Professional Documents
Culture Documents
DICTIONARY
of
COMPUTER SCIENCE
Revised Edition
The Facts On File
DICTIONARY
of
COMPUTER SCIENCE
Revised Edition
Edited by
John Daintith
Edmund Wright
The Facts On File Dictionary of Computer Science
Revised Edition
All rights reserved. No part of this book may be reproduced or utilized in any
form or by any means, electronic or mechanical, including photocopying,
recording, or by any information storage or retrieval systems, without
permission in writing from the publisher. For information contact:
Facts On File books are available at special discounts when purchased in bulk
quantities for businesses, associations, institutions, or sales promotions. Please call
our Special Sales Department in New York at (212) 967-8800 or (800) 322-8755.
MP 10 9 8 7 6 5 4 3 2 1
This dictionary is one of a series designed for use in schools. It is intended for stu-
dents of computer science, but we hope that it will also be helpful to other science
students and to anyone interested in computers. Facts On File also publishes dictio-
naries in a variety of disciplines, including astronomy, chemistry, biology, earth sci-
ence, physics, mathematics, forensic science, weather and climate, and marine
science.
The first edition of this book was published in 2001. This second edition has been
extensively revised and extended and now contains over 2,450 headwords covering
the terminology of computing. A totally new feature of this edition is the inclusion
of over 500 pronunciations for terms that are not in everyday use. A number of ap-
pendixes have been included covering mathematical symbols and notation, file ex-
tensions, domain names, and numerical conversions. There is also a list of Web sites
and a bibliography. A guide to using the dictionary has also been added to this lat-
est version of the book.
We would like to thank all the people who have cooperated in producing this book.
A list of contributors is given on the acknowledgments page. We are also grateful to
the many people who have given additional help and advice.
v
ACKNOWLEDGMENTS
Contributors
Pronunciations
vi
CONTENTS
Preface v
Acknowledgments vi
Pronunciation Key x
Entries A to Z 1
Appendixes
Symbols and Notation
Arithmetic and algebra 251
Geometry and trigonometry 252
Sets and logic 253
Calculus 254
File Extensions 255
Organizational Top-Level 258
Domain Names
Geographic Top-Level 259
Domain Names
United States Subdomain Names 264
Number Conversions 265
Bibliography 273
vii
GUIDE TO USING THE DICTIONARY
Headwords
The main term being defined is in bold type:
Variants
Sometimes a word has a synonym or alternative spelling. This is placed in brackets after
the headword, and is also in bold type:
Here, ‘bistable’ is another term for ‘flip-flop’. Generally, the entry for the synonym con-
sists of a simple cross-reference:
Abbreviations
Abbreviations for terms are treated in the same way as variants:
In these cases, the entry for the full form consists of a reference:
viii
Multiple definitions
Some terms have two or more distinct senses. These are numbered in bold type
Cross-references
These are references within an entry to other entries that may give additional useful in-
formation. Cross-references are indicated in two ways. When the word appears in the de-
finition, it is printed in small capitals:
Hidden entries
Sometimes it is convenient to define one term within the entry for another term:
Here, ‘script’ is a hidden entry under ‘scripting language’, and is indicated by italic type.
The individual entries consists of a simple cross-reference:
Pronunciations
Where appropriate pronunciations are indicated immediately after the headword, en-
closed in forward slashes:
Note that simple words in everyday language are not given pronunciations. Also head-
words that are two-word phrases do not have pronunciations if the component words are
pronounced elsewhere in the dictionary.
ix
Pronunciation Key
/a/ as in back /bak/, active /ak-tiv/ /nk/ as in rank /rank/, bronchus /bronk-ŭs/
/ă/ as in abduct /ăb-dukt/, gamma /gam-ă/ /o/ as in pot /pot/
/ah/ as in palm /pahm/, father /fah-ther/, /ô/ as in dog /dôg/
/air/ as in care /kair/, aerospace /air-ŏ- /o/ as in buttock /but-ŏk/
spays/ /oh/ as in home /hohm/, post /pohst/
/ar/ as in tar /tar/, starfish /star-fish/, heart /oi/ as in boil /boil/
/hart/ /oo/ as in food /food/, croup /kroop/, fluke
/aw/ as in jaw /jaw/, gall /gawl/, taut /tawt/ /flook/
/ay/ as in mania /may-niă/ ,grey /gray/ /oor/ as in pruritus /proor-ÿ-tis/
/b/ as in bed /bed/ /or/ as in organ /or-găn/, wart /wort/
/ch/ as in chin /chin/ /ow/ as in powder /pow-der/, pouch
/d/ as in day /day/ /powch/
/e/ as in red /red/ /p/ as in pill /pil/
/ĕ/ as in bowel /bow-ĕl/ /r/ as in rib /rib/
/ee/ as in see /see/, haem /heem/, caffeine /s/ as in skin /skin/, cell /sel/
//kaf-een/,/ baby /bay-bee/ /sh/ as in shock /shok/, action /ak-shŏn/
/eer/ as in fear /feer/, serum /seer-ŭm/ /t/ as in tone /tohn/
/er/ as in dermal /der-măl/, labour /lay-ber/ /th/ as in thin /thin/, stealth /stelth/
/ew/ as in dew /dew/, nucleus /new-klee-ŭs/ /th/ as in then /then/, bathe /bayth/
/ewr/ as in epidural /ep-i-dewr-ăl/ /u/ as in pulp /pulp/, blood /blud/
/f/ as in fat /fat/, phobia /foh-biă/, rough /ŭ/ as in typhus /tÿ-fŭs/
/ruf/ /û/ as in pull /pûl/, hook /hûk/
/g/ as in gag /gag/ /v/ as in vein /vayn/
/h/ as in hip /hip/ /w/ as in wind /wind/
