIS and SupDocs carry out market-
ing activities in support of their sales pro- SupDocs prepares:
grams, although the programs have relative
strengths and weaknesses: the Monthly Catalog of United States
Government Publications (which indexes
● NTIS produces a variety of specialized publications by author, title, subject, ser-
subject matter searches that have no di- ies/report number, contract number, stock
rect parallel at SupDocs. number, and title keyword);
● SupDocs makes growing use of radio and the 3 times a year Consumer Information
television public service announcements Catalog (which lists consumer publica-
and is revitalizing the GPO bookstores as tions from about 30 Federal agencies that
sales outlets, marketing tools not used by are available free or at minimal charge
NTIS. from CIC);
the quarterly Government Peri&”cals and
NTIS and SupDocs perform reimbursable
Subscription Serw”ces (which lists over 500
services for other agencies: subscriptions to periodicals and recurring
● In fiscal year 1987, NTIS performed about reports published by more than 40 Fed-
$2.5 million worth of services for other eral agencies and sold by SupDocs);
agencies (accounting for roughly 10 per- the 3 times a year U.S. Government Books
cent of total revenues). (which catalogs about 1,000 of SupDocs
● In fiscal year 1987, SupDocs performed best-selling publications); and
about $5 million in reimbursable services, the bimonthly New Books (which lists new
primarily for operating the CIC for GSA SupDocs wd-es items).
(accounting for about 5 percent of total
SupDocs revenues). Information on SupDocs sales items, bibliog-
e If the DLP, also operated by SupDocs, is raphies, and catalogs is available from private
counted as a reimbursable service funded vendors, in both online and CD-ROM formats.
through appropriations, then reimburs- For example, the GPO Sales Publications
able services would be about 25 percent Reference File, which lists all GPO titles cur-
of total SupDocs revenues. rently for sale, is available online to the public
via the commercial DIALOG information re-
The similarities go on. Both NTIS and Sup- trieval service and includes an online ordering
Docs prepare indices or catalogs to govern- capability.
ment documents. NTIS publishes a weekly and
annual Government Reports Announcement Both NTIS and SupDocs primarily use pa-
and Index Journal (known as GRA&I) that in- per and microfiche formats for dissemination,
cludes summaries of government conducted or although NTIS sales of computer tapes, floppy
sponsored research reports. The summaries are disks, and software have been growing, as have
indexed by subject, author, institution, and GPO sales of computer tapes. Both NTIS and
contract number (if applicable). NTIS also pre- SupDocs have international exchange pro-
pares the NTIS Bibliographic Database that grams to encourage the two-way flow of infor-
includes all items in the NTIS archive. In addi- mation between the U.S. and other countries.
tion to government conducted or sponsored Finally, it bears emphasis that, except for the
reports, the NTIS Database includes federally- type of bibliographic and index products men-
generated machine readable data files and soft- tioned earlier, both NTIS and SupDocs depend
ware, U.S. Government inventions available on the Federal mission agencies as the primary
121

source of documents. The agencies (including thought to be small; the overlap between the
Congress for SupDocs purposes) create the doc- NTIS archive and the regional library archives
uments and in many cases handle primary dis- is unknown (as the libraries do receive some
tribution; NTIS and SupDocs as government- NTIS publications).
wide information dissemination mechanisms
Another possible disadvantage is aggrava-
are responsible for, in effect, secondary distri- tion of separation of powers concerns, since
bution through their sales programs (and NTIS is in the executive branch and SupDocs
through the DLP and CIC in the case of Sup-
Docs)~Private vendors also serve as second- in the legislative. Some Federal executive
branch agencies do not like the current roles
ary distributors of selected agency documents.
of GPO and the Joint Committee on Printing
Disadvantages and Advantages (JCP) (as authorized by Title 44 of the U.S.
Code) with respect to agency printing and pub-
The possible disadvantages of improved lishing activities, view those roles as inap-
NTIS/SupDocs cooperation are erosion of the propriate and/or unconstitutional (see ch. 11),
NTIS archive function and aggravation of sep- and oppose any greater role for them. Regard-
aration of powers concerns. Representatives less of the merits or demerits of these concerns,
of the scientific and technical community as the role of SupDocs has not been the primary
well as the Federal science agencies believe that focus of attention or challenge. In fact, some
the NTIS archive or something equivalent is NTIS officials believe that NTIS is handi-
essential to the U.S. research and development capped because, while Federal agencies are
effort and to basic science and technical inno- required to participate in the SupDocs sales
vation. From this perspective, cooperative ini- program, agency participation in the NTIS ar-
tiatives would have to be designed so as to en- chive is voluntary and not required by stat-
sure continuity of the archive. If the DLP is ute. SupDocs seems to be able to work effec-
viewed as part of SupDocs, then SupDocs does tively with many executive branch agencies,
already have an archive function, since the re- even though SupDocs is in the legislative
gional depository libraries maintain a complete branch. Nonetheless, improved NTIS/SupDocs
archive of all government publications distrib- cooperation and especially a consolidation is
uted to them, 60 percent of which in recent viewed by some Commerce Department and
years are in microfiche format. Also, either mi- OMB officials as possibly aggravating con-
crofiche masters or camera ready copy existed flicts over separation of powers, but more im-
at some previous point in time for most of these portantly, from their perspective, further dis-
materials. However, retention of these origi- tancing the creators of the information (the
nals is incomplete, and neither the originals executive agencies) from the disseminators. In
nor the regional depository library archives are this view, the decentralizing tendencies of elec-
available as part of a coordinated sales pro- tronic technologies should be encouraged by
gram. The SupDoc’s Library Programs Serv- placing information dissemination as close as
ice does maintain a collection of microfiche possible to the ultimate users of the informa-
masters procured for the DLP, and plans to tion. The strengthening of centralized dissem-
eventually transfer this collection to the Na- ination mechanisms (whether SupDocs, NTIS,
tional Archives and Records Administration or even governmentwide indices) seems to be
(NARA). Copies of some of these items are feared and resisted, even if centralized dissem-
available for sale on demand. It also should ination would not preempt agency dissemi-
be noted that NTIS has submitted to NARA nation.
a plan that provides for transferring NTIS
It seems plausible that strengthened NTIS-
master microforms to NARA when records are
10 years old. SupDocs cooperation would lead to improve-
ments in indexing, marketing, and interna-
The actual current overlap between the NTIS tional exchange. Perhaps most important, how-
archive and the SupDocs sales program is ever, is the potential improvement in overall
722

strategic posture that could result from im- users to switch to electronic formats, which
proved cooperation. As presented earlier in de- in turn could lead to yet another price increase
tail, NTIS is in a very vulnerable situation. In for paper copies to cover fixed costs with a
contrast, SupDocs has maintained better than smaller sales volume. According to GPO, un-
breakeven operations in recent years, with net der current law, if the Department of Com-
income of $11.4 million in fiscal year 1987 and merce stopped printing the CBD, there would
$5.5 million in fiscal year 1986. However, while be no printing requisition for SupDocs to
in a strong position compared to NTIS, Sup- “ride” (order extra copies) and thus no “addi-
Docs has some emerging areas of vulnerabil- tional copies” for SupDocs to sell.
ity that could become significant in a rather
Other SupDocs best sellers that might be vul-
short period of time.
nerable include (with fiscal year 1987 revenues
Like NTIS, SupDocs could be vulnerable to indicated): the Code of Federal Regulations ($2
electronic competition. For example, one of million), Federal Acquisition Regulations ($1.9
SupDocs largest revenue sources is the Com- million), Tariff Schedules Annotated ($0.9 mil-
merce Business Daiiy (CBD), with subscrip- lion), and DoD FAR Supplement ($0.9 million).
tions generating more than $9 million in Sup-
At the moment, SupDocs sales volume and
Docs revenue, or about 12 percent of total sales
total distribution appear to be holding reason-
revenue in fiscal year 1987. However, the re-
ably steady. Most indicators declined in the
sults of private sector marketing of the CBD
early 1980s, but have since been relatively
online or on CD-ROM suggests that electronic
level. Trends for fiscal years 1981-87 are shown
formats may be preferable for many C13D cus-
in Table 5-14 for SupDocs sales orders, copies
tomers. If the demand for paper copies declined
sold, CIC free orders, CIC copies distributed,
dramatically over the next few years, it is con-
and depository library copies distributed.
ceivable that the Department of Commerce
might stop funding the set-up charges for A detailed analysis of the SupDocs product
printing paper copies. While SupDocs presum- line is warranted to determine if significant vul-
ably could continue to print the CBD itself, nerability extends beyond items such as the
the cost would increase significantly, since CBD and, as discussed in chapter 4, the Rec-
SupDocs now pays only the marginal print- ord and l?e~”ster, that are well suited to elec-
ing cost, but would have to pay the full print- tronic formats. Overall, SupDocs would appear
ing cost if the Department of Commerce ceased to be in a stronger position than NTIS, since
participation. This could put SupDocs in the many of the traditional government reports
position of raising prices for paper copies of and periodicals sold or distributed by SupDocs
a product (the C13D) that clearly is well suited are likely to be best suited to paper formats
to electronic formats, especially online. If the for years to come. Also, SupDocs has poten-
NTIS experience is any guide, higher prices tial opportunities in other areas, such as sales
could further reduce sales and encourage more of government forms. For example, in fiscal

Table 5-14.—SupDocs Sales and Distribution Activity, Fiscal Years 1981-87

Millions of orders or copies


Fiscal year Fiscal year Fiscal year Fiscal year Fiscal year Fiscal year Fiscal year
1981 1982 1983 1984 1985 1986 1987a
Sales ordersb . . . . . . . . . . . . . . . . . . . . 2.0 1.5 1.3 1.5 1.5 1.5 1.6
Free CIC orders . . . . . . . . . . . . . . . . . . 3.9 2.4 2.3 2.2 2.1 2.6 2.2
b
Copies sold . . . . . . . . . . . . . . . . . . . . . 29.8 25.9 24.5 24.8 26.7 27.1 26.7
Free CIC copies distributed . . . . . . . NA 25.7 23.0 14.7 21.9 19.2 21.5
Depository library copies
distributed . . . . . . . . . . . . . . . . . . . . 28.7 20.7 31.9 37.1 36,1 26.7 22.7
aEstlmates.
blncludes CIC sales
SOURCE U.S Government Printing Off Ice, 1988.
123

year 1987, SupDocs sold IRS forms to tax prac- Table 5-15.—Conventional Printing Functions Affected
titioners. About 80,000 orders were processed, by Length of Press Run
yielding a gross revenue of $2.8 million and
net revenue of $1.5 million. Nonetheless, given Affected by length
the strong commitment of many Federal agen- Function of press run
cies to shift to electronic formats over the next Composition . . . . . . . . . . . . . . . . . . . . . . . - No --
Camera. . . . . . . . . . . . . . . . . . . . . . . . . . . . No
few years, especially for statistical, scientific Platemaking . . . . . . . . . . . . . . . . . . . . . . No
and technical, and administrative documents, Film . . . . . . . . . . . . . . . . . . . . . . . . . . . . . No
the SupDocs sales and distribution outlook Plates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . No
Press makeready . . . . . . . . . . . . . . . . . No
bears continuous scrutiny. Press running . . . . . . . . . . . . . . . . . . . . . Yes
Bindery set-up. . . . . . . . . . . . . . . . . . . . . No
NTIS-SupDocs cooperation could be espe- Bindery running . . . . . . . . . . . . . . . . . . . . Yes
Paper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yes
cially synergistic with regard to low-demand Ink . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yes
items. At present, the NTIS product sales line SOURCE F J R;mano, 1988
—-
is dominated by low demand documents, but
NTIS does not have the resources or mecha-
nism to invest in the electronic technology best line searching and printing-on-demand of
suited to low demand dissemination. On the selected pages.
other hand, the SupDocs product sales line is
almost devoid of low demand items, yet Sup- Electronic printing provides cost-effective
Docs does have access to the GPO revolving multiformat output capability and is especially
fund for capital investment in electronic tech- suited to low-volume, shorter, ald simpler doc-
nology (subject, of course, to approval of the uments with straight text or text and tables
Public Printer and JCP and to overall GPO and a minimum of photographs and complex
funding constraints). An NTIS-SupDocs co- line art (high-end systems can handle photos
operative initiative could design an Electronic and art work, although at higher cost). Best
estimates suggest that over half of the docu-
Document System (similar to the DTIC pro-
ments printed by GPO, and about 90 percent
totype) that would meet NTIS needs plus a
of the documents printed by other agencies,
broadening of the SupDocs product line to in-
are 100 pages in length or less. Estimates alSO
clude selected low demand items.
indicate that about 90 percent of all material
The economics of electronic printing-on- is straight text (80 percent) and tables (10 per-
demand for low volume documents are quite cent). The detailed breakout is shown in Table
simple. Many of the cost elements in conven- 5-16.
tional printing are essentially fixed, and are
not affected by the number of copies printed,
Table 5-16.—Estimated Page Length and Content
as shown in Table 5-15. of Government Documents

Thus most costs are independent of the size Other Overall


of the press run, and reducing the length of GPO. . —Government
—— average
the press run increases the per unit printing Page length
10 pages or less . . . . . . . . . 9“/0 13’Yo
cost, all other things being equal. Electronic ~1-50 pages . . . . . . . . . . . . . 23 30
110/0
26
printing eliminates most of the prepress func- 51-10() pages . . . . . . . . . . . . 25 37 31
tions, although the cost of toner (e.g., for laser 101-250 pages . . . . . . . . . . . 20 10 15
251-499 pages . . . . . . . . . . . 15 10
printers) is higher per page than the cost of 500 pages or more . . . . . . . 8 : 7
printing ink. Electronic printing is generally Page content
less expensive per page at volumes of tens to Text ., . . . . . . . . . . . . . . . . . . 79.30/0
a few hundred. In addition, electronic print- Tables ., . . . . . . . . . . . . . . . 11.8
Line art . . . . . . . . . . . . . . . 4.8
ing facilitates electronic linkages bet ween the Photographs ... , ... . . . . 4.1
document database and user terminals for on- SOURCE GPO and F J Romano 1988
124

In sum, many government documents are suit- lated sales program that at present is limited
able for electronic printing if the demand is low. to a few magnetic tapes. An expanded Sup-
Clearly many NTIS document sales items meet Docs program could start to duplicate NTIS.
this criterion. At present, few SupDocs sales A single coordinated governmentwide sales
items meet this criterion, since average sales mechanism presumably would be more effi-
volume is about 2,000 copies (and the average cient and easier for both the participating agen-
GPO press run is 3,000 to 4,000 copies). How- cies and the customers. Many agencies would
ever, a significant number of government doc- still be likely to distribute some computer prod-
uments not presently included in the GPO ucts themselves. A coordinated and possibly
sales program may be viable on an electronic even consolidated NTIS-SupDocs computer
printing-on-demand basis. The combined NTIS product line could also benefit from appropri-
and SupDocs low-demand sales volume could ate use of GPO bookstores, catalogs, and ad-
help justify investment in the necessary equip- vertising, and would fit well with the concept
ment. An Electronic Document System could of a governmentwide index to Federal infor-
be funded out of the GPO revolving fund and mation in all formats.
charged back to SupDocs as depreciation, just Another potential advantage of NTIS-Sup-
like any other SupDocs capital investment.
Docs cooperation would be to improve coordi-
NTIS could reimburse SupDocs for a prorated
nation among all four of the governmentwide
portion of the capital investment, funded out information dissemination mechanisms (Sup-
of NTIS retained earnings (if authorized by
Docs, NTIS, DLP, and CIC) and help insure
Congress).
that statutory requirements are fulfilled. It is
NTIS-SupDocs cooperation could also be also possible that improved cooperation would
synergistic with respect to sales of what NTIS result in reduced total overhead and indirect
calls computer products. As noted earlier, this labor, due to efficiencies in certain manage-
has become a significant product line for NTIS, ment and administrative functions. However,
one of the few showing recent sales growth. a full analysis would require more detailed in-
However, it is likely that only a small fraction formation on NTIS and SupDocs cost and la-
of agency computer products are included at bor force structures.
the present time. SupDocs has initiated a re-
Chapter 6

Information Technologies,
Libraries, and the Federal
Depository Library Program

Clockwise from top left: library shelving with document collection materials; librarian assisting user at reference desk;
librarian assisting user with map collection; and user on an OCLC terminal (photo credits: Documents Center,
Robert W. Woodruff Library, Emory University).
CONTENTS
Page
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................127
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........128
Role of Information Technologies in Libraries . . . . . . . . . . . . . . . . . . . . . . . .131
Use of Specific Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ...132
Online Database Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...133
Library Communication Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134
Electronic Bulletin Boards . . . . . . . . . . .............................135
Optical Disks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................136
Facsimile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................137
Summary . . . . . . . . . . . . . . . . . . . . . ................................138
Federal Depository Library Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..138
Origins and Operations of the Depository Library Program . ..........138
Format of Depository Library Materials: Paper v. Microfiche ... ... ...140
Dissemination of Information in Electronic Format. . . . . . . . . . . . . . . ...142
Online Catalogs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............144

Table
Table Page
6-l. Depository Library Access to Information Technology . . . . . . . . . . . . .133
Chapter 6