/i/ as in fit /fit/, reduction /ri-duk-shăn/ /y/ as in yeast /yeest/
/j/ as in jaw /jaw/, gene /jeen/, ridge /rij/ /ÿ/ as in bite /bÿt/, high /hÿ/, hyperfine /hÿ-
/k/ as in kidney /kid-nee/, chlorine /klor- per-fÿn/
een/, crisis /krÿ-sis/ /yoo/ as in unit /yoo-nit/, formula /form-
/ks/ as in toxic /toks-ik/ yoo-lă/
/kw/ as in quadrate /kwod-rayt/ /yoor/ as in pure /pyoor/, ureter /yoor-ee-
/l/ as in liver /liv-er/, seal /seel/ ter/
/m/ as in milk /milk/ /ÿr/ as in fire /fÿr/
/n/ as in nit /nit/ /z/ as in zinc /zink/, glucose /gloo-kohz/
/ng/ as in sing /sing/ /zh/ as in vision /vizh-ŏn/
x
A
abend /ă-bend/ The premature stopping combination of keys that can be used to
of a program caused by a program error, carry out a specified function. For exam-
system failure, power failure, etc. The term ple, ctrl+V (pressing the control key and
is a contraction of ‘abnormal end’. See also the letter V) is a method of pasting data in
abort. many applications.
1
account
2
A/D converter
pler by cable. Acoustic couplers have now tor display; TFT display) A high-quality
been superseded by MODEMS. color display using LCD technology. For-
merly used only in laptop and notepad
Acrobat Trademark A collection of pro- computers, active matrix displays are now
grams developed by Adobe Systems, Inc. found in flat panel displays for desktop
for reproduction of printed documents in computers.
their original form and for electronic pub-
lications. The system involves conversion Active Server Pages See ASP.
of a formatted file (from a word processor,
DTP application, etc.) into a portable doc- ActiveX /ak-tă-veks/ Trademark A de-
ument format (PDF) file. This file format velopment by Microsoft of a loosely de-
can be read by the application Acrobat fined collection of technologies based on
Reader, which is freely available. The PDF COM used to enable different software com-
file contains all the formatting and fonts of ponents to interact. The most common
the original document and is particularly example is ActiveX controls, which are
useful for electronic transmission or publi- separately developed components that add
cation of documents in which layout is im- functionality to programs or Web pages.
portant (e.g. official publications, legal Examples range from reusable program
documents, etc.). PDF files can be gener- components, such as calendars, clocks,
ated directly from many applications. They and various types of data entry/display
can also be produced from PostScript files tools, to complex special effects for Web
using the application Acrobat Distiller. pages.
PDF files can also have bookmarks and hy-
perlinks embedded and can be indexed. See Ada /ay-dă/ Trademark A programming
also PostScript. language named after Ada, Lady Lovelace,
the mathematician and sometime assistant
active Operating or being used. For ex- to Charles Babbage. Ada was developed as
ample, the active cell in a spreadsheet is the a result of a competition for a new pro-
one selected at a particular time; the active gramming language organized by the US
window in an application is the one af- Department of Defense in the late 1970s. It
fected by input from the keyboard or cur- was designed to be used in EMBEDDED sys-
sor. tems, i.e. in situations where a computer
forms part of a specialized system such as a
Active Directory Trademark A method cruise missile or a manufacturing plant.
of organizing a Windows network. Every Ada is a large high-level language embody-
object on the network – users, groups of ing concepts developed in earlier languages
users, printers, disk DIRECTORIES, network such as PASCAL, but carrying them much
services (e.g. e-mail) – can be arranged in a further. There is now an international stan-
logical structure that controls which ob- dard for Ada.
jects can interact with which other objects
and in what way: for example which users adapter A printed circuit board inserted
can access the contents of a given disk into a computer in order to enable it to in-
directory, which users can manage which terface with a peripheral device, such as a
other users, etc. This logical structure modem or sound card.
need not match the physical structure of
the network. If the sales department of Adaptive Server Enterprise See SQL
a large company is spread across mul- Server.
tiple sites and so multiple LANs, Active Di-
rectory can be configured to make it ap- ADC See A/D converter.
pear as a single unit to the rest of the
network. A/D converter (analog to digital con-
verter; ADC) A device for sampling an
active matrix display (thin film transis- ANALOG SIGNAL and converting it into an
3
adder
equivalent DIGITAL SIGNAL, i.e. for convert- address 1. A number used to specify a
ing a continuously varying signal, such as a LOCATION within computer memory, usu-
voltage, into a series of binary values that ally to allow storage or retrieval of an item
are suitable for use by a computer. See also of information at this location. The words
D/A converter. ‘address’ and ‘location’ are actually used as
synonyms, as in ordinary speech. In back-
ing store an address identifies a specific
SECTOR on a magnetic disk or a specific
A 0110 BLOCK on a magnetic tape. In main store an
B 0100 address identifies a particular location
holding a WORD or a BYTE. When a location
S 1010
is specified it is said to be addressed. In
C 0100
some computers, processor REGISTERS
and/or INPUT/OUTPUT DEVICES can also be
Binary addition addressed.