Information Technologies, Libraries, and


the Federal Depository Library Program

SUMMARY
Chapters 6 and 7 explore the role of libraries, microcomputers, online databases, library
and particularly those participating in the U.S. communication networks, electronic bulletin
Government Printing Office (GPO) depository boards, facsimile, and optical disks. Next, the
library program (DLP), in the dissemination history of the depository program is briefly re-
of Federal information to the public. ] The viewed, followed by a description of current
program is a cooperative activity between the dissemination efforts in the Library Programs
Federal Government and approximately 1,400 Service. Three topics concerning access to gov-
libraries. The government provides copies of ernment information are examined in detail:
government-produced materials free of charge dual format which concerns the distribut-
to the libraries; the libraries, in return, provide ion of selected materials in paper and mi-
housing for the documents and access to this crofiche:
information free of charge to their patrons. provision of government information in
DLP is a principal avenue of access to govern- electronic formats to depository libraries
ment information for the public. It is recog- through a pilot project program; and
nized as one of several guaranteed channels the development of online catalogs in de-
of public access to government information pository institutions.
established by Congress in support of our
democratic form of government, and serves in The three topics are concerned with meeting
part as an ‘‘information safety net for mem- the information and format needs of users,
bers of the public. This safety net is changing while at the same time facing and resolving
because of the increasing use of information new financial issues.
technologies by Federal agencies in support OTA has found that depository libraries are
of agency programs. This use is influencing the increasingly incorporating new technologies in
way in which agencies conduct their business, support of user services and operations. The re-
and how citizens access government infor- sults of the General Accounting Office Survey
mation. of Federal Information Users, when compared
This chapter examines how libraries employ to earlier depository library data, indicate a
a variety of information technologies to sup- strong and growing technology base in deposi-
port their mission of “allowing people to uti- tory institutions. For example, 83 percent of
lize information. First, the chapter reviews those surveyed have access to microcomputers
the role of libraries in the dissemination of gov- with modems for online access, 95 percent have
ernment information in the United States. This access to microfiche readers with printers, 41 per-
is followed by a discussion of key technologi- cent have access to a CD-ROM reader, and 36
cal trends and applications relevant to libraries percent have access to a mainframe computer
in general and to depository libraries in par- facility. The survey also found that these same
ticular. The technologies examined include institutions intend or wish to expand their use
of information technologies within the next 3
~ ~ ~ this rewrt, use of the phrase “depositor~ program” years to support user information needs. OTA
refers onl~’ to the CI PO depositor~ libra~’ program. has concluded that information technologies,
‘hl, Turoff and M. Spector, “ I,ibraries and the 1 replications
of Computer Technology, proceedings of the,1 FI1’.S ,\’ati~~nal if appropriately planned and executed, hold the
Computer (’(reference, 1’01. 45. 1 9 7 6 , promise of helping to achieve the original goals

127
128

and intent of the depository program through ians and information providers are experiment-
enhanced access to government information. ing with current electronic capabilities and fu-
ture opportunities in order to meet user
Information technologies are changing how
information needs. For example, it appears
libraries function and how users seek informa-
that since government information has been
tion. Many libraries are deploying the elec-
integrated into library collections through on-
tronic technologies to become gateways to in-
line catalogs, use of the information has in-
formation with the use of local, State, regional,
creased significantly.
and national networks and information
services—both public and private. The rela- While these technologies present the user
tively recent, rapid introduction of new infor- with different types and levels of access, they
mation applications, such as full-text online also present both the librarian and user with
retrieval of networked information services new cost concerns and format decisions.
and CD-ROM tools, demonstrates that librar-

INTRODUCTION
People need information to perform a vari- responsibility of every American to be in-
ety of daily tasks, to participate in govern- formed.3
mental deliberations, to vote, to be effective Recognition of the importance of an informed
members of a community, to make business citizenry has been affirmed since the found-
decisions, and more. As the largest collector ing of the country, and continues through the
and disseminator of information in the United enactment of new laws such as the Freedom
States, the Federal Government is responsi- of Information Act, Government in the Sun-
ble for creating and disseminating much of this shine Act, and the law establishing the DLP.
“information” used by the public. Information As stated by Senator Lausche during hearings
reaches the public through a number of for- on the Depository Program in 1962:
mal and informal, complimentary and competi-
tive channels. These range from agency pro- Although it may sound trite, an intelligent,
grams with specific dissemination charters to informed, populace has been, is and will con-
private sector services, and from public inter- tinue to be the fundamental element in the
est group efforts and the media to libraries— strength of our Nation. Contributing greatly
to that intellectual strength is the so-called
State and local public libraries, libraries in aca- Government document, designed to dissemi-
demic and research institutions, special nate to the American public important infor-
libraries, and Federal libraries. mation relative to the activities and purposes
Many of these channels are supported by the of its Government.4
Federal Government in recognition of the im- There is also the understanding that: “equally
portance of public access to government infor- important is their (the people’s) ability to ac-
mation. This is a basic tenet of U.S. society cess all other types of information, informa-
and is considered vital to the functioning of
our democratic form of government. As stated
by Jefferson: ~Letter to Cd. Charles Yancey from Thomas Jefferson, July
6, 1816.
If a nation expects to be ignorant and free ‘U.S. Congress, Committee on Rules and Administration,
Subcommittee on the Library, Depositor-y Libraries, Hearings
in a state of civilization, it expects what never on S.2029 and H.R. 8141 To Revise the Laws Relating to De-
was and never will be . . . if we are to guard positor-y Libraries, 87th Cong., 2d session, Mar. 15-16, 1962,
against ignorance and remain free, it is the p. 25.
—-

129

tion that has a direct bearing on the quality net’ to the public, an existing institutional
of life our citizens enjoy. mechanism that guarantees a minimum level
of availability of government information to
In addition to democratic and quality of life
all members of the public, ” and “the Federal
principles, the DLP serves the business com-
Government shall rely upon the depository li-
munity, which is important to local, State, and
brary system to provide free citizen access to
national economies. Congress, through the
public information. ”
establishment of the DLP, specifically recog-
nized the need for a guaranteed channel of ac- There are many classes of government in-
cess to government information by citizens, formation collected for a variety of purposes,
and in Title 44 describes the purpose of the and these are disseminated to the public
program as an avenue of dissemination of gov- through the DLP. Some information is referred
ernment information free of charge to the to as process, core, or basic information such
public: as that found in the Federal Register and Con-
gressional Record, executive and congressional
The depository library system is a long-
established cooperative program between the budgetary information, and the like. This in-
Federal Government and designated major li- formation is recognized as both a product of
braries throughout the United States under the operation of government and a necessary
which certain classes of Government publica- element to maintaining an educated and in-
tions are supplied free of cost to those libraries formed citizenry. As noted by members of the
for the purpose of making such publications Subcommittee on the Library, “Government
more readily accessible to the American publications generally serve two main pur-
public.” poses. In the first place they have a functional
The primary mission of the program as set value in the agency which issues them. Sec-
out in the 1977 Guidelines For the Depository ondly, and often quite as important, they have
Library System is: “. . . to make U.S. Govern- an educational value which makes their avail-
ment publications easily accessible to the gen- ability y to the American public a highly desira-
eral public and to insure their continued avail- ble objective. In the course of fulfilling
ability in the near future. The Guideh”nes a.lso their missions, agencies collect information.
note that the materials will be forwarded to Some agencies, such as the Bureau of Census,
the participating institutions” without delay, ” collect information on the population as their
again to insure timely access to information mission; other agencies, such as the Depart-
by citizens.7 There are two other elements of ment of Transportation, collect information in
program mission: use of government docu- order to effect policy and regulation. This same
ments by the academic/research community; information is then used by a variety of com-
and educational needs and use. munities—business and industry, academia,
and others—for a variety of purposes.
The Office of Management and Budget (OMB)
also recognized the importance of the program The Federal Government has long recog-
in Circular A-130 and noted that: “depository nized the importance of libraries as a channel
libraries provide a kind of information ‘safety for disseminating information it has collected.
The role of libraries in society, and the unique
“Testimony of <Joseph Duncan on behalf of the 11A in U.S. role of libraries in support of the “public good,
Congress, Committee on [~overnrnent Operations. Electronic
Collection and Dissemination of Information b?’ Federal .4gen- “Office of Management and Budget, ‘ ‘Management of Fed.
cies: A Policy 0~’er\riew’, 99th Cong., 2d CWSS., House Report eral Information Resources”, Circular No. A-130, Dec. 12, 1985,
NO. 323, 1986, p. 52. and ‘Improved Management and Dissemination of Federal In-
‘U.S. Congress, Senate Committee cm Rules and Administra- formation: Request for Comment, ” Federal Register, \rol, 45,
tion, op. cit., footnote 4, p.1. June 9, 1980, p. 38462.
‘Depositor IJibrary Council, Guidelines for the Depositor> WJ. S. Congress Committee on Rules and Administration,
I.ibrar?- S~rstem GPO: il$rashington, DC: Oct. 18, 1977). p. 1. Senate Report No. 1587, 87th Cong., 2d sess., 1962 p. 8.
130
.-

have been well defined. Libraries perform a of Congress (LOC) was established and is now
number of tasks in our society-’ ‘conserving the largest library in the world. It continues
and preserving our cultural heritage,”] 1 pro- to be the principal library for Congress. In the
viding education resources to various pub- late 1850s the DLP was established to make
lics, and disseminating government informa- congressional and other governmental infor-
tion. “The library. . . collects all the knowledge mation more broadly available to the general
of society, all the information, unedited, un- public. The establishment of a depository
screened, unrewritten, and instead of broad- library system was further affirmation by Con-
casting it to the masses, organizes, and directs gress of the need for a sound distribution sys-
that information to the individual. ”]z A s tem for government documents through li-
noted by Curley, “Libraries do not serve braries.
merely individual, informational, and recrea-
In addition, two national libraries were
tional interests, but are part of the essential
established-the National Library of Medicine
fabric of our society– its fragile cultural and
(NLM) began in 1836; and the National Agri-
social ecology. ‘1{] Libraries and librarians pro-
cultural Library (NAL) was created in 1862
mote access to all types of information and rep-
with the establishment of the U.S. Department
resent user interests and information needs. A
of Agriculture. A variety of other information
library collection, regardless of format, reflects
dissemination mechanisms were subsequently
the information needs of its users, whether they
created, expanding the number of avenues for
be the local community, academic, research, spe-
citizens to receive government information—
cial interest institution, State, or region.
the National Archives in 1943, now known as
Today , there are over 8,000 public libraries, the National Archives and Record Adminis-
s,000 college and university libraries, 88,000 tration (NARA); the Federal Library Commit-
elementary and secondary school libraries, tee in 1965, now known as the Federal Library
2,700 Federal libraries, and 11,000 private and and Information Center Committee (a cooper-
other special libraries in the United States. ative organization of Federal libraries); the Na-
tional Technical Information Service (NTIS)
This number and diversity are due in large
in 1970 (its predecessor, the Office of Techni-
part to Federal Government recognition of the
cal Services, was created in 1946); and other
importance of access to information through
Federal depository programs such as the Pat-
libraries. Since the founding of the Nation,
ent Depository Library Program. In addition,
there has been government support of libraries.
a series of congressional actions led to in-
The Continental Congress arranged with the
creased Federal involvement in libraries and,
Library Company of Philadelphia to receive
expanded the role of libraries in the provision
needed information for its members, and the
First Congress of the United States arranged of information to citizens.
access to the New York Society Library for Since the Library Services Act (LSA) was
similar purposes. In April 1800, the Library passed in 1956, the relationship between the
Federal Government and libraries has ex-
“ )Public good is the concept that the “good” for society is panded markedly. Libraries are one means by
greater than the well-being of certain individuals within it; see which the Federal Government seeks to pro-
I.ibraries, Coalitions and the Pubfic Good, E.J. Josey, cd., (New vide educational resources, services, and op-
York, NY: Neal-Schuman Publishers, Inc., 1987).
‘‘ Robert Wedgeworth, “A Library Agenda for the 1980’ s.” portunities to both a broad populous and to
in An Information Agenda for the 1980 ‘s, Carlton C. Rochell, specific segments of society. LSA provided li-
cd., Proceedings of a Colloquium, (Chicago: American Library brary services to rural areas, and the Higher
Association, June 17-18, 1980), p. 94.
‘JJohn N. Berry III, “The Public Good: What Is It?” Education Act of 1957 authorized funds for
Libraries, Co&”tions and the Public Good, E.J. Josey, cd., (New the purchase of books, periodicals, and other
York, NY: Neal-Schunlan Publishers, Inc. 1987), p. 10. library materials; library training programs;
] ‘Arthur Curlev, “Towards a Broader Definition of the Pub-
lic Good,’” I,ibrar;es, Coalitions, and the Pubiic (iood, kj.J. Josey, and R&D for new ways to program, process,
cd., (11’ew }’ork, NY: Neal -Schuman Publishers, Inc. 1987), p. 36. store, and disseminate information. The Li-

131

brary Services and Construction Act (LSCA) tion of government information. As noted in
provides services to rural areas and allows congressional hearings on the depository li-
funding for facilities’ construction, enhancing brary program:
of interlibrary cooperation, and increased serv-
The Clovernment is able to make such infor-
ice for physcially handicapped, disadvantaged, mation available to the citizenry due in large
and bilingual individuals. 11 LSA, the Higher measure to the splendid cooperation of the
Education Act, and LSCA have enhanced the American library profession. This is a ser~’ice
libraries’ ability to serve the general population, to the Nation which its libraries ha~re per-
and with various government information dis- formed in the past, are presently performing,
semination programs, serve to strengthen and and are anxious to perform in the future to a
reinforce the role of libraries in the dissemina- greater degree and in a more comprehensive
manner. 13

1“~1.s, congress, committee on Rules and t’+dm]rli~[r:it ion,


op. cit., footnote 4, p. 26.

ROLE OF INFORMATION TECHNOLOGIES IN LIBRARIES


All Libraries employ a variety of information better fulfill their missions. As stated by ~Bris-
technologies in support of their mission of ‘al- coe et al.:
lowing people to utilize information.’’”) The Technology has already changed the tradi-
following section discusses the role of technol- tional way in which libraries operate, and this
ogies in libraries and reviews a few key infor- trend will continue. The library needs to per-
mation technologies and current applications. sist in its role as a knowledge institut,ion––
Emphasis has been placed on those technol- mankind’s archive and encyclopedia—while
ogies found in depository institutions. providing the necessary services of an infor-
mation broker: computer literature searching,
Although over time the physical form of in- information retrieval, and document de.
formation has varied from manuscripts to au- livery .18
diovisuals, to online service, and to other tech-
nologies, the need of the librarian to access this As libraries increasingly employ the technol-
information for users has remained constant. ogies and expand access to all types and forms
A library is an institution that acquires, man- of information, the role of the library and in-
ages, and disseminates information. Moreover, formation specialist will not diminish. In fact,
‘‘a library is a bibliographic system regardless the current role will likely increase. The advent
of the situation in which it is placed, and the of “user friendly’ software available to users
task of the librarian is to bring people and for accessing electronic information systems
graphic records together in a meaningful rela- will increase the number of users in libraries
tionship that will be beneficial to the user. “l; and elsewhere, and at the same time many
users will still require information specialists.
Information technologies offer libraries op- For example, specialists in government infor-
portunities and capabilities for enhancing their mation will: assist users in identifying sources
current services and for allowing libraries to to search, provide users with some assistance
in using search technologies, and/or in some
cases actually perform the search for users.
“’Turoff and Spector, op. cit., 1976.
‘ ; Pau!ine W i l s o n , A Ccn]rnunit> Elice and tbe /)ul)ll’(, [,i- “P. 13riscoe, et al., “Ashurbanipal’s P;nduring Ar~het~’p~:
brar.v: The li.~es of in forma~ion in I,eadership (W’estport, (’T: Thoughts on the I~ibrary Role in the Future, ” College and I<e-
19771, p. xii. search Libraries, March 1986, pp. 121-126.
132

These same technologies, by integrating gov- This, in turn, has affected libraries and their
ernment information into the full library col- ability to purchase new systems.
lection, will increase both the use of govern- These technologies and technological appli-
ment information and the use of the total cations are merely machines or processes for
resources of the library and other local, State, distributing information-the content does not
regional, and national information resources.
vary, though one can do more and different
Information technologies are not “new” to things with information in electronic form than
libraries. A broad range of technologies have in paper form. As noted by the Commission
been employed by them for years and have af- on Freedom and Equality of Access to Infor-
fected all aspects of library operations and mation:
services. In fact, it has been noted that:
. . . the new technology not only gives poten-
“Almost every function carried out in a library tial users quicker and more convenient access
has been altered to some extent by electronics, to wider bodies of information, including ins-
computerization, and telecommunications. tantly current information, than can be pro-
Software is available for most aspects of li- vided by print alone; it also gives the user a
brary operations: circulation, inventory, acqui- new kind of abilit y to search through and man-
sitions, periodicals, cataloging, and reserves. ipulate the information, and in effect to cre-
The use of technologies for information user ate new information by the selection, combi-
services has resulted in the formation of library nation, and arrangement of data. Moreover,
networks, and has spurred the development the user can alter the data in a kind of two-
of national databases, thus allowing faster and way transaction.23
more efficient access to information.20 “The A variety of technologies are found in de-
changes brought about by advances in tech- pository libraries, though not always in the
nology have been so extensive that it is diffi- documents collection. The amount or types of
cult to assess their total effect, but it is clear technologies available reflect, in some respects,
that libraries are in a stage of fundamental the parent institution. Twenty-three percent
transformation. “2] Generally, library automa- of the depository libraries are public libraries,
tion refers to systems and technologies that 55 percent are academic research institutions,
provide improved access to resources within 7 percent are Federal libraries, 11 percent are
a library, whereas information automation law school libraries, and 4 percent are special
refers to systems and technologies that pro- institutional affiliations such as special librar-
vide access to resources outside the library. ies and historical societies.
A growing range of information technologies
are regularly employed in all types of libraries, Use of Specific Technologies
though the cost of some of these needed tech- In a 1984 survey of depository libraries, the
nologies is still prohibitive for many libraries,
Ad Hoc Committee on Depository Library Ac-
due to fiscal constra.ints.22 Library funding
cess to Federal Automated Databases (ap-
comes from a number of sources, including
pointed by the Joint Committee on Printing
State, local, and Federal governments, all of
[JCP]) concluded that:
which have experienced reduced revenues.
. . . there is a wide array of computer equip-
“Barbara Moran, Academic Libraries, The Changing Knowl-
ment already in place in depository libraries
edge Centers of Colleges and Universities (Washington, DC: or their parent institutions, and that many of
Clearinghouse on Higher Education, 1984) p. i. the libraries regularly make use of time-shar-
“)U.S. Congress, Office of Technology Assessment, informa- ing services for searching databases, both
tion Technology and Its Impact on American h’ducation (Wash- Government and non-Government.
ington, DC: U.S. Government Printing Office, November 1982)
p. 238. ~ ‘American Library Association, Commission on Freedom
‘l Moran, op. cit., footnote 19. and Equality of Access to Information, Freedom and L’qualitj’
“For those institutions unable to afford a “new” technology, of Access to Information (Chicago, IL: 1986), p. 31.
the user’s access to desired information may be limited as some 24U.S. Congress, Joint Committee on Printing, Provision of
information is not available in more than one format. Federal Government Pubh”cations in Electronic Format to De-
(continued on next page)
133

Since that survey, more depositories have brary automation tasks. In addition, there
adopted information technologies.25 As indi- were over 140,000 microcomputers in elemen-
cated in the GAO Survey of Federal Informa- tary and high school libraries. Recent survey
tion Users, for the 403 responding of the 451 data, including the GAO Survey of Federal In-
depositories surveyed, libraries were equipped formation Users, indicate further growth and
as shown in Table 6-1. purchases by libraries. A recent survey discov-
ered that the mean expenditure spent on li-
Depository libraries employ one or more of
brary automation per library over the past 5
the following technologies and/or technologi- years was $38,000. As in the 1984 survey,
cal applications: microcomputers, online data word-processing software continues to be the
services (bibliographic, numeric and others), most popular software, followed by software
networks such as OCLC (Online College Li- for database management purposes and sta-
brary Center) and RLIN (Research Libraries tistical uses in academic, public, and special
Information Network), automated information libraries. School libraries prefer word process-
systems, electronic bulletin boards, optical ing as well, though statistical, database, in-
disk technologies such as videodisk and CD- ventory, graphics, and spreadsheet software
ROM, facsimile, and microfiche and related are also used in these institutions. PC’s are em-
equipment. (A discussion of microfiche can be ployed in support of administration, catalog-
found in a following section on the format of ing, and reference purposes the majority of the
materials in the depository library program. ) time.2G
These are the primary technologies and tech-
nological applications in use today and those Online Database Services
most likely to be found in libraries within the
next 5 to 10 years. Online database services, such as DIALOG,
BRS, and other computerized retrieval sys-
In a 1984 survey, over 5,000 public libraries, tems, cover a wide array of continually expand-
1,600 academic libraries, and more than 7,000 ing subject areas. Each database is a compila-
special libraries were using microcomputers for tion of textual, statistical, and/or bibliographic
a variety of information automation and li- information. Bibliographic and referral data-
bases are sometimes called reference data-
bases, whereas numeric and textual-numeric
(continued from pre~ious pagel
databases are called source databases. In 1979-
pository Libraries, Report of the Ad Hoc Committee on Depos- 80 there were 400 databases, 221 database pro-
itory Library Access to Federal Automated Databases (Wash- ducers, and 59 online services available. By
ington DC: U.S. Government Printing Office, 1984) p. 3.
‘ > Discussions with Joseph McClane, Chief Library Inspec-
1987, there were 3,169 databases, 1,494 data-
y~
tion Team, LPS, and Mark Scully, Director, I.ibrary Programs base producers, and 486 online services.
Service, U.S. Government Printing Office. Dec. 8, 1986. These services allow rapid access to informa-
tion sources, can integrate information for the
Table 6“1 .—Depository Library Access to Information user, permit libraries greater flexibility in a
Technology choice of format, and provide access to previ-
Number of Libraries ously unavailable information. Use of these
Technology with Equipment services also allows the library to be less de-
Microcomputer without ‘ - pendent on paper or hard-copy indexing ma-
modem . 283 terials. These services are a primary means of
Microcomputer with modem
for online access . . . . 337 accessing certain types of government infor-
Microfiche reader without mation not found elsewhere (e. g., government
printer . . . . . . . . . 352
Microfiche reader with printer 384
CD-ROM reader . . . . . . . . . . . . 169 “’Sur\’e\’ data from Cahners Research, September 1986, and
Videodisk player . . 72 “Table 1,-” I,ibrm-j. Journal, No\ember 1986, p. IJC8.
Mainframe c o m p u t e r 149 ‘7 Cuadra Associations, Director~’ of Oniine Databases (New
SOURCE GAO Federal Information User Survey, 1988 York, NY: 1986) \’ol. 7. No. 3, p. ~.
134

information found only in an online format, data to the extent desired. In general, the GAO
such as some Bureau of the Census data). Survey of Federal Information Users found
that depository institutions use online serv-
Online bibliographic services usually require
ices primarily for bibliographic and statisti-
a trained searcher to search the databases ef-
cal information. Regular library use of many
fectively, and also to limit search time and
of the Federally generated databases available
associated costs. A number of vendors and in-
through commercial vendors is limited because
stitutions, such as NLM and academic insti-
tutions, have introduced user-friendly software of the relatively high costs. Online systems,
such as DIALOG and BRS, have introduced
that may reduce both the need for trained
new services for “after hours” users that can
searchers and the costs of online searching.
substantially reduce the costs of online search-
Pricing policies for online services vary. ing, if a library can accommodate requisite
Some services charge a monthly fee (e.g. $200 scheduling changes.
per month), as well as connect time ranging
from $4.00 to $45.00 per hour and system use Library Communication Networks
charges ranging from $.03 to $.90 per unit of
computer processing time. There may also be Two or more libraries may form communi-
disk storage costs incurred with certain serv- cation networks utilizing information technol-
ices. Prices of online services are most com- ogies to enhance the exchange of materials, in-
monly based on hourly connect charges in addi- formation, or other services. The formation of
tion to telecommunication costs for access to local, State, regional, and national networks
the network. These connect charges range from has significantly altered the operation of
$15 to $300 per hour. If offline printing occurs, libraries. There are several types of networks–
the user will typically pay per citation o r bibliographic utility, regional service organi-
page.” Online services are reexamining zations, and others (which include State-wide
connect-time pricing due, in part, to the in- publicly funded networks, local or geographi-
crease in transmission speeds. With the in- cally concentrated multi-institutional net-
crease in transmission speeds (from 110 bps works, and sub-regional subject-oriented
to 300 bps in the 1970’s to up to 2,400 bps or networks). AMIGOS, SOLINET, CLASS, and
higher today), users can perform more in-depth the like are regional service networks that fa-
searches, download, or print in a more cost- cilitate the expansion of the bibliographic util-
effective manner. NLM and Mead Data Cen- ity. Although bibliographic utilities began as
tral have revised their pricing schedules to ac- a means for libraries to reduce costs of cata-
count for this shift. For example, NLM now loging, their primary function today is for shar-
has a lower connect fee, and charges accord- ing of resources. One example of a biblio-
ing to the characters transmitted and the work graphic utility is OCLC, a major computer-
performed on a given search by the NLM based cooperative network with over 7,900
computer. members and employed by all types of libraries
nationally and internationally. The OCLC net-
A number of Federal agencies produce data- work assists librarians in acquiring and cata-
bases consisting of original statistical infor- loging materials, ordering custom-printed cat-
mation. Agencies such as the Bureau of Census alog cards, initiating interlibrary loan, locating
provide computer tapes of their information, materials in member libraries, and gaining ac-
sometimes, in lieu of the paper format. Use of cess to other databases. More and more depos-
these numeric databases allows the librarian itory libraries are using the OCLC database
to both provide the needed information to the for reference purposes to assist in searching
patron directly and be able to manipulate the for government documents. The GPO Library
‘* Ibid., pp. v-vi. Division catalogs government documents into
135

OCLC where they can be searched by member tion located elsewhere. The Wisconsin Inter-
institutions. library Service (WILS) network is one exam-
ple of the growing use of bulletin boards in
These networks are undergoing changes in
libraries. The WILS network is used by over
their structure and functions. Areas affected
one-half of the 55 member libraries, a combi-
include autonomy for members, changes in the
nation of public library systems and State li-
telecommunication infrastructure, decen-
brary resource centers, in the Wisconsin library
tralized versus centralized control, the devel- system. WILS can handle over 90,000 requests
opment of more integrated systems for librar-
a year. Users note the following advantages:
ies that permit less reliance on the utility and
greater emphasis on local resource sharing, and it is inexpensive and, in fact, is less costly
finally, the debate concerning ownership of than the previous system;
data found in the shared cataloging databases. it offers increased speed of communi-
As a result of network changes, libraries are cation;
being changed as well.]’) many members had the necessary equip-
ment (microcomputers and modems) and,
Automated information systems are those
therefore, it did not require special equip-
that assist the librarian in performing specific ment or hardware purchases;
library tasks such as circulation, inventory, ac- it has the capability to store and track the
quisitions, cataloging, administration, budget- requests in a database;
ing, personnel, and more. Many depository it reduces the amount of paper used to sup-
libraries use OCLC to perform many of these
port the ILL system; and
tasks because they lack access to other dedi- it enhances microcomputer use by library
cated systems or necessary software. An ex- staff. 31
ample of an automated information system at
NLM is DOCLINE. This is the Library’s auto- Libraries are also subscribing to bulletin
mated interlibrary loan request and referral boards containing government information.
system that automatically routes an inter- These boards contain timely information
library local request through the Regional produced by agencies. For example, the SRS
Medical Library Network. Requests for titles Remote Bulletin Board System (RBBS) of the
found in SE RLINE, the Library’s online data- National Science Foundation contains infor-
base of approximately 66,000 serial titles, are mation on financial and human resources for
also automatically routed, based on the hold- science and engineering activities. Also in-
ings of SE RHOLD, NLM National Biomedi- cluded is information concerning current
cal Serials Holding database, which contains studies of the Foundation, announcements of
the holdings of 2,276 libraries. available publications, and comprehensive sta-
tistical tabulations. Specific data contained
Electronic Bulletin Boards within the file include: “Federal Funds for Re-
search and Development, ” “Scientific and
Libraries are employing electronic bulletin Engineering Expenditures at Universities, ”
boards in support of library operations such ‘‘Employment and Demographic Characteris-
as interlibrary loan (ILL), resource-sharing tics: U.S. Scientists and Engineers, ” and “In-
functions, and for access to current informa- ternational Comparisons of Science and Tech-
nology Data, among others .32 The GAO

‘g13ecause GPO has been inputting to OCLC since July 1976,


a limited amount of retrospective searching is possible, though ‘ ] Cathy Moore, “Do-It-Yourself Automation: Interloan
it has been extensively noted that these earIjr GPO cataloging Bulletin 130ards, ” Library Jouma), }10}. 112, NO. 18, NOV. i, 1987.
records contain numerous errors. “National Science Foundation, “Remote Access to Science
‘ f’ Moran, op. cit, footnote 19. Resources Studies Data”, 1987.
136

Survey of Federal Information Users found a service application in videodisk format. The
minimal use of electronic bulletin boards by MINI MARC cataloging system is published
those surveyed. The predominant library use on two videodisks containing over 2.1 million
was for press releases and statistical data. Library of Congress MARC records– 1.5 mil-
lion MARC records on 52,900 video frames on
Optical Disks the first disk, and over 600,000 MARC records
on 27,000 frames and 17,000 video frames of
In a 1985 survey by Link Resources Corp., index data on the second. 36 The videodisk is
7.6 percent of the libraries contacted had one updated twice a month. ALDE (Applied La-
or more videodisks or CD-ROMs. Sixty-five ser Disk Efficiencies) Publishing produces the
percent of those responding forecast a pur- United States Code (USC) and the Code of Fed-
chase of optical disk technology by 1990.33 eral Regulations (CFR) using digitally encoded
The GAO Survey of Federal Information Users videodisks. These materials are available on
found that 169 of the 403 depository library disk and can be broken out into specific areas
respondents had access to a CD-ROM player. of Titles of Interest. For example, Title 26 (tax
Libraries are adopting optical disk technol- code) of the CFR is available annually with
ogies for both operational or technical services monthly updates. 37 Another example is
purposes and for reference services. In fact, IAC’S Government Publications Index on vid-
the”. . . library and information communities eodisk, which indexes the Monthly Catalog
are at the forefront of testing the various op- from 1978 to the present with monthly
tical media-videodisk, CD-ROM, and optical updates.
digital disk–in digital data publishing and
storage applications. These technologies CD-ROM, an optical storage device, “uses
can provide improved access to a variety of the differential reflection of light from a mirror-
like disk surface as a means of reading infor-
information tools and sources, are a means of
mation. 8 The following factors make CD-
preserving important documents and informa-
ROMS increasingly popular, particularly in
tion, and appear to be popular with users.
libraries and for database creators:
Optical disk technologies include videodisks,
compact audio disks, CD-ROMs, optical digi- ● storage capacity,
tal disks, and others. This discussion will fo- ● durability and stability,
● cost compared to magnetic tape and mi-
cus on videodisks and CD-ROMs. With regard
crofiche,
to videodisks, the very large storage capacity
● fixed searching costs,
and the ability to carry both video and audio
● the ability of users to perform the searches
information, are the two key characteristics
that make videodisks attractive technologies themselves without a trained librarian to
for libraries. There are a number of types of assist, and
videodisks with different capabilities. The la- size and compactness of the disk.
ser optical videodisk is the most accepted tech- Despite a lack of common information access
nology. One indication of wider acceptance of and retrieval standards, an increasing number
this technology is the recent drop in the price of vendors are introducing database services
of products as more data files are introduced on CD-ROM.
and competition increases.35
Use of a CD-ROM usually requires an inter-
The MINI MARC produced by Library Sys- active system consisting of a microcomputer,
tems and Services is an example of a technical a ROM disk, and a disk drive. Reference ma-
SSJudY Mc@een ad RichMd W. Boss, videodisk and @~i-
cal Disk Technologies and Their Apph”cations in Libraries, 1986 “Ibid., p. 115.
Update (Chicago, IL: American Library Association, 1986), p. ‘TIbid., p. 127.
105. ‘ hDonald Case and Robert Powers, Optical Disk Publication
]’Ibid., p. 3. of Databases: A Review of Applications for Acaderm”c Libraries,
]’Ibid., pp. 9-36. (Washington, DC: Council of I.ibrary Resources, 1986), p. 4.
137
.-

terials and large textual or statistical data- cards, transmit records, and more. Brodart
bases are ideal candidates for the CD-ROM for- markets the Le Pac: Government Documents
mat in some libraries and information centers. Option on CD-ROM. This service also uses
Reference materials are especially well suited GPO/LOC MARC records, and provides a pub-
to CD-ROM because they save shelf space and lic access catalog of about 230,000 titles of de-
do not require frequent updating. pository and nondepository titles from 1976
to the present on an annual subscription ba-
Books in Print and Ulrich Periodicals sis with bi-monthly updates. Auto-Graphics
Directory are now available from R.R. Bowker GDCS also produces a government documents
in CD-ROM format. In a joint venture with catalog on CD-ROM with monthly cumulated
Online Computer Systems Inc. who developed updates.
the search software, the Books in Print Plus
service includes all of the multivolume BIP, There are a number of other factors to be
the Subject Guide to BIP, BIP Supplement, considered by libraries as this technology is
Forthcoming Books and Subject Guide to introduced. CD-ROMs cannot be updated un-
Forthcoming Books, in addition to names and less a new disk is mastered. Therefore CD-
addresses of book publishers. This is contained ROMS are not practical for time-sensitive data,
on one disk. Ulrich Plus on CD-ROM includes Access time to CD-ROMs varies, and this may
68,000 periodicals, in alphabetical order by ti- limit the number of users able to use the sys-
tle, in 557 subject categories. tem concurrently. Different databases require
different access software and indexing struc-
Online databases are also available on CD- tures. The use of different search and retrieval
ROMS. The primary advantage of having these software packages by vendors results in diffi-
databases on disk is that the user may sit at culty for librarians when “putting up’ a new
a terminal for any length of time and not in- disk. This requires additional expertise and
cur high connect charges. This allows the un- training on the part of the libraries. Finally,
trained user to perform his/her own search. a microcomputer or PC and a CD-ROM reader
This user-oriented characteristic of CD-ROM are necessary, and this may represent addi-
explains some of the technology’s popularity. tional expense to the library. However, many
In fact, many libraries find the need to place libraries already have or will be purchasing
a time limit on the workstations due to the microcomputers.
popularity of using these disk files. AGRI-
COLA, the database compiled by NAL con-
taining citations on agriculture and related Facsimile
topics, is available on CD-ROM from $950 an- Facsimile is the transmission of printed in-
nually with a quarterly update. Another gov- formation (e.g., a letter, order form, interlibrary
ernment-generated database, ERIC (Educa- loan request) from one locale to another by
tional Resources Information Center), is also
encoding the printed materials into digitized
available from $1,750 with quarterly updates. form. The information is converted (or decoded)
The acceptance by users of the CD-ROM tech-
back to its original form once it is received.
nology has been rapid, and as a consequence, Current generation digital facsimile machines
vendors are quickly respondng through the are able to transmit one to three sheets of 81,’z
introduction of new products. by 1 l-inch paper per minute. This is a substan-
The Library Corporation markets Library tial improvement over analog machines that
of Congress (LOC) MARC databases in disk were only able to transit one page every 6 to
format. The BiblioFile Catalog Production Sys- 7 minutes. Facsimile machines area very quick
tem contains over one million Library of Con- method of relaying information between Iibrar-
gress MARC records on four disks. The user ies. The NLM facsimile program is an exam-
can search, edit, create, and save MARC ple of how this technology is currently used.
records, display the catalog card image, print NLM and a number of medical libraries are par-
138