The reference to an address appears in a
adder An electronic device – a LOGIC CIR- MACHINE INSTRUCTION. An instruction is
CUIT – that adds together two numbers. It normally made up of a combination of an
can also subtract two numbers using two’s OPERATION CODE and some way of specify-
complement arithmetic (see complement). ing the OPERAND (or operands) upon which
The numbers are represented as binary sig- the operation is to be performed. This is
nals that are fed into the device (see digital commonly done by specifying the MACHINE
signal). The binary output represents the ADDRESS of the operand. The process is
sum of the numbers. known as DIRECT ADDRESSING.
When two binary (or decimal) numbers Direct addressing is one of a variety of
are added, each step of the addition gener- processes by which the storage location to
ates a sum digit, which forms part of the be accessed is determined from the ma-
answer, and a carry digit, which is passed chine instruction. Alternative ADDRESSING
to the next step. In the table, S is the sum MODES are used in cases where the machine
and C the carry resulting from the addition address is too large to be comfortably in-
of two binary numbers A and B, equal to cluded in a machine instruction, or where it
0110 (i.e. 6) and 0100 (i.e. 4). This is a type is not necessary or not possible to assign an
of TRUTH TABLE. explicit address.
Within the adder, addition is performed 2. The means used to specify the location
electronically on corresponding pairs of of a participant in a computer NETWORK; in
bits (the two least significant bits, the two particular, the sequence of characters used
next least significant bits, etc.). The cir- to specify the destination of an e-mail.
cuitry involved in this must be able not
only to add a particular pair of bits to pro- address bus See bus.
duce a sum bit but also to add the carry bit
from the previous step in the addition. The addressing mode Any of various meth-
sum bits together form the final answer ods by which an ADDRESS can be specified
that is output from the device. The final in the address part of a MACHINE INSTRUC-
carry bit is discarded or is used to detect TION. The methods include DIRECT AD-
OVERFLOW. DRESSING, IMMEDIATE ADDRESSING, INDIRECT
ADDRESSING, INDEXED ADDRESSING, and REL-
add-in card See expansion card. ATIVE ADDRESSING, all of which are used in
computer systems. Whatever method is
add-on 1. A device that is external to a used, the computer’s hardware applies ap-
computer and plugged into it in order to in- propriate rules to derive an ABSOLUTE AD-
crease its capabilities. DRESS. See table overleaf.
2. A small program that enhances a larger
one. address part A part of a MACHINE IN-
4
Advanced Research Project Agency Network
ADDRESSING MODES
STRUCTION that usually contains only an therefore important that their passwords
ADDRESS, or information on where to find are known only to authorized people.
or how to construct an address. There may
be more than one address part in an in- Adobe Type Manager /ă-doh-bee/
struction. In a JUMP INSTRUCTION one of the (ATM) Software produced by Adobe
addresses may specify the location of the Systems, Inc. used to control PostScript
next instruction to be executed. FONTS.
5
Advanced Technology Attachment
6
AND operation
7
animation
8
arithmetic operator
9
arithmetic shift
ARITHMETIC OPERATORS
Operators
Operation
In general use In Basic
addition + +
subtraction – –
multiplication × or . *
division ÷ or / /
exponentiation 52 5↑2
would thus be written as 7*2 and 6/3 in array refers to the set of storage LOCATIONS
most high-level languages. Some languages in which the data items are placed, keeping
have separate operators for integer division their original arrangement.
and remaindering (see integer arithmetic). The set of locations forming an array is
Some languages do not have an operator referenced by a single IDENTIFIER, chosen by
for EXPONENTIATION. the programmer. Each element in an array
An example of how arithmetic opera- (i.e. a location or its contents) can be spec-
tors are used, in Basic, is as follows: ified by combining one or more subscript
D = SQR((X(1) – X(0)) ↑ 2 + X(2)*3) values with the identifier. Subscripts are
This is equivalent to usually integers and are generally placed in
D = √((X1 – X0)2 + 3X2) brackets after the identifier. The number of
subscripts required to specify an element
arithmetic shift See shift. gives the dimension of the array.