ticipating in an interlibrary loan program provide timely and accurate information to li-
whereby NLM will send up to 20 pages of li- brary users in a variety of formats and for vari-
brary material to a member library in support ous purposes. For instance, as noted earlier,
of emergency patient care. A small amount of surveys show that all types of libraries are pur-
information is relayed quickly-this is not a chasing microcomputers in increasing numbers
printing-on-demand program for lengthy doc- for a variety of purposes. The GAO survey
uments. The project is limited to emergency demonstrates the growing technology base in
medical care for a number of reasons: a broader depository libraries and how new technologies
project could overwhelm the NLM interlibrary such as CD-ROM are becoming more widely
loan staff, the cost could be prohibitive, and accepted and used.
the majority of requests are satisfied by the
regular interlibrary loan program. Another ex- Most importantly, information technologies
ample is the use of facsimile machines by GPO permit access to a much greater range of in-
field offices. Field offices send notices of print- formation and resources, including govern-
ing requisitions via facsimile to the GPO Li- ment information through vendor (profit and
brary and GPO Sales Program. The GPO Li- not-for-profit) services. New types of Federal
brary and Sales Programs select items to be information resources, such as statistical/nu-
included in their respective programs and ad- meric databases from the Bureau of the Cen-
vise the field offices via facsimile of the items sus, are now online and available to libraries
and number of additional copies to be printed. through the use of information technologies
and vendors. Newer technologies such as CD-
Summary ROM are moving quickly from the marketplace
to libraries as producers place more and more
In summary, information technologies in- services in a CD-ROM format. Libraries are
dividually and collectively are changing the na- experimenting and employing these technol-
ture of access to govermnent documents via ogies in support of their operations, which, in
libraries and have the capability to improve turn, permits the user greater access to needed
access to government information. They can information.

FEDERAL DEPOSITORY LIBRARY PROGRAM


In 1813, Congress established a system for Origins and Operations of the
the distribution of congressional literature, and Depository Library Program
this system developed into the depository li-
brary program-a significant avenue for dis- There are approximately 1,400 Federal de-
semination of government information to the pository libraries in the United States and re-
public. The program has experienced a num- lated territories. These libraries provide Fed-
ber of changes since its inception, and is still eral publications without charge to the general
changing as participating libraries and man- public. This program is the primary avenue or
agers of the program at GPO debate how to “safety net ~ for dissemination of govern-
best serve the users of the depository system. ment information to the general public.
The following section provides a brief descrip- The DLP originated in 1813 when a resolu-
tion of the origins of the program and its oper- tion was passed authorizing the printing of ad-
ations. This is followed by a discussion of three ditional copies of’ congressional literature for
specific topics: 1) dual format distribution (pa- distribution to State governments and legis-
per and microfiche), 2) the dissemination of in- latures. The following year, the American An-
formation in electronic format, and online
catalogs. “’Office of Management and Budget, op. cit., footnote 8.
. . .—— .—— -

139

tiquarian Societ y became the first depository to the State Historical Society of Wis-
library. Responsibility for the distribution of consin;
materials shifted among a number of govern- expanding the variety of government doc-
ment agencies prior to resting with GPO. Con- uments available for distribution; and
gressional resolutions in 1857 and 1858 af- establishing a reporting mechanism to
firmed the distribution of congressional ascertain the libraies’ condition (the Bien-
materials to institutions such as libraries and nial Survey became the reporting vehicle).
colleges, and Members of Congress designated
There have been two changes to the 1962
organizations within their districts as deposi-
Federal Depository Act. The highest appellate
tory institutions. In 1895, a new printing act
court of each State became exempt from the
was passed, incorporating the old legislation
requirement of public access in 1972, and law
and placing responsibility for bibliographic
schools were eligible to become depositories
control efforts, distribution, marketing of pub-
under the law designation in 197S, ~: This leg-
lic documents, and the DI.P in the office of the
islation has expanded the total number of
Superintendent of Documents at GPO.*’) This
libraries in the program, since some of those
legislation also specified that certain (not in-
law libraries already participating became
ternal, confidential, or administrative) execu-
members under the “law” designation thus al-
tive materials were to be included in the de-
lowing for new participants under the separate
pository program. In addition, the act called
congressional designation. Another effect has
for a catalog to be published each month list-
been a substantial increase in law schools par-
ing government documents published the pre-
ticipating in the depository program; almost
vious month. A number of other points in the
one-half of the new depositories between 1976
legislation were central to the DLP-attaining
and 1985 were accredited law schools. The ap-
status as a depository library could be gained
pointment of librarians and knowledgeable in-
either through congressional designation or
dividuals to a Depository Library Council be-
through legal designation; and the Superinten- gan in 1972 in an effort to assist the I>ublic
dent of Documents could now “investigate” Printer and the Superintendent of Doc-
depositor-y libraries and evaluate their hold-
uments. 3
ings vis-a-vis the program. It was not until
1923 that depository libraries were able to se- One description of the Library Programs
lect those government documents most appro- Service is that of a “production shop, ’
priate to their clientele. 1’ From this perspective, its purpose is to act as
a transfer agent of government documents
The Federal Depository Act of 1962 revised
from Federal agencies to the member deposi-
the previous legislation by:
tories. By law (as stipulated in Title 44, all
. increasing the number of possible deposi- documents produced by an agency that are not
tory libraries; confidential, not for internal use, or not con-
● establishing a system of regional libraries cerned with national security belong in the de-
(two per State), which were to maintain pository program. In fiscal year 1986, 66,367
a permanent collection and provide inter- titles or 27 million copies of government doc-
library loan and reference services; uments were distributed to depository libraries.
providing for the transfer of certain doc- GPO staff state that the workload of the pro-
uments within New York and Wisconsin gram has remained relatively constant for sev-
to either the ~New York State Library or
‘-i bid., pp. 5-8.
“’’l’he General I)rinting Act of 1895, ch, 23, 28 SLat 601 (codi- 1‘Hernon, McClure, and Purcell, ~;p. cit., fmtn~te 42, p. 1 i.
fied as amended in scattered Sections 44 [J. S.C.). .jn earlier (’ouncil was formed in the 1960’s.
‘ i Peter Hernon, Charles hfc(’lure, and (jar~ Purcell, G P O ‘‘I)iscussion with Mark Scull~’, Director, I.ibrar~’ Programs
L)epositor.\r Librarj’ Program .4 Descriptit’c .4nai\’si.~ (Norwood, Ser\’ic’e, and [jonald Fossedal, Superintendent of Documents,
N,J: i4blex Publishing Corp.. 19X5), pp. 5-8, [’. S [~t)~ernment Printing Office. Dec. 8, 1986.
140

eral years. GPO staff estimate that an addi- through a legislative branch appropriation, is
tional 5 percent of the 66,000 titles or about responsible for the cost of distributing these
3000 titles are fugitive documents–those be- materials to member institutions if GPO prints
longing in the program but not included by the the documents. If another agency prints doc-
agencies .45 uments on its own premises or elsewhere, that
The operating cost of managing the deposi- agency is then responsible for the cost of print-
ing copies for depository distribution, with
tory program is provided by the GPO in the
GPO bearing the distribution costs.
annual budget. In fiscal year 1987, the bud-
get for the depository program was $19.7 mil- Over the past several years LPS, the Depos-
lion, and the fiscal year 1988 estimate is $20.2 itory Library Council, and members of the de-
million. DLP is managed by the Superinten- pository library community have debated the
dent of Documents. The principal mission of availability of government information in
this office is to “distribute government docu- different formats in the depository library
ments, and information about them for the program. There are two debates regarding
three branches of government. ”4G The DLP is format— the dual format debate that concerns
managed directly by the Library Programs materials distributed in paper and microfiche
Service (LPS), within the Office of the Superin- with libraries selecting either format; and the
tendent of Documents. The Joint Committee debate about inclusion of government elec-
on Printing (JCP) oversees the policies and tronic information products in the program.
overall direction of the program. Both debates are concerned with meeting user
Until recently, the Gu.i”deLines for the Depos- preferences on format, with the costs of pro-
itory Library System recommended that viding these products, and with ensuring ac-
libraries (other than regionals that receive one cess to government information regardless of
copy of all documents distributed) select a min- format. The focus of both debates is the ac-
imum of 25 percent of available documents, cessibility of the information and availability
and approximately 50 percent of the deposi- of the information.
tories select no more than 25 percent of the
available government documents. It is pre- Format of Depository Library
dicted that “. . . the U.S. Government Print- Materials: Paper v. Microfiche
ing Office will distribute approximately 20,000
paper documents and 43,000 on microfiche Materials sent to depository library partici-
each year. “47 For those libraries selecting the pants are either in paper format, microfiche,
minimum number of government documents, or a combination of both (although only
this represents approximately 15,000 docu- regionals can receive a title in both formats).
ments per year-requiring an enormous invest- Beginning in the early 1970s, the JCP and GPO
ment in space, collection maintenance, and began to explore the advantages and disadvan-
staff time by participating libraries. GPO, tages of instituting a microfiche publishing
program for depository materials. In 1977, fol-
— lowing a number of library surveys and com-
‘ s Fugitive documents continue to be a problem for the pro- mittee evaluation efforts, the JCP gave per-
gram, although members of the Library Programs Servicce be- mission to GPO to begin conversion of selected
lieve the number is declining. However, it has been noted by
members of the depository library community that the number depository materials to a microfiche format to
of fugitive documents is increasing, at the same time that the effect cost savings for the program and for par-
number of materials in the depository library program is de- ticipating libraries. Private information pro-
creasing.
%overnment Printing Office, Government Printing Office, viders objected to this practice at the time be-
Superintendent of Documents Description, Draft, (GPO, 1986), cause it was their stated position that the
p. 1. library community was already well served by
‘TDonald Case and Kathleen Welden, “Distribution of Gov-
ernment Publications to Depository Libraries by Optical Disk, private sector firms. At issue was the differ-
Government Publications Review, vol. 13, 1986, p. 314. ence in the scope and amount of materials to
.—

141

be offered by GPO in contrast to those avail- rials will remain in hard copy. In August 1983,
able from the private sector. Members of this the Superintendent of Documents issued SOD
community, as represented by the Information 13, a list of criteria for determining which doc-
Industry Association (11A), believed that pro- uments were more appropriate in microfiche
vision of free microfiche to depository institu- or paper format. Criteria include physical char-
tions would undermine their business, and acteristics (color, size, etc.,), timeliness, au-
voiced concerns that the Federal Government dience, frequency and type of use, savings in
would be the “sole” information provider to space, historical significance, and reference
libraries and other users of Federal infor- value. This directive recognized that certain
mation. documents are more suited to either paper or
microfiche and some documents to both for-
Since that time, the LPS has adopted a pol- mats. Depository librarians also recognize that
icy of providing more and more documents in
some conversion to microfiche is helpful in or-
microfiche format, primarily for financial rea-
der to reduce program costs, save space in par-
sons. Reduced production and postage costs
ticipating libraries, and make more informa-
of microfiche, compared to paper, allow sav- tion available to the public. The Depository
ings for the program. Many libraries have Library Council and the Public Printer con-
adopted microfiche to both achieve greater ac- tinue to work together to identify materials
cess to a broader range of government materi-
that can be converted to microfiche and those
als and reduce their maintenance costs. Hous-
that must remain in dual format (that is, dis-
ing of paper can be quite costly. In turn, use
tributed in both paper and microfiche). The
of microfiche has reduced the financial burden JCP passed a resolution on April 9, 1987 sup-
on GPO. In the spring of 1986, 54 percent of porting choice of format for depository insti-
the materials sent by GPO to member institu- tutions.
tions were in microfiche, and the number is in-
creasing. By December of 1986, 61.2 percent Dual format documents are the most heav-
of the materials were in microfiche.~s In addi- ily used titles in the majority of depository
tion, a number of agencies send their micro- libraries, and “are the fundamental records of
fiche materials directly to library participants, Government.’’” Secondly, it is important
based on interagency agreements resulting in that libraries receive dual format items such
a more decentralized operation. The Depart- as the F’edez-al Register in a timely fashion so
ment of Energy sends copies of microfiche that users can respond to proposed regulations
concerned with technical R&D information within a 30-or 60-day timeframe. The delay re-
directly to participating depository institu- sulting from conversion from paper to micro-
tions, and the U.S. Geological Service (USGS) fiche format and subsequent shipment can
ships cartographic microfiche materials for sometimes make a timely response impossible.
themselves and the Defense Mapping Agency Third, the format of some key documents, such
(DMA). The Equal Employment Opportunity as the Code of Federal Regulations, does not
Commission (EEOC) also provides agency de- lend itself to use in the microfiche format.
cisions in microfiche to depositories via its con- Given the high usage of key documents, the
tractor, IHS. EEOC pays the production costs need for receiving these documents in a timely
and is responsible for sending out the materi- fashion, and ease of access to information con-
als; GPO reimburses the EEOC for postage tained in the paper documents versus micro-
costs. fiche, it is understandable why a paper format
is preferred.
There is a continuing debate between the
LPS, member institutions, and the JCP over Library use of microfiche has a number of
what proportion of materials and which mate- advantages and disadvantages. On the pIus

1(]
“Discussion with Mark Scully, Director. I,ibrary Programs Conversation with Judy Myers, University of Houston Li-
Services, GPO Dec. 8, 1986. brary. tJune 1’7, 1987.
142

side, microfiche is an enormous space saver; transfers materials from the government to
consequently, more government information participating institutions. To others, the pro-
can be made available at the depository. Small gram is one that provides timely and inform-
colleges and public libraries in particular ben- ative government materials to citizens in sup-
efit from the distribution of government ma- port of the principle of public access. To those
terials in microfiche, and access is improved adhering to the access philosophy, the adop-
since many of these institutions otherwise tion of microfiche as the predominant format
could not afford to store the materials. The use negates both the accessibility and timeliness
of microfiche also permits libraries to retain objectives of the program.54 Dissatisfaction
more information for longer periods of time or with the microfiche format by library patrons
permanently. It is projected that, “. . . libraries and the added delay of conversion from hard
that accept all depository publications distrib- copy to microfiche are cited as critical
uted over the next 20 years will require an esti- factors.”
mated 7,500 linear feet of hard copy storage An added difficulty in resolving the dual for-
and 2,500 linear feet of microfiche storage. mat distribution debate is the poor but im-
However, microfiche also has disadvantages. proving relationship between the managers of
Librarians are finding that patrons prefer pa- the DLP and members of the depository library
per to microfiche as there are: community. There has been some improvement
in the relationship since the LPS began upgrad-
. . . problems with viewing and reproduction ing the quality of the GPO cataloging tapes,
equipment (that) have resulted in user com-
plaints of eye strain and unsatisfactory paper the inspection program, and pertinent train-
copies.51 ing programs and seminars, among other areas.
On the other hand, GPO’s failure to resolve
The cost to the patron is at least double when problems with its microfiche contractor has
duplicating pages from microfiche, compared exacerbated its relationship with depository
to copying paper documents, and the range of libraries.
costs to the library for the purchase of a mi-
crofiche reader/printer from Kodak, for exam- Dissemination of Information in
ple, is between $1,500 and $5,000, plus main-
Electronic Format
tenance fees. Also, there are added difficulties
in the organization and bibliographic control Microfiche and hard copy materials are the
of fiche.52 Another consideration is that con- only formats employed to date by GPO in the
version of a document to microfiche by GPO transfer of government information to depos-
adds 4 to 8 additional months to the process- itory institutions (except for the planned ship-
ing time prior to the document being shipped ment of the Bureau of the Census CD-ROM
out. GPO has stated that time-sensitive ma- “Test Disk No. 2“ to the depositories).’6
terials will not be included in the microfiche
program due to this extra delay. 5:’ GPO is currently reformulating agency pol-
icies with respect to electronic dissemination
The dual format issue exacerbates two some- in the depository program. Prior agency deci-
what competing and contradictory philoso- sions to withhold electronic information from
phies of the depository library program. To
many, the GPO program is simply one that ‘iIIA supports the continuation of the dual format programs
because it recognizes the need for items used more heavily to
50Case and Welden, op. cit., footnote 45, p. 315. be available in paper due to ease of access, patron preference,
5’ Ibid. and timeliness.
‘*Discussions at the Depository Library Council Meeting, ‘sDiscussions and meetings, GODORT, American Library
Oct. 15-17, 1986. Association Midwinter Meeting, Jan. 16-19, 1987, Chicago, IL.
~.~Discussion with Mmk &ully, op. cit., footnote 45. An IJPS “t’GPO has agreed to “ride” the Census “Test Disk No. 2“
microfiche contractor has defaulted on the contract, causing order and ship copies of this CD-ROM to all depository institu-
extensive delays in the creation and distribution of microfiche tions. The Census of Agriculture for 1982 and the Census of
to the depositories. Retail Trade by Zip Code will be included on this disk.
143
-.