The simplest array is a single sequence
arithmetic unit See arithmetic and logic of elements. This is a one-dimensional
unit. array, only one subscript being necessary
to select a particular element. For example,
ARP See address resolution protocol. a list of people’s ages could form an array
named AGE; if the subscript of the first el-
ARPANET /ar-pă-net/ (Advanced Re- ement is 1, the age of the eighth person in
search Project Agency Network) A net- the list is found by specifying AGE(8). The
work introduced in 1969 connecting a subscript may be a VARIABLE or an EXPRES-
small number of research institutions, SION. A one-dimensional array is also
funded by the US Department of Defense. known as a vector.
ARPANET used PACKET SWITCHING and pi- In a two-dimensional array (also called
oneered many of the protocols used in net- a matrix), the elements are arranged in the
work operation. It was the forerunner of form of a table with a fixed number of
the INTERNET. rows and a fixed number of columns. Each
element is distinguished by a pair of sub-
array One form in which a collection of scripts; the first subscript gives the row
related data items can be stored in com- number, the second gives the column num-
puter memory. The data items in an array ber. For example, A(3,7) refers to the el-
are arranged in a particular order or pat- ement in row three and column seven of
tern and are all of the same type, for exam- the array A. Again, the subscripts may be
ple all integers or all real numbers. This variables or expressions.
collection of data items is referred to as an The values of a subscript range from a
array. More usually, however, the word lower limit (usually 1 or 0 unless otherwise
10
assembler
specified) to an upper limit. These limits organisms. Computer simulations are used
specify the total number of elements in an to show how self-replicating organisms
array, and are called bounds. The bounds might behave.
of an array can be declared in various
ways, depending on the programming lan- ascending order The arrangement of a
guage. In Basic, for example, a DIMEN- set of data in order starting with the lowest
SION statement is used: character value, as in A to Z or 1 to 50.
DIM X(4,10)
This is a declaration of a two-dimensional ASCII /as-kee/ American National Stan-
array, X, with 5 rows and 11 columns, dard Code for Information Interchange, a
since the default lower bound in Basic is standard code for the interchange of infor-
zero. mation between computer systems, data
communication systems, and associated
arrow keys Four keys on a KEYBOARD equipment. Since it is a standard code
that are labeled with up, down, left, and (rather than one developed by a particular
right arrow symbols. The arrow keys can manufacturer), it allows equipment of dif-
be used for control of the CURSOR on a dis- ferent manufacturers to exchange informa-
play screen. tion. It is thus widely used. ASCII encoding
produces coded characters of 7 bits, and
artificial intelligence (AI) The branch hence provides 27, i.e. 128, distinct bit
of computer science concerned with pro- patterns. These 128 characters make up
grams that carry out tasks requiring intelli- the ASCII CHARACTER SET: they consist of
gence when done by humans. Many of ALPHANUMERIC CHARACTERS, the SPACE
these tasks involve a lot more computation CHARACTER, SPECIAL CHARACTERS, and
than is immediately apparent because CONTROL CHARACTERS. The character set
much of the computation is unconscious in and the control characters are explained
humans, making it hard to simulate. Pro- in the table overleaf. The binary encodings
grams now exist that play chess and other shown in the table are for the character
games at the highest level, take decisions b7b6b5b4b3b2b1
based on available evidence, prove theo- Thus the encoding for G is 1000111 and
rems in certain branches of mathematics, the encoding for g is 1100111. ASCII be-
recognize connected speech using a limited came the de facto standard character set
vocabulary, and use television cameras to for most computers and has been incorpo-
recognize objects. Although these exam- rated in such larger character sets as UNI-
ples sound impressive, the programs have CODE. The coding from values 129 to 256
limited ability, no creativity, and each can is sometimes called extended ASCII. See
only carry out a limited range of tasks. also ISO-7.
There is still a lot more research to be done
before the ultimate goal of artificial intelli- ASP (Active Server Pages) A SCRIPTING
gence is achieved, which is to understand LANGUAGE developed by Microsoft for
intelligence well enough to make comput- SERVER-SIDE SCRIPTING. ASP was first re-
ers more intelligent than people. In fact leased in 1996 and the latest version,
there is considerable controversy about the ASP.NET, is part of the .NET software-de-
whole subject, with many people arguing velopment platform.
that the human thought process is different
in kind to the computational operation of aspect ratio The ratio of width to height
computer processes and so cannot be re- for a rectangular shape, such as an illustra-
produced using computers. See also expert tion or window.
system; robot.
assembler A program that takes as
artificial life A branch of ARTIFICIAL-IN- input a program written in ASSEMBLY LAN-
TELLIGENCE research concerned with inves- GUAGE and translates it into MACHINE CODE.