the program were based on a GPO interpreta- ment data files. The Subcommittee on Legis-
tion of previous legislation concerning the de- lative of the House Committee on Appropria-
pository program, specifically section 1901 of tions supported dissemination in CD-ROM
the 1962 Depository Act. The opinion by formats in the fiscal year 1989 I.egislative
former GPO general counsel Garrett Brown Branch Appropriation Bill and requested a
determined GPO policy: copy of the GPO-JCP plan. In addition, the
Committee noted that online access and other
. . . the Depository Library Act [of 1962] does
not direct [the] Superintendent of Documents formats (excepting CD-ROM) may require ad-
make published documents available in all pos- ditional funding, and requested that GPO sub-
sible formats to the libraries. It was the intent mit any future electronic dissemination plans
of Congress that only printed publications to the Subcommittee on Legislati\’e. -’(’
would be made available to depositories, :); The JCP position on the dissemination of
(3PO now supports the position that, while it government information in electronic form re-
cannot require agencies to submit electronic sulted in part from the deliberations of the Ad
products for distribution through the deposi- Hoc Committee on Depository Library Access
tory system, agencies may voluntarily submit to Federal Automated Databases. The Com-
electronic products to GPO. Also, those elec- mittee’s efforts were based on a request from
tronic products available in paper or microfiche the JCP to: “. . evaluate the feasibility and
format can be disseminated to depositories in desirability of providing access to Federal Gov-
electronic format since these materials have ernment information in electronic formats to
already been ‘‘published. ” depository libraries. ” The Ad Hoc Committee
has considered a variety of formats, databases,
The recent plan to distribute a Bureau of the and institutional arrangements for the pro\’ i-
Census CD-ROM to depositories prompted the sion of agency information for the past A
Public Printer to request approval from the years.’”)
JCP and clarification of the Committee’s views
concerning disseminatiOn of government infor- This advisor~ committee intended to recom-
mation in electronic formats to depositories. mend to Members of the JCP certain agency
In a March 25, 1988 letter to the Public Printer, databases for depository distribution in online
Congressman Frank Annunzio, Chairman of the or CD-ROM format to test electronic dissemi-
JCP, affirmed both the Committee’s support nation to depositories. These recommendations
of the Census project and the position that the were to be based, in part, on proposals made
GPO’s authority as required by Title 44, UiN”ted by each agency to the JCP. As of December
States Code, extended to the “production and 1986, 16 proposals were received by the JCP
distribution of Government publications in from Federal agencies hoping to participate
these new formats. “;)” in the pilot program. (;’ These proposals
ranged from provision of 4 possible databases
GPO and the JCP recently developed a re- from the U.S. Geological Survey-the Geologic
search plan that identifies selected electronic Map Index, the Library System Catalog, the
data files as products for depository distribu- Mineral Resources Data System, and the Se-
tion. This plan was approved by the JCP on lected Water Resources Abstracts–to 3 data-
.June 29, 1988. The demonstrations involve a bases from the Department of the Treasury–
ombination of online and CD-ROM govern-
“’’U.S. Congress, Commitw, on ,\ppropriations, Legiskt~\(~
,- [). S. ”Congress, ,Joint Conlmittee on I]rinting l’r~~~i.sion of
Appropriations Bill, 1989 Report to accompanJ, H. R. 4587,
[’edt’ral {;o\’c3rnment ])utlficaLiOns in Eiectronl(’ [<’<)l”nl:jt t (~ [~tI-
Report No. 1W621, I()()th C’on~,. 2d session, 1988.
posj[or~” l.ihrarie.s, Report of the .Ad Hoc Comfllit we ()[1 h?poS- ‘(’The Ad HOG Committee is c{)rnpc)sd of indit~iduals from
it~r~’ I.lbrar}T Access to Federal Automated Databases (\$’ash- government agencies and representatives of pertinent ass{)-
ington, 1)(”: U S. (iovernrnent Printing office, I 984), pp. 112-113. ciations.
“l, etter from the ~][J1lOr~lbh? Frank Annunzio, Chairman, “‘The tJCP is still recei~’ing new pilot proj(lct proposals” from
Joint Commit tee on I’rint ing CO the Ilonorable RaJph Kennick- Federal agencies interested in pro~iding electronic products t ()
ell, ,J r., the I’ub\ic I)rinter, hlar. 2.>, 19H8. depositor: nlembers.
144

the Financial Management Database, the In- tronic format is seen as a continuation of the
ternal Revenue Bulletin, and the Taxpayer In- current multi-tiered approach to disseminat-
formation Publications. ing government information: provision of in-
formation directly to the individual by govern-
One goal of the pilot and demonstration pro- ment, provision of information via the private
jects was to permit depository institutions ac- sector through a number of services, and pro-
cess to some agency data not previously avail- vision of information through the GPO docu-
able or data that were lost once converted to
ment sales program and the DLP. This three-
an electronic format. It would also open up the tiered approach recognizes that there are both
depository program to government informa- different markets and different users for this
tion in electronic form. The JCP passed a reso- information, and that these three modes of de-
lution on April 9, 1987, accepting the recommen- livery are not necessarily competitive and, in
dations of the Ad Hoc Committee in principle
many respects, are complementary.
and “urged” GPO to initiate pilot projects.
Despite the April 9, 1987 resolution, a fiscal Some database producers and services ob-
year 1987 funding request of $800,000 for the ject to the inclusion of electronic formats in
initial round of pilot projects was deferred by the depository program as proposed in the pi-
the Appropriations Committees of both the lot project program. The private sector posi-
House and Senate. GPO did create the Infor- tion is represented, in part, by the Informa-
mation Technology Program within LPS, with tion Industry Association (11A), a trade
internal funds, to prepare the depository pro- association with over 450 members from the
gram for electronic projects, gather informa- publishing and information sectors of the econ-
tion on Federal agency electronic programs, omy. These businesses employ information
and assist internal LPS operations. technologies to supply users, both public and
private, with all types of information. The 11A
The introduction of electronic formats to the has argued that provision of government in-
depository library program has been charac- formation in electronic format via depository
terized as: libraries, as proposed in the pilot project pro-
. . . an opportunity to make Government infor- gram, would compete with existing private sec-
mation useful and more timely, and, . . . an op- tor online services, and that, if electronic for-
portunity to achieve a higher level of service mats are included in the depository program,
to constituents.G2 they should be provided by private vendors.
This proposal has been endorsed by the Amer- The Association has further stated that the
depository program should comply with OMB
ican Library Association, Special Library
Association, American Association of Law Circular A-130 (though the legislative and ju-
dicial branches of government are not legally
Libraries, Association of Research Libraries,
National Coordinating Committee for the Pro- subject to A-130), and that the goals of the de-
motion of History, Medical Library Associa- pository program should be developed and re-
tion, Cartographic Users Advisory Council, viewed in much greater detail. Some members
of the 11A also contend that, if government
and others representing thousands of libraries
information in electronic format is dissemi-
around the country. Many depository librar-
nated through the depository program, private
ians also view the pilot projects as a chance
vendors will be unable to compete fairly and
to test a variety of electronic formats, and dis-
cover which one or combination of technologies will suffer adverse economic consequences.
and formats (electronic, paper, and microfiche)
are appropriate for different kinds information. Online Catalogs
Finally, provision of information in an elec-
Some government information is available
‘)zDiscussions, American Library Association Midwinter to depository libraries in electronic formats
Meeting, Chicago, II.., Jan. 16-19, 1987. through a number of private and not-for-profit
145
— ————-——— —— ———

database and vendor services, such as DIA- . . . the resources required to catalog items and
LOG, BRS, and OCLC, and the number of these to maintain card catalogs in even a moderate-
services is growing. The majority of the depos- sized institution are so extensive that libraries
itory libraries have access to at least one of the have frequently chosen not to catalog docu-
database systems, such as DIALOG or BRS, ments in order to contain these costs. f’-’
and the majority are also planning future on- Whereas previously the combination of tradi-
line catalogs.’” Since 1976, LPS use of OCLC tionally understaffed and low-budget docu-
allows depository institutions and others to ment departments could not afford the enor-
search OCLC and other online services for gov- mous cost of cataloging the materials, new
ernment documents for cataloging purposes, technologies now allow many to catalog both
for downloading into library catalogs, and as new and retrospective documents.
a limited reference tool.
There are a number of commercial services
LPS is the “center of authority” for the cat- available to libraries for cataloging of govern-
aloging of Federal documents (employing ac- ment documents, including retrospective ma-
cepted Anglo-American cataloging rules terials. For example, Marcive and Brodart pro-
[AACR2]), and is responsible for producing vide machine-readable tape, a microfiche
original cataloging records of Federal docu- catalog, or catalog card set records to deposi-
ments in a timely fashion. Once cataloged at tory libraries. The library identifies by a GPO
GPO, the record is available online immedi- item number those documents requiring a rec-
ately. Each week, OCLC sends the computer ord, and the vendors will supply the record in
tapes to GPO where they are consolidated by the desired format. Vendors are also provid-
the GPO Data System Service. Four computer ing this service for retrospective government
tapes are again consolidated to produce the documents. This type of service presents the
MonthluY Catalog of the Um”ted States Govern- participating institution with new opportun-
ment Publications. These GPO MARC tapes ities for creating online catalogs of Federal doc-
can be purchased from GPO and the Library uments, as these tapes can be loaded into a
of Congress by commercial firms and libraries. library’s local online catalog.’’”
As more and more libraries adopt informa- Some GPO cataloging records, particularly
tion technologies, the promise of online cata- from July 1976 to 1984 (when GPO began to
logs is particularly appealing for government include corrections made during the Monthl&v
document collections. It has been noted that, Catalog production process), contain errors
“three developments seem to have had the that have not been corrected.” GPO does not
widest impac~ on the overall effects of auto- generate retrospective corrections on the
mation in academic libraries: the growth and OCLC tapes for users, unlike the Library of
development of bibliographic utilities, the Congress and other Federal library institu-
changes brought about in information retrieval tions. The added expense to a library of iden-
by the use of online databases, and the more
recent development of online public access cat- 6sRoseann Bowerman and Susan Cad3, “Go\ernrnent P u b -
alogs. ‘f’f The 1981 Depository Library Bien- lications in an Online Catalog: A Feasibility Stud~,” fiotern-
nial Statistical Summary found that only 70 rnent Publications, December 1984, p. 331,
66Conversation with Judy hl~ers. op. cit., footnote 50.
depository libraries (or 6 percent of all deposi- ‘;TFor more information, see: LJudj E. Myers, ‘‘The (;o\ern-
tory libraries) catalog all government docu- ment Printing Office Cataloging Records: Opportunities and
ments received, while 666 depositories (or 56 Problems, ” Government Information Quarterl~r 2 ( 1985), pp.
27-56; Bowerman and Cad~’, op. cit.. footnote 65: Nlary Sue
percent) catalog less than one-tenth of items Stephenson and Gary Purcell, ” Current and Future Direction
selected. It has been noted that: of Automation Activities for t] ,S. (’, merriment Depvsi~ur~ Col-
lections, ” Government Information QuarterI.}r 3 ( 1986), pp. 191-
“‘Con\rersation with .Joseph hlcl’lane. Chief. 1.ibrar~ Inspec- 199; and Margaret ,Johnston Powell, Deborah Smith, and Ellen
tion Team, I.ibrar~r I]rograrns Ser\rice, U, S. (lo fernment Print- Conrad, “The Use of OCLC for Cataloging IJ. S. Go\’ernment
ing office>, Nm’ember 19H6. Publications, A Feasibility’ Studj, ’ (;ol’ernnlent Public ation.q
‘ ‘hloran, op. cit.. footnote 19, p. S. Re~riem’ ( 1987), pp. 61-76.
146
— —.. -—

tifying and correcting a record is quite high– percent increase in documents usage once
almost $4.50 per corrected record versus $1.40 records were included in the circulation
per high-quality record such as those produced system .70
today. For example, it would cost about $495,000 In conclusion, the availability of retrospec-
to examine, identify, and correct the 110,000 tive GPO cataloging tapes and private and not-
GPO cataloging records at the University of for-profit vendor services, combined with the
Houston in order to include these records in increasing number of technologies in deposi-
the online catalog. Error-free, the cost of in- tory institutions, permits these institutions to
clusion in the online catalog would be substan- catalog their government documents in a more
tially reduced to $154,000G8 cost-effective manner. This, in turn, increases
At those depositories where online catalogs access by patrons to government documents.
are being created and/or catalog records are In addition, it also allows these libraries to con-
being generated, government documents are sider machine-readable catalogs. The advent
becoming more accessible as cataloging of online catalogs in libraries in the next 5 to
records are now integrated into the main cat- 10 years will revolutionize government docu-
alog, and “. . . usage rates are going Up. As ment collections, as they will allow subject
early as 1984, Trinity University noted a 300 access to these materials by users utilizing elec-
tronic capabilities, and integrate the govern-
“ h Ibid, Judy E. Myers.
“ S ’Discussions, American Library Association Midwinter
ment information into the rest of the library
Meeting, Jan. 16-19, 1987. There are materials that are still not collection.
accessible through the program; GPO does not catalog all ma-
terials it distributes, such as the DOE materials, and there are
no plans for creating machine-readable records for those depos-
itor-y materials that predate 1976. 7“Bowerman and Cady, op. cit., footnote 65, p. 341.
Chapter 7

Alternative Futures for the


Depository Library Program

Photo credft ” Documents Center, Robed Woodruff Library, Emory University

Documents librarian assists users with the Documents Center online catalog.
CONTENTS
Page
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................149
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .........151
Alternative I: Status Quo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........151
Alternative II: Electronic Depository Library Program . ...............153
A Subalternative for Distributing Electronic Formats. . ..............157
Alternative III: Reorganized Electronic Depository System . ...........158
Disseminating Electronic Information Products–Two Case Studies . . . . ..160
C o n g r e s s i o n a l R e c o r d on CD-ROM . ...............................161
Federtd Re~”ster Online . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........164
Issues Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......171
Dissemination Formats in the Depository Program . .................171
Changing Costsof the Depository Program . .......................173
Reorganized Depository Program . . . . . . . . . . . . . . . . . . . . . ............175
Changing Rolesof Stakeholders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..176

Tables
Table Page
7-l. Estimated Costs Per Library Per Year for Distribution of
the Bound C’ongz-essional Record to Depository Libraries,
Various Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..............163
7-2 Estimated Costs Per Library Per Year for Distributionof the
Federa.lRe@ster, by Format . . . . . . . . . . . . . . . . . . . . . . . . . . .........167
7-3 Queries and Cost Data for Online Patent Information, 1987.........167
7-4 Estimated Reproduction and Distribution Costs, Per
Magnetic Tape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ....170
Chapter 7
Alternative Futures for the
Depository Library Program

SUMMARY
This chapter discusses several alternative brary considers . . . the expanded availability
futures for the U.S. Government Printing Of- of documents to those depository libraries as
fice (GPO) depository library program (DLP) absolutely essential if the purpose intended by
and examines their implications for the depos- Congress in the establishment of the original
itory program and users of Federal informa- program is to continue to be served.’” The
tion. The three alternatives considered are: debate today concerns not only additional ma-
terials but additional formats, and whether in-
● maintaining the status quo (that is, the
cluding electronic formats is consistent with
program as currently operating, exclud-
the legislative history and statutory author-
ing most electronic information products);
ity of the 1962 act.
● incorporating electronic information prod-
ucts into the current depository library With the increasing number of electronic dis-
program; and semination projects in agencies as well as re-
● reorganizing the depository program in lated private sector offerings, the impetus for
the 2- to 10-year time period, to accom- including electronic information in the depos-
modate electronic formats and the adop- itory program is strong. Electronic products
tion of current and emerging information enhance access to many types of information,
technologies by libraries.l and failure to include these products within
the depository library program could create
Included in this chapter are case studies of elec- or exacerbate inequities in public access to such
tronic delivery oft wo data files—the Congres- information. The Joint Committee on Print-
sional Record and Federal Register. This chap- ing (JCP) has interpreted provisions of Title
ter also discusses several key issues concerning 44 of the United States Code as extending to
the future of the DLP. government information in all formats and has
In 1962, Congress revised the laws relating endorsed pilot and demonstration tests and de-
to the depository program. Throughout the livery of government information products in
hearings and debate on the proposed legisla- electronic formats. The Subcommittee on Leg-
tion, Members and those testifying noted the islative, Committee on House Appropriations,
“vital role” of the depository program in the has approved the distribution of CD-ROMs to
dissemination of government information to depository libraries. Thus, it seems clear that
the American people. One of the revisions ac- some electronic products will be included in
cepted by Members was a provision to extend the depository program. However, the num-
to the depositories access to additional ber and types of products are likely to be quite
government-produced materials, and Members limited, in the absence of further congressional
noted that “. . . the Subcommittee on the Li- guidance, since at present GPO is not empow-
ered to require agencies to submit their elec-
1 Throughout this Chapter D PI. refers onl}r to the U.S. (~ ov- tronic products for depository distribution.
ernment Printing Office depository librar~.”program; and the Thus, without further policy action, erosion of
reorganized DLP alternative is based in part on the proposal
developed by the Association of Research Libraries Task Force
on Government Information in Electronic Format. For more
information see: Association of Research Libraries, Technol- IU. S. Congress, Committee on Rules and Administration,
ogy & U.S. Government Information Poh”cies: Catalwvsts for New Revising the I.a ws Relating to Depositor?’ Libraries. Senate Re-
Partnerships (J4’ashington, DC: ARL, October 1987). port No. 1587, 87th Cong., 2d sess., 1962,p. 25.

149
150
-—

public access to government information via the depository library participants. There will
the depository libraries is likely to continue. be a need to periodically review depository li-
brary policies as new formats are added, espe-
OTA has also concluded that the likely intro- cially since budgetary constraints are not likely
duction of electronic information to the deposi- to permit multiple formats for many govern-
tory library program may require an examina- ment information products. Difficult decisions
tion of the current structure of the program and will need to be made about which formats for
the nature of the relationships between deposi- the different products should be distributed
tory participants and the government. Commit- to depository libraries.
tees of Congress, member libraries, and GPO will
need to assess the current organizational struc- In the longer-term, Congress may wish to
ture and operation of the depository library pro- consider a reorganization or restructuring of
gram and determine if it is the most effective the current depository program in light of elec-
and efficient system for users to access govern- tronic information dissemination options now
ment information. Pilot projects and the like will or likely to become available. This assumes
provide useful information regarding user infor- that there is a fundamental need to reorganize
mation needs, financial costs, administrative re- the depository program to account for changes
quirements, and levels of usage. These pilots will in how users access information and how li-
also assist the committees, GPO, and member braries provide information to users. A reor-
institutions in designing new delivery and ganized depository program presumably would
financing arrangements, particularly in light incorporate the ‘lessons learned’ from the pi-
of the evolving nature of both libraries and the lot projects and demonstrations. To this end,
telecommunications infrastructure. the pilots and other resource-sharing projects
would assist depository librarians, GPO, the
An important reason for electronic demon- JCP, and other congressional committees in
stration projects is to better understand the discussing and redesigning the current struc-
issue of costs to users, to government, and to ture of the depository program to best serve
depository institutions. If the basic underlying the end-user—the public. Other issues, such
principle of the depository program is to retain as how best to serve the needs of rural users
free access to government information for users, of government information and whether the
then Congress needs to appreciate that there depository program should remain within GPO
may be additional costs associated with the in- or be moved elsewhere (e.g., to the Library of
troduction of certain electronic services, and Con- Congress), also could be examined during the
gress may have to assist depository libraries and reorganization discussions.
GPO in designing and financing ways to make
this information available to the public. Case In the final section of this chapter, four
studies of two electronic data files, the final closely related policy issues are examined.
or bound Congressional Record in CD-ROM These issues concern the need for developing
format and the Federal Re~”ster online, are pre a clear information policy on access to govern-
sented to illustrate possible delivery modes and ment information in electronic format through
costs. depository libraries. In formulating policy it
will be important to consider the following spe
Distribution of selected government infor- cific questions or issues:
mation products in CD-ROM format such as
the bound or final cumulated edition of the Con- ● Should government information in all elec-
gressional Record could improve access to such tronic formats be disseminated to the pub-
information, and could be a cost-effective dis- lic through the depository library program?
semination mechanism for the Library Pro- Are the principles of free access still appli-
grams Service (LPS) for certain data files. cable to the depository program, or are
There could be some additional equipment and there new costs associated with the intro-
training costs associated with this format for duction of electronically formatted prod-
151

ucts such that user fees or new funding ● Does the increasing shift to electronically
mechanisms need to be considered? formatted information require a reexami-
Can the current depository system accom- nation of the composition and relation-
modate new responsibilities for electronic ships of the stakeholders in the depository
formats or should a new institutional program, and especially the role of the in-
structure be considered? and formation industry?

INTRODUCTION
The first three major sections of this chap- 1. acquisition of materials;
ter explore three possible alternatives for the 2. classification of materials;
depository library program. These are main- 3. cataloging of materials;
taining the status quo, establishing an elec- 4. distribution of materials;
tronic depository library program, and devel- 5. monitoring of member depositories.
oping a long-term, reorganized electronic The discussion of monitoring includes consid-
depository library program based on new and eration of internal LPS operations and is pre-
emerging technologies and the changing na- sented from the perspective of the quality of
ture of libraries and information needs of users. library service and access to Federal infor-
The analysis of the alternatives and their mation.
possible implications is intended to facilitate The fourth section in this chapter discusses
an understanding of the possible choices avail- the possible dissemination of two data files to
able to policy makers. The three alternatives the depository libraries in electronic formats—
are presented and evaluated in some detail. In the bound Congressional Record in CD-ROM
evaluating the alternatives, the effects of pro- format, and the Federal Register online. Fi-
posed changes resulting from the use of new nally, the fifth and last section in this chapter
technologies are given considerable attention. discusses four key issues relevant to the fu-
Each of the alternatives is structured in ture of the DLP.
terms of the five main functions of the Library
Programs Service (LPS):

ALTERNATIVE I: STATUS QUO


The status quo is defined as a continuation DLP, within the Superintendent of Docu-
of the current roles and activities of the DLP ments (SupDocs) at GPO, would continue to
as described in chapter 6. The discussion be- distribute government documents to approx-
low assumes that no major congressional or imately 1,400 participating depository insti-
executive actions are taken for the next few tutions. The amount of government informa-
years and: tion that should belong in the program is
projected to increase, but the actual amount
● GPO disseminates information in paper distributed would probably decrease for two
and microfiche formats with a few CD- reasons—agencies failing to place paper doc-
ROMS and a few online files; and uments in the program (fugitive documents),
and an increase in the percentage of electronic
● depository libraries receive information products falling outside the program. The de-
from Federal agencies in paper and micro- centralized practice of agencies shipping ma-
fiche formats with few electronic formats. terials directly to participating depository in-
152

stitutions would likely increase, as in the case with a charter to initiate agency electronic pi-
of Department of Energy and Equal Employ- lot projects for the depository program. How-
ment Opportunity Commission agreements ever, no monies were appropriated by Congress
that are typical of arrangements between GPO, for this program. The program focus instead
libraries, and the agencies (see ch. 6 for more has been on internal operations, such as auto-
information). mated shipping lists, a claims-processing sys-
tem, and other microcomputer-based systems
Classification of materials. There would be in support of LPS operations. Additional staff
no changes in or effects on the classification
time has been spent gathering information on
of materials.
other agency electronic information programs
Cataloging of materials. There would be no and a few electronic projects such as the
changes in the cataloging of materials. Census Disk. Over the next few years, under
the status quo alternative, the role of ITP with
Distribution of materials. The bulk of gov-
respect to dissemination of electronic formats
ernment documents distributed to depository
would continue to be limited.
libraries would continue to be in microform for-
mat. Dual format (paper and microfiche) would The ability of LPS to accomplish its mission
continue for selected congressional and execu- would be eroded to the degree that:
tive branch materials. GPO and the library ● electronically-formatted government in-
community would likely revisit the debate over
formation was unavailable to the public
the choice and cost issues raised by dual
through the depository program;
format. ● the agencies became even more dependent
The Superintendent of Documents would on NTIS rather than GPO/SupDocs as a
maintain the practice of selling GPO tapes to dissemination mechanism for electronic
vendors at a nominal fee. Neither these tapes products; and
nor the bulk of electronically-formatted mate- ● the agencies relied on contractors, inter-
rials from other government agencies would agency agreements, or private sector ar-
be distributed to depository institutions (ex- rangements rather than GPO for elec-
cept for a few CD-ROMs and online products). tronic dissemination in general.
Depository institutions in need of electroni-
Although the number of selective depository
cally-formatted information would presumably institutions in the program would likely in-
purchase this information from vendors or
crease, several regional libraries (those libraries
through other arrangements directly with the
receiving and permanently maintaining all gov-
agencies, for example, Bureau of the Census
ernment documents) would likely drop mem-
or National Library of Medicine (NLM).
bership in the depository program (as is hap-
Monitoring of the member institutions and ef- pening currently). The increase in membership
fectiveness of the program. The budget of the of selective depositories would be due to the
LPS would likely remain relatively constant. minimum selection requirement that allows
If Federal agencies move away from GPO serv- participating institutions to select only those
ices (for whatever reasons, such as an increase government documents as appropriate for
in electronic products in lieu of paper) and, as their patrons. The decline in the number of re-
a consequence, fewer government documents gional depositories would be attributed to the
were available to the Sales Program, the GPO growing amount of government information
sales could be reduced. This could in turn af- that would need to be maintained permanently
fect the amount of monies redirected from net and the escalating costs of participation.
sales revenues to LPS to partially reduce the
As GPO’s role diminished, the role of the pri-
need for appropriated funds.
vate sector in the provision of government in-
The LPS Information Technology Program formation to the public would likely expand.
(ITP) was established in the summer of 1987 An increasing percentage of information in
153
. — .

electronic formats would be available only be more difficult for patrons to locate. To the
through private vendors. Depositories would degree that depository library users were de-
have two choices: either pay vendors for gov- nied effective access to enhanced electronic ver-
ernment information not available through the sions of core governmental process informa-
depository program or directly from agencies, tion such as the Congressional Record, equity
or not provide direct access to these materials of access would be further reduced. As stated
for their patrons.’ The costs to member de- by Frantzich:
positories would increase and continue to rise The current hard copy version of the Rec-
as agencies moved to greater reliance upon elec- ord is particularly inflexible. While users gen-
tronic formats and private sector services. The erally want a full picture of a debate on a par-
costs to those depositories opting to provide ticular subject or the actions of a particular
access would continually rise as agencies Member, these are scattered throughout the
moved to greater reliance upon electronic for- text and over a number of different docu-
mats and private sector services. Also, depos- ments. The ability to use new technology to
itories could incur increased costs for online “cut and paste” a tailored document would
searching and additional reference services to greatly increase the usability of the material
the extent that librarians and information in the Record. 4
specialists needed to check a greater number Under this alternative, overall government
of sources for government information. Region- costs would likely increase since government
als could face additional costs to the extent at all levels (including Federal as well as State
that selective depository institutions were un- and local depository libraries) would not be re-
able to provide specific information to patrons ceiving needed Federal information through
and as a result referred inquiries to the regional the depository program and would have to
libraries. maintain it through other more expensive
Under the status quo alternative, public ac- means.
cess to government information via depository Under the status quo, with a greater num-
libraries would likely be continually eroded and ber of agency information products available
reduced. Equity of access would be adversely in electronic formats, GPO would be unable
affected to the extent that patrons of deposi- to comply fully with a legislative mandate of
tory institutions would have to pay for access providing access to government information
to government information. Also, many of the to the public through the depository library
agency electronic information products could program.
‘Not all government information in electronic format would
be available through private sector services, since some or many
types of information would not be expected to produce mone- ‘Stephen Frantzich, “Public Access to Congressional Infor-
tary benefits for vendors. This information could be permanently mation in the Technological Age: Case Studies. Draft OTA
lost to the public. contractor report, OTA, September 1987, pp. 50-51.

ALTERNATIVE II: ELECTRONIC DEPOSITORY


LIBRARY PROGRAM
This alternative assumes that the existing ● each depository would select the type and
DLP would be extended to include government number of formats; and
information products in electronic formats as
well as paper and microfiche. The program ● OMB WOLIM issue a circular requiring
would be managed as it is now. In addition, agencies to comply with the depository
this alternative assumes that GPO would serve program for all government information
as the disseminating agency for the depository regardless of format (within current ac-
program, and: cepted guidelines for those materials that
154

are not confidential, administrative, or for it might be more advantageous for GPO to ship
internal use). the CD-ROMs directly to the depositories.
Under this alternative, the Superintendent Classification of materials. The introduction
of Documents would approve the inclusion of of electronically formatted materials should
electronic products from Federal agencies for not require any significant changes in current
dissemination to depository institutions, in LPS classification procedures. The format type
addition to paper and microfiche products. Un- –paper, microfiche, CD-ROM disk, diskette,
der this alternative, it is likely that microfiche or online tape—would need to be noted as it
would still account for the bulk of products is now. It would be necessary for the originat-
within the program. Dual format would prob- ing agencies to clearly define the source and
ably continue for a minimum number of prod- nature of the electronic material so it can be
ucts, and fewer paper documents would be properly classified and assigned a correct
available to member institutions. LPS would number.
be able to provide a choice of files in a variety
Cataloging of materials. GPO employs
of formats to members of the program, but
AACR-2 (Anglo-American Cataloging Rules),
these files would not always be available in all
the accepted standard for cataloging developed
formats.
by the library community. The cataloging of
This alternative could have the following ef- new media is already an accepted practice in
fects on the five major functions of LPS. the library community. GPO’s integration of
new media into the depository program would
Acquisition of materials. Many of the current
procedures for acquiring materials would con- require training of LPS cataloging staff and
informational assistance to depository insti-
tinue. GPO would receive tapes, disks, or dis-
kettes from the issuing agency, duplicate the tutions to make library catalogers aware of the
new media in-house or via a contractor, or de- changes in format. GPO has developed Cata-
loging Guidelines that describe preferred rou-
positories would receive the materials directly
tines for inputting records into the Online Com-
from the agency or an agency contractor pur-
puter Library Center (OCLC), use of AACR-2,
suant to an interagency agreement.
serials procedures, and the like. GPO would
The information format would require indi- be required to update these guidelines to in-
vidual decisions by the agencies and GPO to clude procedures for handling electronic
determine if GPO would operate in a centrali- formats.
zed or decentralized mode. For example, a data
Distribution of new materials. Overall, the
file developed by or for an agency could be re-
current distribution procedures would remain
ceived by the depository via GPO, from the
with some modification for materials in an on-
issuing agency directly, or from an agency con-
line format. Diskettes and CD-ROMS would
tractor. This data tape could also be a “raw”
present few, if any, new distribution problems
data tape or one with “value-added” software
to LPS. However, distribution of online data
included. Each mode of delivery to the deposi-
files could present a variety of problems, de-
tories and each format has advantages and dis-
advantages to the program, the participating pending on whether the mode of operation was
institutions, and the agencies. Thus, it could centralized or decentralized.
be more advantageous for libraries to work The addition of electronic formats might af-
directly with the Bureau of the Census for cer- fect the selectivity of the depositories. Depos-
tain data tapes. This could require additional itory institutions are becoming increasingly
infrastructure within agencies where the selective in the number of and kind of govern-
amount of electronically formatted informa- ment information products they receive. The
tion is significant. Another example would be inclusion of electronic products in the program
CD-ROMs of the bound Congressional Record would not change this trend, and might even
or of Bureau of the Census materials, where increase selectivity. As with paper and micro-
155

fiche, librarians would need to examine the gov- of Federal Information Users noted a grow-
ernment materials available in electronic for- ing technology base in depository institutions.
mat and explore the choices for their- patrons. If GPO were to provide electronically format-
Whereas the initial cost of adding a document ted materials, it is likely that more and more
in microfiche is minimal (not counting the costs depositories would, over time, invest in needed
of storage and maintenance), the cost of equip- equipment just as they did for microfiche.
ment and software development for electronic
Each depository institution would be in a
formats could be higher, at least initially. With
different stage of development concerning the
electronic formats, depositories would have
use of information technologies in support of
new choices to make concerning the informa-
depository library programs. For example,
tion needs of users. many university libraries already have CD-
Monitoring of member institutions and effec- ROM equipment, whereas many smaller librar-
tiveness of the program (as it relates to qual- ies do not. On the other hand, CD-ROM tech-
ity of service and access). The introduction of nology is inexpensive, very user-friendly, and
electronic files to the depository program and requires minimal staff and user training. It
to GPO would require the development of ad- does, though, require some training and knowl-
ditional GPO in-house expertise in information edge in order to use different software pack-
technologies. For example, GPO could contract ages effectively.
for the mastering and production of CD-ROMs, A determination would need to be made on
produce CD-ROMs in-house (equipment per- what kind and level of support GPO should
mitting), or obtain the necessary number of provide including, costs and other implications.
CD-ROMs from the agencies. Regardless of the For example, GPO could offer the depository
choice, an overall understanding of CD-ROM libraries a series of comprehensive seminars
technology, production, and use would be on equipment and training, and/or form a team
needed to ensure an effective program. of information technology consultants similar
The Information Technology Program (ITP) to the current depository inspectors. This team
would need to be expanded. The new respon- would assist member institutions introduce
sibilities of this office could include develop- electronic formats to the library staff and
ment of training programs for depository li- users.
braries that focus on equipment purchases, use Congress could consider a one-time equip-
of new electronic services, and awareness of ment grant for depository library CD-ROM
electronic information products available from equipment but would need to address stand-
the government. This training component would ardization issues. Depository participants and
be critical to the success of this alternative, and GPO are likely to be unable to regularly up-
would require increased funding. grade their CD-ROMs (for financial constraints
alone), yet, at the same time, CD-ROM tech-
Overall, large institutions such as the Asso- nologies are constantly changing (both hard-
ciation of Research Libraries (ARL) members, ware and software capabilities). One possible
academic institutions, and State libraries side effect of an equipment grant, if made,
(which collectively account for over 50 per cent would be to encourage and accelerate standard-
of the depositories) would be better prepared
setting, since the government would be buy-
than smaller institutions to accept products ing for up to 1,400 institutions.
in electronic format. These institutions already
have much of the needed equipment and on- Some reconsideration of retention policies
going training programs for staff. Acceptance wou[d be necessary with the introduction of
of electronic products probably would be electronic files. These new policies would be
slower at smaller institutions, primarily due applicable to regional depositories that are now
to lack of necessary equipment, training, and required to retain all government materials per-
an adequate financial base. The GAO Survey manently. There are over 50 regional institu-
156
-.