tigations into the behavior of living Each instruction in assembly language is
11
ASCII CHARACTER SET
b7 0 0 0 0 1 1 1 1
b6 0 0 1 1 0 0 1 1
b5 0 1 0 1 0 1 0 1
b4 b3 b2 b1
0 0 0 0 NUL DLE space 0 @ P ` p
0 0 0 1 SOH DC1 ! 1 A Q a q
0 0 1 0 STX DC2 ″ 2 B R b r
0 0 1 1 ETX DC3 # 3 C S c s
0 1 0 0 EOT DC4 $ 4 D T d t
0 1 0 1 ENQ NAK % 5 E U e u
0 1 1 0 ACK SYN & 6 F V f v
0 1 1 1 BEL ETB ′ 7 G W g w
1 0 0 0 BS CAN ( 8 H X h x
1 0 0 1 HT EM ) 9 I Y i y
1 0 1 0 LF SUB * : J Z j z
1 0 1 1 VT ESC + ; K [ k {
1 1 0 0 FF FS , < L \ l |
1 1 0 1 CR GS – = M ] m }
1 1 1 0 SO RS . > N ^ n ~
1 1 1 1 SI US / ? O — o DEL
12
asynchronous transmission
usually converted into one machine in- will thus cause a new value to be placed at
struction. The input to the assembler is the appropriate storage location(s). The as-
called the source program; the output is signment statement is fundamental to most
called an OBJECT MODULE. The translation programming languages. It is indicated by
process is known as assembly and the pro- a special symbol, such as = or :=. This is
gram that is translated is said to have been called the assignment operator. The sym-
assembled. The entire program must be as- bol used depends on the language.
sembled before it can be executed. See also The expression on the right of the as-
compiler; interpreter. signment operator is given to the variable
whose name appears on the left. If, say, the
assembly language A type of PRO- value of a variable C is to be increased by
GRAMMING LANGUAGE that is a readable one, this is expressed in Basic, Fortran, C,
and convenient notation (in human terms) and Java as
for representing programs in MACHINE C=C+1
CODE. Assembly language was originally and in Pascal and Algol as
devised to alleviate the tedious and time- C := C + 1;
consuming task of writing programs actu- and is read as C becomes C + 1. In a pro-
ally in machine code. It may be used gram FLOWCHART the symbol → is often
nowadays, in preference to a HIGH-LEVEL used to indicate an assignment, as in
LANGUAGE, for reasons of speed or com- C→C+1
pactness.
Assembly language is the most com- associative store (content-addressable
monly used LOW-LEVEL LANGUAGE. Differ- store) A storage device in which a LOCA-
ent forms are developed for different TION is identified by what is in it rather
computers, usually by the computer manu- than by its position. The location contains
facturers. The features of a particular as- a particular item of data – a search word –
sembly language thus reflect the facilities for which a search can be conducted by the
of the machine on which it is employed. storage device. This can be achieved in var-
Each MACHINE INSTRUCTION has a corre- ious ways. The desired data in the location
sponding assembly language instruction. is in close association or proximity to the
The programmer can use alphabetic OPER- search word. Associative stores are used,
ATION CODES with mnemonic significance for example, as part of a VIRTUAL STORAGE
(e.g. LDA is commonly used for ‘load ac- facility.
cumulator’), and can use symbolic names,
personally chosen, for ADDRESSES of storage asymmetric digital subscriber line /ay-
locations or registers (see symbolic ad- să-met-rik/ See ADSL.
dressing). Addressing modes can also be
expressed in a convenient way. In addition asynchronous /ay-sink-rŏ-nŭs/ Involv-
assembly language allows the use of vari- ing or requiring a form of timing control in
ous number systems (e.g. DECIMAL, OCTAL, which a specific operation is begun on re-
HEXADECIMAL NOTATION) for numerical ceipt of a signal to indicate that the preced-
constants, and allows the programmer to ing operation has been completed.
attach LABELS to lines of the program. Operations do not therefore occur at regu-
A program known as an ASSEMBLER lar or predictable times. Compare synchro-
translates the assembly language program nous.
into machine code. It can then be executed
by the computer. asynchronous transmission A form of
transmission in which data is sent as it be-
assignment statement A program comes available. The time at which the
STATEMENT that assigns a new value to a start of transmission occurs is arbitrary, al-
VARIABLE. Each variable is associated with though the rate at which the bits compris-
a particular LOCATION or group of loca- ing the data are subsequently transmitted is
tions in memory. An assignment statement fixed.
13
ATA
ATA (Advanced Technology Attach- trail may detect attempts at violating the
ment) A standard INTERFACE used to con- security of the system by, say, unautho-
nect disk drives and other storage devices rized reading or writing of data in a file.
to a computer. Although in use from the
mid-1980s the first standard, ATA-1, was availability The actual AVAILABLE TIME
only approved in 1994; the latest version is expressed as a percentage of planned avail-
ATA-7. Early versions of ATA were more able time. Any shortfall will be the result of
widely known as IDE or EIDE. ATA is also faults or breakdowns. In a modern com-
known as Parallel ATA (PATA) to distin- puter system an availability that is not in
guish it from SERIAL ATA, which has super-
the high nineties (say less than 98%) is
seded it in most new machines.
cause for concern.
Atlas See second generation computers.
available time The amount of time in a
ATM See Adobe Type Manager. given period that a computer system can be
used by its normal users. During this pe-
attachment See e-mail. riod the system must be functioning cor-
rectly, have power supplied to it, and not
audit trail A record showing the occur- be undergoing repair or maintenance.
rence of specified events in a computer sys-
tem. For example, an entry might be made AVI (audio video interleave) A Win-
in an audit trail whenever a user logs on or dows standard format for audio and video.
accesses a file. Examination of the audit Files of this type have the avi file extension.