tions—a mix of public and private institutions There would be some reduction in other cur-
and academic, public, and State libraries. Stor- rent costs, such as for storage of paper and
age guidelines for new formats at these insti- microfiche. For example, the cost of microfiche
tutions, particularly for online files, would be cabinets is very high-$3,000 per year, includ-
an issue if the old requirements were retained. ing maintenance and space considerations—
The two key issues that would need to be ad- and the reduction in the amount of microfiche
dressed would be the development of guide- could be a benefit to regional depository insti-
lines for online storage, and the development tutions.
of guidelines regarding the costs to regionals
for provision of online information to other in- All depository institutions that accept elec-
stitutions. tronically formatted products might face ad-
ditional costs from participating in the pro-
A new institutional structure for the depos- gram. These costs would result from one or
itory program is evolving with the emergence more of the following: 1) staff training, 2) equip-
of a set of “supra” regionals. The role of some ment purchase (beyond that possibly provided
regional depositories has developed into one by GPO, 3) costs of online searching, 4) local
of providing service support to other deposi- mounting and/ or downloading of government
tories, including staff time and equipment. information, and 5) increased user support. The
Also, some regional institutions do not house specific electronic format would affect the level
all government materials received, but instead of costs to the user, the library, or the govern-
assume responsibility for these documents ment. For example, providing the Congres-
regardless of the location. This change has sional Record retrospectively on a CD-ROM
occurred due to increased cooperation among disk to all depository institutions (if mastered
members, with “statewide” institutions ex- by GPO, replicated by a contractor, and dis-
panding their participation. There are a num- tributed by the depository program), would
ber of advantages to this emerging structure: likely impose some additional financial respon-
● increased integration of government doc- sibilities on depositories and actually reduce
uments into library collections, costs to the government if provided in lieu of
● greater resource sharing, paper or microfiche. However, access to an
● spreading out the “burden” of the serv- agency online data file might involve additional
ice support functions and costs, and costs to the depositories and/or the govern-
● improved flexibility of storage re- ment when telecommunication costs are fac-
quirements. tored in.
This growth in cooperation and flexibility Some regional depositories would face addi-
within the depository system is very impor- tional financial burdens, and some depositories
tant and should be beneficial as the amount might reconsider membership as the costs of
of electronic information increases. 5 resource-sharing increased. An increasing
number of depositories not receiving certain
Other impacts of Alternative II. Under Alter- electronic files might turn to the regional de-
native II, there could be substantial savings pository libraries for that information. If this
in GPO production and distribution costs if occurred, it could be difficult for these regional
more government information products were libraries to maintain policies of free access to
available in CD-ROM format and not produced government information.
in paper and microfiche (e.g., for certain Bu-
reau of the Census materials). There could be Under alternative II, overall access to gov-
additional costs incurred depending on the ernment information would be expected to im-
number of products distributed in an online prove. But access would be enhanced only if
format as this format can be more expensive. depository libraries could accommodate and
support electronic formats and develop re-
“Based on discussions with GPO, LPS staff, November 1987. source-sharing procedures for those libraries
157

that did not have the necessary technologies, ● GPO would need to design criteria for
funding, and training within a State or region. selection of library participants. Telecom-
munications permit access to online infor-
This alternative would increase the need for
mation regardless of geographic location,
clarification of the roles and legal authorities
so geographical concerns need not limit
of GPO, the depositories, and the agencies.
the libraries selected. The type of library,
Agencies presumably would be required to pro-
the facilities, equipment and training pro-
vide electronically formatted information to
grams in the library, and the networking
the depository program. Clear guidelines and
capabilities to other depositories are ex-
possibly legislation would be necessary.
amples of criteria or factors that could be
employed by GPO to select participants.
A Subalternative for Distributing ● GPO would need to specify the responsi-
Electronic Formats bility of participating libraries regarding
the need to maintain paper and microfiche
A subalternative of Alternative II would
copies of these data files to guarantee ac-
make selected electronic products available to
cess to government information, and for
specified depositories not via GPO but via an
archival purposes.
agreement with private sector or not-for-profit
● GPO would need some assurance (on be-
services. The Public Printer has previously pro- half of all participants) of length and level
posed a pilot project whereby private sector of commitment by the vendor, and the
and not-for-profit services would disseminate vendor would require some assurance as
selected government generated tapes to a sub- to the commitment by GPO to the con-
set of depository libraries at little or no cost
tinuation of this program. For example,
to the government.h The private sector serv- would this be a pilot project or a program
ice would add value to five suggested data- that would seek congressional endorse-
bases: congressional bills, Congress~onal Rec- ment and financial support for, say, the
ord, Federal Register, Code of Federal next 3 years?
Regulations, and the Monthly Catalog of ● For an option such as this to be success-
Um”ted States Government Publications, in re- ful, the vendor would likely already have
turn for free receipt of the tapes. The private
considerable market share in one or sev-
sector service would be required to accept all eral of the data files and within the library
five tapes, because the total value of the five community.
tapes would provide increased access for users ● The value of the duplicated tapes and the
and increased value to the vendor. The cost “charge” against the depository program
of the tapes would be debited to the deposi- would need to be factored into the overall
tory program. costs of the program.
For this or similar proposals to be seriously A determination of proprietary rights in
considered, a number of issues would require the ‘value added’ information and ensu-
clarification. For example: ing use by libraries would be necessary.
● Previous “barter” agreements between
● GPO would need to determine a level of agencies and private sector services gen-
public access to the electronic information erally have been unsuccessful, and con-
that would be considered viable and appr~ gressional committees with relevant juris-
priate. Would the combined value of the diction generally have not supported
tapes minus the va.badded costs provide projects of this nature. Congressional con-
a sufficient level of access to the public
cerns would need to be alleviated prior to
and sufficient incentive to the vendors? implementation of such a proposal.
Arrangements concerning telecommuni-
61.etter from Ralph E. Kennickell, Jr., Public Printer, to
Honorable Frank Annunzio, Chairman, tJoint Committee on cations charges and the like would need
Printing, Dec. 10, 1987. to be specified.
158
.—— —

Under this subalternative, public access to rights in the value added by vendors would
government information would appear to in- need to be addressed. Should or could these
crease. Electronic information would be avail- rights be waived, or should restrictions on pub-
able to the depositories at little or no immedi- lic use of such value added information be ac-
ate cost to participating institutions. Some cepted? Any restrictions could adversely af-
insight would be gained concerning usage fect the ability of libraries to share this
levels and the overall costs that would be asso- information with other depositories and users.
ciated with a larger electronic program. There Overall, this subalternative would change the
would be minimal costs to government, at least current relationships between the government
at the outset. and the depository libraries. Congress, GPO,
and the libraries would need to consider care-
There are also disadvantages associated with fully the implications of including new “part-
this subaltemative. With the pilot project, only ners’ in the depository library program. Fi-
a selected number of libraries would partici- nally, barter-type arrangements such as this
pate and even those would have only a mini- suba.lternative have not been successful in
mum level of access. It is unclear whether and other agencies, for example, the Patent and
under what conditions this suba.lternative Trademark Office (PTO) and the Securities and
could be extended to all depository libraries. Exchange Commission (SEC). Another possi-
The costs to the government while minimal bility would be for the government to simply
with the pilot project would increase substan- pay the vendor for services rendered at a bulk
tially over time as the number and types of discount rate.
files expanded. The question of proprietary

ALTERNATIVE III: REORGANIZED ELECTRONIC


DEPOSITORY SYSTEM
This alternative presents one of many pos- ● Basic Service–libraries with small gov-
sible future directions for the depository pro- ernment document collections and gate-
gram once electronic pilots have commenced way access to electronic information lo-
and user needs, usage patterns, and cost fac- cated elsewhere. Basic service would be
tors are more fully understood. This alterna- characterized by self-help, on-demand
tive seeks to describe a longer-term reorganized service, and high cost per transaction, but
view of the LPS incorporating current and small fixed cost.
emerging technologies and reflecting the ● Intermediate Service—libraries with a
changing roles of libraries. The composition larger government document collection,
of participating libraries could be reviewed and including some electronic information and
reorganized, consistent with meeting user gateway access to other electronic infor-
needs and optimizing use of resources. This mation located elsewhere. Intermediate
alternative is based on the recent Association service would be characterized by some
of Research Libraries (ARL) proposal for res- vah.wadded information development and
tructuring the depository program. This pro- increased mediation between information
posal has been put forth by the ARL Task resources and information users.
Force on Government Information in Elec- ● Full Service—libraries with a complete
tronic Format for discussion purposes only. government document collections and a
The ARL proposal defines a three-tiered sys- full range of electronic information and
tem of libraries and library responsibilities. gateways to other resources. Full service
Three new levels of service would be desig- would be characterized by support from
nated: Basic, Intermediate, and Full: “related, locally available databases, ”
159

value-added services, development of soft- ated by a member institution. LPS would then,
ware packages and similar “approaches in turn, notify other participants in the pro-
which would change wholesale Govern- gram (e.g., through Administrative Notes) that
ment information into retail Government the products were available. The depository
information, higher fixed costs, and institutions could also rely on other networks
lower per transaction costs.7 and bulletin boards to convey this information.
Under this alternative, LPS would continue Cataloging of Materials. The cataloging dis-
to distribute government information to par- cussion found in the previous section would
ticipating institutions. Electronic products apply to this reorganized electronic option.
from Federal agencies would be included in the
Distribution of New Materials. In addition to
depository program (and the Monthly Cata-
the distribution mechanisms discussed in the
log of United States Government Publications)
previous section, a core collection of materi-
in addition to paper and microfiche products.
als for basic and intermediate levels of service
Dual format would continue for a minimum
would need to be developed by LPS in concert
number of products. Microfiche would prob-
with the depository institutions. Distribution
ably continue to account for the bulk of mate- systems or new resource-sharing procedures
rials in the program. LPS would be able to pro- between basic, intermediate, and full service
vide a choice of files in a number of formats,
libraries would need to be developed by par-
but these files would not always be available
ticipating institutions and the LPS. These pro-
in all formats. The full-resource institutions cedures would include a description of the
would be assuming many new responsibilities, responsibilities of each service level, financial
for example, local mounting of tape files. The obligations, interlibrary loan procedures, and
new focus would be on the ability to access in- the like. ITP could be responsible for assist-
formation as required from a host of available ing intermediate-and full-service institutions
resources. With the reorganized structure LPS:
with new technological applications, and pro-
would not require the same amount of viding current information on activities in Fed-
printed or microfiche products from GPO; eral agencies, such as the development of new
would need to expand the functions oft he electronic information products and appli-
ITP; cations.
would determine with depository institu- Monitoring of member institutions and effec-
tions the “core” collection for basic and tiveness of the program (as it relates to qual-
intermediate services; and ity of service and access). Most of the new tasks
would work closely with depository mem- noted in the previous discussion of Alterna-
bers in developing criteria and infrastruc- tive II would apply here as well—the expan-
ture for the reorganized system. sion of ITP to assist libraries in choice of for-
Acquisition of Materials. Most of the current mats, technological applications, and liaison
procedures for acquisition of materials would with other Federal agencies; new retention pol-
remain in effect. The discussion in Alternative icies for electronic formats; and the possible
II would apply to this alternative. provision of CD-ROM equipment. Overall, pol-
icies for bibliographic searching, cooperative
Classification of Materials. The previous dis- acquisition, interlibrary loan, document deliv-
cussion of Alternative I I would apply to this ery services, reciprocal borrowing privileges,
reorganized electronic alternative. LPS and the referral and reference services, and the stor-
intermediate-and full-service depositories could age and preservation of government materi-
consider the value of devising a system where- als would need to be modified or created.
by LPS would be notified of any value-added
products, software products, or the like cre- The establishment of a new infrastructure
for the depository program would probably re-
‘Association of Research Libraries. op. cit., footnote 1, p. 22. quire changes in the monitoring responsibili-
160

ties of LPS. Depository members and GPO to- Other impacts of Alternative III. It is diffi-
gether would need to define the goals and cult to determine if there would be savings to
objectives of the new system, define the government under this alternative, without
responsibilities of each level of service, and de- detailed cost-benefit studies. Reducing distri-
fine the responsibilities of GPO and an over- bution of paper and microfiche would save
all framework for monitoring performance of money. However, there would be transition
the depository program. The current system costs as well as new equipment and training
employed in a number of regions, whereby re- costs (e.g. resulting in the shifting of funds
gional depositories have assumed some respon- from distribution functions to ITP within
sibilities for the level of service in their region LPS).
or State, might be applicable in the new sys-
tem. In this instance, full service libraries would Access to government information would be
with intermediate libraries assist new libraries improved under this alternative. A reorganized
wishing to join the depository system and electronic program would: 1) facilitate access
would regularly evaluate the services needed to print-based materials and electronic infor-
and those already provided for the region. mation, 2) expand and improve access to a host
of online information services and products,
The depository institutions would need to and 3) encourage a new level of sophisticated
consider carefully which level of service under manipulation of information electronically
the reorganized system would best serve their (government and nongovernment information).
organization and patrons. There is a wide var- The reorganized structure would permit an in-
iance in technological sophistication among the formation network to develop among deposi-
libraries in the current depository system. The tories, allowing for increased efficiency and ac-
same variance would be evident in a reorgan- cess to information resources on a national,
ized system, and many libraries would not be State, and local scale.
capable of providing “gateway” services with-
out guidance and support from affiliated de- This alternative would increase the need for
pository members. The reorganized structure clarification of the roles and legal authorities
would likely streamline the current program of GPO, the depositories, and the agencies.
and permit the development of a network or Agencies presumably would be required to pro-
system of depository institutions, recognizing vide electronically formatted information to
that there is a need for different levels of the depository program. Clear guidelines and
service. possibly legislation would be necessary.