14
B
back end 1. A program that is used to The time taken to retrieve a particular
convert output from a general-purpose item from backing store and transfer it di-
program into a specific form. For example, rectly into main store must be brief. RAN-
a graphics program might produce picture DOM ACCESS to these items is thus necessary
information in a quite general form, which rather than SERIAL ACCESS. MAGNETIC DISKS,
is then processed by a back end specific to CDS, and DVDS provide random access and
the sort of output device required. There are extensively used for on-line backing
would be a different back end for each de- storage. MAGNETIC TAPES are serial-access
vice – plotter, VDU, graphics printer, etc. devices and are used as off-line backing
2. See engine. store. Programs and data can be trans-
ferred from tape to disk for processing, or
background printing The printing of a can be copied from disks to tape.
document at the same time as other
processes are being carried out. backslash The \ character on a key-
board. In MS-DOS and Windows systems
background processing The process- the backslash is used to represent the root
ing of computer programs with a low pri- directory and also to separate directory
ority in a way that does not allow names and filenames in a pathname.
interaction with the user through the key-
board. Programs run in this way are called backup A file, device, or system that can
background programs and are processed at be used as a substitute in the event of a loss
times when programs with a higher prior- of data, development of a fault, etc. A
ity are not using the computer. See also backup file, for example, is a copy of a file
foreground processing. that is taken in case the original is uninten-
tionally altered or destroyed and the data
backing store (backing storage; sec- lost. It is not stored ON-LINE. When a copy
ondary store) Storage devices in which is made it is said to back up the original
programs and data are kept when not re- version. Such a backup is a type of DUMP.
quired by the PROCESSOR (or processors) of
a computer. Programs can only be exe- Backus-Naur form /bak-ŭs nor/ See
cuted, however, when they are in the com- BNF.
puter’s MAIN STORE. A program plus
associated data is therefore copied from backward compatibility (downward
backing store into main store when re- compatibility) Compatibility with earlier
quired. The storage CAPACITY of main store versions or models of the same product. A
is much smaller than the capacity available program is backward compatible if the
as backing store. Only the programs that files and data created by previous versions
are currently being executed are (or need to of the program can be used; an operating
be) in main store. The use of backing store system is backward compatible if existing
means that a computer has access to a large application programs continue to operate
repertoire of programs and to large correctly. A computer is backward com-
amounts of data, and can call on these as patible if it can run the same applications
and when it needs them. and operating systems as the previous
15
bad sector
model. Sometimes, as new technology or other digits give the manufacturer code
techniques are developed, backward com- and identify the product.
patibility cannot be totally maintained,
software and hardware both become obso- barrel printer (drum printer) See line
lete, and the user has to upgrade the sys- printer.
tem. See also upward compatibility.
base (radix) The number of distinct dig-
bad sector A SECTOR on a MAGNETIC its (and possibly letters) used in a particu-
DISK that has become unusable for holding lar NUMBER SYSTEM. DECIMAL NOTATION has
data. This can be for a variety of reasons, base 10, BINARY NOTATION has base 2,
usually some form of hardware fault. The OCTAL NOTATION has base 8; the digits used
management of such sectors is one of the in these three systems are 0–9, 0 and 1, and
functions of a computer’s operating sys- 0–7 respectively. HEXADECIMAL NOTATION
tem, which can find and mark bad sectors has base 16 and uses the digits 0–9 and the
on a disk so that they can be ignored. six letters A–F. The base of a number can
be indicated by means of a subscript as in
bak A file extension commonly used to 10110, i.e. 101 in decimal notation
indicate that the file is a backup. In many 1012, i.e. 101 in binary notation, i.e. 510
applications a backup file is often pro-
duced automatically when files are saved; base address, base address register
it is usually the previous version of the file. See relative address.
band printer See line printer. Basic /bay-sik/ A group of similar high-
level PROGRAMMING LANGUAGES whose
bandwidth /band-witth/ The amount of name is derived from the words beginners
data that can be transmitted in a fixed time
all-purpose symbolic instruction code. The
period. In digital transmissions it is meas-
original Basic was developed in the mid-
ured in bits (or bytes) per second and in
1960s for INTERACTIVE computing, which
analog transmissions as cycles per second,
was a new concept at the time. Many com-
or hertz.
puter manufacturers and software houses
subsequently developed their own dialects,
banner A small advertising region on a
and the emergence of the microcomputer
Web page that contains a link to the adver-
tiser’s own Web site. It generally spans the increased the diversification to the point
width of the page but is not usually more where Basic was often extremely difficult
than an inch tall. to move from one computer to another.