DISSEMINATING ELECTRONIC INFORMATION


PRODUCTS–TWO CASE STUDIES
The previous sections examined three pos- Register online. These files were selected for
sible alternatives for the depository library several reasons. First, the Congressional Rec-
program: maintenance of the status quo, an ord and the Federal Register represent core
electronic depository library program, and a or process government information. Second,
reorganized electronic depository library sys- these files: have been identified by depository
tern. This section will examine two formats (on- librarians as useful and/or desirable in elec-
line and CD-ROM) and delivery mechanisms tronic format; are extremely popular with high
for specific government data files as test or regular usage; and are files found in most
projects for disseminating electronic formats depositories. Third, providing these files in
through the DLP. The data files described are electronic formats clearly improves and en-
the Congressional Record in CD-ROM format hances public access, and in some cases time-
for the bound, cumulated file, and the Federal liness, compared to paper and microfiche for-
mats. The bound or final Congressional Record to 72 hours after the printing of each Record.
in CD-ROM format and the Federal Register A number of vendors acquire these tapes from
online could be made available to depositories the GPO, add value to the existing version,
as described in Alternatives II or III. Finally, and sell this enhanced information to clients.
the JCP recently announced that the bound A yearly subscription to the Congressional
Record will be available on CD-ROM through Record tape service costs $29,300, and each
GPO. tape can be purchased for $175. Microfiche co-
pies of the Record are produced by a GPO con-
Congressional Record on CD-ROM tractor and are then distributed by GPO. These
microfiche copies are not available as quickly
In the 1983 Ad Hoc Committee on Deposi- as either the paper or electronic formats.
tory Library Access to Federal Automated
Databases survey of depository institutions, The Congressional Record is recorrected and
depository librarians identified the Congres- new printing plates are created to produce the
sional Record as a key data file which, if avail- bound, permanent copy or final edition of the
able in electronic format, would enhance ac- Congressional Record. The bound Record is a
cess by patrons to government information. number of years behind. The most recent
The Congressional Record is received by most bound volumes published cover 1982 (volume
depositories, is currently available in paper or 128) and 1985 (volume 131), with current ef-
microfiche (dual format) from GPO, and is forts focused on 1983, 1984, and 1986. The
available online for a fee through several com- most current index available is for 1980. The
mercial vendors. 1981 index is in production and expected in
1988; the 1982 index is scheduled for comple-
The Congressional Record contains the daily tion in late 1988. The cumulated, final, bound
record of House and Senate floor proceedings Congressional Record represents the only cor-
as well as schedules of other congressional rected edition of the Record and is important
activities and actions. A new Record i s for archival, historical, and sometimes legal
produced nightly and is available to the pub- purposes. (For more information on the issues
lic the following morning. It has been stated relating to the Record, see ch. 8.)
that, “a distinguishing feature of the Record
is its timeliness."8 GPO receives scheduling
Bound Congressional Record on CD-ROM
information, prepared remarks and inserts
from Members, floor debate transcripts, bill There are a number of possible options for
texts, and other documents and melds this ma- mastering and replicating a CD-ROM disk of
terial into a 200-to 300-page document every the bound Record; for example, by GPO, by
night that Congress is in session. a commercial vendor, or by a combination of
the two. Several new internal production steps
The material is accepted by GPO in numer- will be necessary to produce a disk. Once com-
ous formats (electronic, printed, and hand- plete, the yearly cumulative Congressional
written) which are then entered in the data- Record on CD-ROM, produced either by GPO
base by GPO staff. This new electronic ver- or a contractor, could then be shipped to the
sion is used to produce the printing plates for depositories through normal distribution chan-
the printing of the Congressional Record in nels. The disk could also be available through
hard copy.9 The electronic database in the the Superintendent of Documents for a fee (the
form of magnetic computer tapes is corrected usual cost plus 50 per cent).
and then made available for purchase through
the Superintendent of Documents within 24 The corrected daily Record tapes produced
by the GPO Office of Information Resources
‘Frantzich, op. cit., footnote 4, Management will be the digitized data used
‘Due to time constraints of the printing process, errors in the
electronic tapes are not corrected immediately. For more infor- for the creation of the CD-ROM. GPO man-
mation see Frantzich, op. cit., footnote 4, p. 35. agement is currently considering the lease of
162

a CD Publisher system that is capable of recon- • delivery mode—including format, equip-


figuring (reindexing) a file structure and pre- ment needs, training needs, etc.; and
paring the file for one or more disks. This file • costs—including startup, equipment,
would then be ready for a contractor to mas- staff, operational needs, etc.ll
ter and replicate CD-ROMs for GPO distribu-
Data requirements. Under the current guide-
tion to the depository libraries and/or sale lines, GPO only offers to depositories the mi-
through the Superintendent of Documents.10
crofiche format with a paper index of the bound
GPO management has determined that GPO
final Congressional Record. If the CD-ROM
staff will not develop the needed retrieval soft-
bound Record were available, libraries could
ware itself, but will purchase the software from
choose among the two formats (microfiche or
a vendor. Producing the software on CD-ROM
CD-ROM) for a limited amount of transition
may eliminate the need for a separate index time. This transition time would provide data
to the Record because of the search and re-
to determine user preferences regarding the
trieval capabilities inherent in CD-ROM soft-
format of the bound or final Record. (A limited
ware. On the other hand, it is also argued that
number of printed copies would be available
there may be the need for both the Congres- for purchase through the Superintendent of
sional Record Index and the CD-ROM search Documents.) The bound Congressional Record
and retrieval software because the index pro- serves as an important historical, archival, and
vides additional reference points and “human
legal tool. Member institutions would need to
judgment” not found in the software. determine their institutions’ needs regarding
The average amount of data in the Record access (printed or electronic) and transition
per year is: 37,594 pages representing over 500 time between different formats if switching
million bytes of information (for 1985 as a sam- from paper and microfiche to CD-ROM or mi-
ple year). These figures do not include an in- crofiche to CD-ROM. Many libraries may be
dex to the bound version. GPO is considering in the position of housing paper, microfiche,
many possibilities. Two under consideration and electronic versions of the Record for ar-
are: one year, one volume of the bound version chival purposes.12 Many of the same transi-
of the Record, without the index, plus re- tion issues addressed in the late 1970s and
quired/necessary software on one CD-ROM; 1980s, as libraries incorporated microfiche into
and one year of the Record on one CD-ROM, their collections, would apply here.
plus a floppy disk that would contain the soft- There is no agreement on the longevity of
ware for accessing and manipulating the data
optical disks, with estimates ranging from as
residing on the CD-ROM. Because GPO has
little as 10 years to as long as 100 years. Also,
not developed such a disk before, staff are un-
although CD-ROMs may endure for up to sev-
certain as to the amount of data that can fit
eral decades or longer, the equipment used to
on the disk and what constitutes the “best”
“read” these products may quickly become
approach.
outdated. Format longevity is important for
Certain crosscutting criteria can be applied archival purposes because one goal of some de-
to each data file and delivery mode to describe positories is to provide a continuous and com-
and present the opportunities and drawbacks plete record of government information. The
of each format option. These criteria are: importance of maintaining a usable and com-
plete Record file reflects several needs—
• data requirements —including complete-
ness, size, and use of data, timeliness, etc.;
11 These criteria are based in part on questions proposed by
10 At this time, GPO does not intend to master and replicate the ARL Task Force on Government Information in Electronic
CD-ROM products. GPO believes that it would not be cost- Format for use in evaluating pilot projects; Report No. 1, App.
effective for the agency to invest in such equipment or neces- 1, Oct. 30, 1986, Draft No. 1.
12
sary manpower at this time. If the need for and use of CD-ROM GP0 could begin production of CD-ROM formats begin-
products by the Federal Government increased radically and ning with the 1983 bound Record. However, GPO would be un-
requires substantial production capabilities, then GPO would able at this time to retrospectively convert earlier (pre-1983)
reconsider its position. Record tapes to CD-ROM products.
163

historical research, research on a political po- trieval of information would increase the use
sition, and, increasingly, determination of of the Record.
legislative intent by the courts, agencies, law-
yers, and others. Delivery and costs. There would be few, if
any, new requirements or equipment needs for
Timeliness is not a critical issue for the LPS to deliver this information in CD-ROM
bound Record, though use of the CD-ROM for- to the depositories. As noted in Table 7-1 the
mat probably would reduce the current back- estimated per-library cost for provision of the
log. As with the replication of microfiche, GPO bound Congressional Record is $632.83 for pa-
would rely on private sector contractors to per format; $33.74 for a hard copy of the in-
master and replicate the disks. The schedules dex of the Record; $83.62 for a microfiche copy;
and reliability of the firms chosen as well as and $10.05 for a CD-ROM plus floppy disk (one
GPO contract specifications would, in many of two possibilities under investigation). If
respects, determine the turnaround time from GPO used commercial access software with the
GPO to the depositories. disk, there might be an additional software
The availability of the bound Record in a CD- license fee, although it would likely be mini-
ROM format would enhance and improve ac- mal. According to GPO, the overall cost of pro-
cess by users to those files. The number of ducing the microfiche master of the bound Rec-
users simultaneously using Congressional Rec- ord is $5,047.50, and the estimated production
ord information would not necessarily increase, cost of the CD-ROM master for the bound Rec-
but ease of access to the file would increase ord is $1,700. GPO would not require supple-
mental funding to produce the CD-ROM for
dramatically. This would be particularly true
when compared to the microfiche format. In the bound Record, if this were the only format
addition, by its very nature, indexing would produced.
be built into the disk file, whereas with micro- However, member depository libraries would
fiche there is a separate index (still maintained need to assess their CD-ROM information ac-
in paper for congressional and depository use), cess and equipment needs. The GAO Survey
and searching is more cumbersome and time- of Federal Information Users found that over
consuming. The CD-ROM format would in- 40 per cent of those surveyed have a CD-ROM
clude print-on-demand capabilities similar to player or access to one. Those libraries with-
those in use today for microfiche reader/ out CD-ROM players would need to invest
printers. It has been noted that: about $600-$700 per player. The GAO survey
A major limitation of using the Record in also found that 283 of 451 depository libraries
its current form is the limited indexing and have (or have access to) a microcomputer with-
the difficulty of finding materials. Whereas, out a modem, and 337 of the 451 have a micro-
the ability to create new subsets of data makes
Table 7-1 .—Estimated Costs Per Library Per Year
an electronic database very powerful and much for Distribution of the Bound Congressional/ Record
more valuable than a paper catalog. Searches to Depository Libraries, Various Formats
of the database become easier, faster, cheaper,
and more thorough. 13 Paper Paper Microfiche CD-ROM
Copies Index Copies Copies
In general, the bound Record is not one of the Printing Cost ... . $569,70$30.30 – –
most heavily used items in a depository, but Production Costs — — $28.27 –
it is one that 1,305 of the 1,393 libraries main- Duplication Costs
tain and one that is used by patrons. It is ex- CD-ROM ... ... – – – $3.00
pected that improved indexing and easier re- Floppy Disk — — — $5.00
Postage . . . . . . . . . . . $55,30 $3.13 $.85 $1.49
Handling . . . . . . . . . $7,83 $.31 $54,50 $.06
} ‘Frmtzich, op. cit., footnote 4, p. 42 and, Stephen Frantzich, Documentation . . . — — — $.50
“Public Access to Congressional Information: The Potential and
Pitfalls of Technology I?nhanced Access” OTA contractor re- Total . . , . . . . . . . . $632.83 $33,74 $83.62 $10,05
port, January 1987, p. 17. SOURCE: U.S. Government Printing Office 1987
164

computer with a modem for online access However, this equipment would be used for
(many libraries have more than one microcom- numerous tasks and many information prod-
puter). Those libraries not having a microcom- ucts, not just the Record.
puter, or not having adequate access if the
equipment is located elsewhere, would need to In summary, there appear to be numerous
invest in a microcomputer as well, at a cost advantages to using the CD-ROM format for
of about $1,200 to $1,400. If CD-ROM becomes the bound Congressional Record:
a major format for depository distribution, ●
the large textual database lends itself to
many libraries may wish to invest in a com- the CD-ROM format;
plete CD-ROM system (player, microcomputer, ● the information is not current data and,
and printer, at a total cost about $2,500-$3,000 therefore, does not require regular up-
per system) for dedicated use. dating;
GPO/LPS training needs would be rather
● the efficiency and ease of access to the in-
minimal. The LPS training role could be formation would improve with this for-
directed toward assisting member libraries mat, compared to either paper or micro-
choose equipment, providing or developing ad- fiche products;
ditional software applications, and arranging
● library shelving needs would be reduced;
training seminars for participating library
● there could be substantial cost savings for
staff. the GPO/Library Programs Service, de-
pending on the format options; and
Depository library training requirements ● for some libraries, the ability to combine
would be greater. Libraries would need to pro- the historical data on disk- and current
vide both hands-on training sessions for staff data online would present exciting new ac-
and at least minimal assistance to users. The cess possibilities and potential.
amount of training and assistance required
would depend, in part, on the software pack- The disadvantages of adopting the CD-ROM
age provided or developed by GPO and its ease format would be:
of use. Libraries that have provided some user ●
the need for some libraries to purchase one
CD-ROM training and instruction note im- or more pieces of equipment;
provements in user capabilities and search- ● the need to provide physical space for CD-
strategy success.14 ROM work stations for microcomputers,
An important consideration with the intro- printers, and CD-ROM players; and
duction of any service is to factor in, as well ● finally, the need for some or many libraries
as possible, the life cycle costs. The shift to to maintain collections of the Record in
a CD-ROM format for the bound Record could paper, microfiche, and CD-ROM formats.
result in a three-format collection for many
institutions (for archival and preservation pur-
poses): 1) maintenance of paper format for cur- Federal Register Online
rent information, 2) microfiche for the retro- The Federal Register is one of the core or
spective collection, and 3) CD-ROM for the process documents included in the collections
bound Record. The combination of formats of most depository institutions. The Federal
would require the use of different equipment Register is a dual format item (available in mi-
and possible upgrading of equipment (particu- crofiche or paper from the GPO), and is avail-
larly for CD-ROM players), all with associated able online (all or parts thereof) through sev-
purchase, lease, and/or maintenance costs. eral commercial services for a fee. The Code
of Federal Regulations (CFR) is also available
14 For more information see K.J. Pearce, ‘‘CD-ROM: caveat
in CD-ROM format from VLSOPTEXT. VLS
Emptor, Library Journal, vol. 113, No. 2, Feb. 1, 1988, pp. 37- plans to offer a combination CFR and Federal
38; and Linda Stewart and Jan Olsen, “Compact Disk Data-
bases: Are They Good For Users?, Online, vol. 12, No. 3, May Register on disk quarterly with “seamless” ac-
1988, pp. 48-52. cess to an online Federal Register file.
165

The Federal Register is a daily publication Frantzich has noted that:


of the government that documents executive
branch regulations (proposed and final), It is unreasonable to expect individuals and
presidential directives, meetings, and policies organizations to comply with the rules and reg-
ulations of government without timely access
(proposed and final). The classes of documents to the relevant details. A prime purpose of the
found in the Register are grouped under four Federal Register is to solicit comments and
headings or categories: inform the interested public about meetings
on proposed regulations.16
1. the President’s section consisting of ex-
The Federal Register is regularly cited by
ecutive orders, proclamations, and other
depository librarians as a key document that
presidential documents;
is needed on an up-to-date basis; 1,040 libraries
2. rules and regulations, which include the
receive paper copies of the Register and 363
administrative actions pursuant to stat-
receive microfiche copies, both via LPS. In the
utory law;
GAO survey, depository librarians identified
3. proposed rules, that provide an avenue
the Federal Register online as one of the most
for notification of new rulemaking and for
useful electronic services that could be pro-
interested parties to comment on draft
vided. The Federal Register has been described
rules; and
as one item received by depositories that can
4. notices, which include miscellaneous
“never arrive soon enough. ” If not received
agency material, advisory activities and
in paper format, it is clearly less useful in mi-
opinions, meetings, and the like. 15
crofiche due to the time lag and the inherent
limitations of the microfiche format.
Like the Congressional Record, the Federal
Register is produced daily by GPO, and an elec- LPS requests that depository members re-
tronic database is created by GPO for use in tain at least the current and previous year’s
the printing process. Also, like the Record, the editions of the Federal Register on file. Mem-
hard copy of the Federal Register takes prece- ber libraries also retain the current year of the
dence over both electronic and microfiche ver- Code of Federal Regulations (except for Title
sions. The microfiche version is replicated (by 3). Much of the pertinent material printed in
a GPO contractor) and distributed 24 hours the daily Register is eventually included in the
following the printing, and the corrected elec- Code of Federal Regulations. Some regional
tronic tapes are available up to 72 hours fol- depository libraries keep retrospective micro-
lowing the hard copy release. Final corrections fiche collections of the Federal Register. Use
are made by GPO in the electronic database of these back files has been described as mini-
during a lull in the printing process. These data mal due, in part, to the difficulty in using the
tapes, once corrected, can be purchased on a microfiche format.
yearly subscription basis for $37,500, or on a
daily basis for $175 per tape from the Superin-
Federal Register Online Delivery
tendent of Documents. The daily Register con-
tains an index, and a cumulated index is pro- If the Federal Register were to be provided on-
duced monthly. Indexing of the Reqister is line to depository libraries, there are several
automated. The average number of pages per possible delivery options:
year in the Federal Register is 52,000, repre-
senting 416 million bytes of information, in- Option 1: Centralized delivery. Depositories
cluding the GPO printing codes. would have direct access to the Federal
Register data file maintained by GPO,
with GPO providing minimal value-added
enhancements to the basic data and with
15 Frantzich, “Public Access to Executive Agency Informa-
tion in the Technological Age: Case Studies, OTA contractor
report, February 1988, p. 8. 16 Frantzich, Ibid.

<
“ , — ‘- . , -
166

libraries using commercial dial-up tele- able within hours of the printed Register. 18
communication lines;17 GPO would need to determine how much data
Option 2: Decentralized delivery. GPO would to maintain online-for example, the past year
duplicate and provide Federal Register or two of the Register. Users would be required
computer tapes to a select number of de- to use paper or microfiche copies of the Regis-
positories; these depositories would, in ter for certain dated materials-for example,
turn, locally mount the data and make the those more than 6 months or a year old—
information available online to participat- instead of relying on the online file.
ing libraries in a designated region; and Data requirements. Online access to the Fed-
Option 3: Subscription basis. Depository eral Register would greatly improve and en-
libraries would access the Federal Regis- hance access to and timeliness of the informa-
ter data file via a commercial or not-for- tion for patrons. Receipt of the Register in a
profit vendor with a subscription subsidy timely fashion is one requirement of its use.
(full or partial) provided by GPO. Unlike direct access with CD-ROM technology,
Each of these options will be considered in online access could require a trained informa-
terms of the same criteria applied in the previ- tion specialist. Although there are user-friendly
ous discussion of the Record on CD-ROM: data software packages available, the telecommu-
requirements, delivery, and costs. The bulk of nication costs associated with online access can
the information presented in the discussion of be high, and these costs could be reduced if
centralized delivery applies to the discussions a trained librarian performed the search.
of decentralized and subscription delivery. Although access to the information in the
Choice of these three delivery options for dis- Federal Register file would be improved, un-
cussion does not preclude other possible op- less the libraries have high speed modems,
tions. It is important to note that decisions users’ searches would likely be limited and the
concerning the Federal Register are made by information would be downloaded and printed
the Office of the Federal Register (OFR). De- offline. This adds an additional step to access-
cisions relating to format and dissemination ing the information.
are determined by the Administrative Com-
mittee of the Federal Register, whose mem-