Both ANSI and ISO standards have been
bar code A pattern of parallel lines of defined for Basic.
variable width and spacing that provides Basic is a simple language, is easy to
coded information about the item on learn, and is of particular importance to
which it appears. The codes are read by rel- beginners. It allows easy modification of
atively simple equipment, known as bar- programs since the program text is kept in
code readers, using optical or magnetic main store during execution. Basic is nor-
sensing techniques. Bar-code readers com- mally interpreted although it may also be
monly use a laser beam that scans back- compiled (see interpreter; compiler).
ward and forward, with a photoelectric All dialects of Basic should include the
cell to detect the signal reflected from the following components:
code. Bar codes are found, for example, on REM introduces a COMMENT line;
goods sold in supermarkets, where they are INPUT performs input from the keyboard
used to identify the product and its cost at or a file;
the checkout point and to update stock and READ performs input from a DATA state-
sales records. The first digit indicates the ment elsewhere in the program;
country of origin (0 = USA, 5 = UK), and DATA introduces a line of data values;
16
bcc
PRINT performs output to the screen or a count the estimated time to run the job and
file; its demands on resources such as main
IF introduces an IF THEN ELSE STATEMENT; store or tape drives. If a job is particularly
FOR introduces a LOOP; large it may not be run until the middle of
NEXT terminates a LOOP; the night. Batch processing was formerly
LET introduces an ASSIGNMENT STATEMENT the usual method of working, when com-
– may usually be omitted; puter time was a very expensive commod-
DIM declares ARRAYS; ity and had to be used efficiently. Such
GOSUB calls a SUBROUTINE; considerations no longer apply and mod-
RETURN returns from a SUBROUTINE to the ern computer systems, in particular micro-
statement after the GOSUB; computers, do not rely on the use of batch
STOP interrupts execution of the program; processing. The term has come to be used
END terminates the program; more informally, to describe the BACK-
SQR square root function; GROUND PROCESSING of one or more jobs
INT function truncating real numbers to under the control of a job control language
integers – the fraction is discarded: or batch file, possibly at a set time rather
ABS function returning the absolute value than immediately the job is submitted.
of a number, i.e. without a sign;
SGN function returning –1 for negative baud /bawd/ A unit for measuring the
numbers and 1 for positive numbers. speed at which SIGNALS travel in a com-
Apart from these components there are puter or in a communication system such
many statements and commands that as a telephone link. One baud is normally
occur in some dialects and not in others, assumed to be equal to one BIT per second,
and some that mean entirely different the signal being a sequence of the binary
things in different dialects. See also Q digits 0 and 1. One baud may however be
Basic; Visual Basic. equal to one symbol per second or one digit
per second. In general the baud is equal to
bat The file extension of an MS-DOS the number of times per second that the
BATCH FILE. signaling system changes state. Signaling
speeds generally vary from a few tens of
batch file A text file used to contain a thousands of baud (for, say, data transmis-
list of MS-DOS commands, which may sion by MODEM) up to tens or hundreds of
also need parameters. If this file has the millions of baud (for, say, data transmis-
bat extension then, when the filename and sion by satellite). The unit is named for the
values for the parameters (if necessary) French engineer J. M. E. Baudot (1845–
are typed at the command prompt, the 1903), who first developed the teleprinter.
instructions are processed in order.
AUTOEXEC.BAT on a Windows system is bboard /bee-bord, -bohrd/ See bulletin
an example of a batch file. Windows NT, board.
2000, and XP also support an enhanced
type of batch file that uses the extension BBS Bulletin-board system. See bulletin
cmd. board.
batch processing A method of organiz- bcc (blind carbon copy; blind courtesy
ing work for a computer in which items of copy) A directive in an e-mail program
work are queued up, and the OPERATING specifying where to send additional copies
SYSTEM takes one job at a time from the of the e-mail without the knowledge of the
queue and processes it. Each job must be main recipient. Copies are usually sent to
entirely self-contained – it must not require bcc addressees for information purposes
any intervention from the person who sub- only and the recipients are not necessarily
mitted it. This is because the order in which expected to take any further action. The
the jobs are run is at the discretion of the mail header does not contain bcc ad-
operating system, which will take into ac- dressees (compare cc).
17
BCD
E E
C C
C C
E E
E E
C C
C
C
E E
BCD See binary coded decimal. Bezier curve /bez-ee-ay, bay-zee-/ A type
of curve used in computer graphics. A sim-
benchmark A way in which the perfor- ple Bezier curve is defined by four points.
mance of computer systems – hardware Two of these are the end points of the
and software – can be compared. It is nor- curve. The other two are control points
mally in the form of a specially designed and lie off the curve. A way of producing
test or problem. For example, a system the curve is to first join the four points by
may be subjected to a known workload three straight lines. If the midpoints of
and the time taken to complete it is meas- these three lines are taken, together with
ured; this time can then be compared with the original two end points, a better ap-
the times achieved by other systems. The proximation is to draw four straight lines
systems are said to have been bench- joining these points. The midpoints of
marked. A multi-user system may be
these lines can then be used to get an even
benchmarked by another computer simu-
better approximation, and so on. The
lating the action of tens or hundreds of
Bezier curve is the limit of this recursive
individual users all going through a pre-
process. Bezier curves can be defined math-
determined sequence of actions.
When using a benchmark to evaluate ematically by cubic polynomials. A similar
two or more computer systems, it is impor- type of curve obtained by using control
tant that a prospective purchaser uses a points is known as a B-spline.
benchmark that reflects the sort of use to
which the system will be put. bidirectional printer /bÿ-dă-rek-shŏ-
năl/ See character printer.
beta /bay-tă, bee-/ Describing new soft-
ware or hardware that is in its final stages binary /bÿ-nă-ree, -nair-ee/ Composed
of development, containing most if not all of or relating to two components. In com-
of its intended functionality. It may not be puting the term refers to the base 2 num-
totally reliable but is ready for testing in bering system, in which values are
controlled situations by a few customers or expressed as combinations of the digits 0
individuals to iron out any unidentified and 1. Thus, 2 in decimal is written as 10
problems before general release. in binary. See binary code.
18
binary fractions
decimal digit 0 1 2 3 4 5 6 7 8 9
BCD code 0000 0001 0010 0011 0100 0101 0110 0111 1000 1001
binary code A rule for transforming binary data Any data that uses BINARY
data, program instructions, or other infor- NOTATION. All computer data is stored in
mation into a symbolic form in which only this way. The term is also commonly used
the two binary digits 0 and 1 are used. to denote data that is stored using all the
Once encoded there is no way of distin- bits available: for example, a series of the
guishing an instruction from a piece of numbers 0 to 255 decimal in a system with
data. The process of transforming letters, 8-bit words. It is distinguished from data
digits, and other characters into sequences that uses only the numbers 32 to 126 deci-
of binary digits is called binary encoding, mal, which is generally represented by ap-
and the representation used or produced is propriate ASCII characters.
called a binary encoding. The BINARY NO-
TATION used to represent numbers in a binary digit Either of the two digits 0
computer is a binary encoding. The repre- and 1. See bit.
sentation of characters using, for example,
the ASCII scheme is another example. See binary encoding See binary code.
also code.
binary file A file made up from a se-
binary coded decimal (BCD) A CODE quence of 8-bit data or executable code.
by which each decimal digit, 0, 1, 2, …, 9 Such files differ from text files, which can
is transformed into a particular group of 4 be read by eye, in that they are usually in a
binary digits (or bits). In the standard form form readable only by a program.
of BCD, each decimal digit is represented
by the group of 4 bits whose value is equiv- binary fractions Fractions expressed in
alent to the decimal digit, as shown in the binary notation. Binary fractions are usu-
table. The decimal number 58.271 would ally handled in FLOATING-POINT NOTATION.
thus be coded They can however be representated in
0101 1000.0010 0111 0001 FIXED-POINT NOTATION, e.g. .0101 or
The form of a decimal number in BCD is 11.101, where the positional value de-
usually different from its representation in creases from left to right by powers of 2:
BINARY NOTATION. the leftmost bit after the point has the value
19
binary logic
2–1 (i.e. ½ or 0.5), the next bit to the right Conversion from binary to decimal can
having the value 2–2 (i.e. ¼ or 0.25), and so be achieved by multiplying each bit in the
on. The fraction .011 therefore has the number by the relevant power of 2 (the
value least significant (rightmost) bit having a
(0 × 2–1) + (1 × 2–2) + (1 × 2–3) positional value of 20, i.e. 1), and adding
The decimal conversion is thus them together. The above binary number,
.25 + .125 = .375 10100, is thus equivalent to the decimal
Conversion of a decimal fraction to bi- number 20.
nary is shown in the diagram. It is done by Conversion from decimal to binary is
repeated multiplication by 2, the bit result- achieved by repeated division of the deci-
ing before the point being removed to form mal number by 2. The sequence of remain-
the bits of the binary fraction. The process ders forms the bits of the binary number,
is continued until a zero fraction is ob- the first remainder being the least signifi-
tained or the required accuracy is obtained. cant bit, and so on. Division stops when
See also binary notation. the quotient becomes 0. The steps in such a
conversion are shown in the diagram. See
also binary fractions; floating-point nota-
23 ÷ 2 =11 remainder 1 (rightmost bit) tion.
11 ÷ 2 = 5 remainder 1
5 ÷2 = 2 remainder 1 binary number See binary notation.
2 ÷2 = 1 remainder 0
1 ÷2 = 0 remainder 1 (leftmost bit) binary operator (dyadic operator) See
operator.
Conversion of decimal 23 to binary equivalent,
10111 binary representation (bit representa-
tion) The representation used within a
computer for numbers, CHARACTERS, and
binary logic 1. The electronic compo- MACHINE INSTRUCTIONS, and having the
nents used in a computer system to carry form of distinctive sequences of the two
out LOGIC OPERATIONS on binary variables BITS 0 and 1. See also binary code; binary
(i.e. quantities that can take either of two notation; complement; floating-point nota-
values). The components consist of LOGIC tion; character set.
GATES and LOGIC CIRCUITS.
2. The methods and principles under- binary search See searching.
lying this implementation of logic opera-
tions. binary signal See digital signal.
binary notation (binary system) The binary system Any system involving
number system employed in computing to just two possible values, two alternatives,
represent numbers internally. It uses the or two items. One example is the binary
two digits 0 and 1 and thus has a BASE 2. number system, i.e. BINARY NOTATION, in
These digits are known as binary digits (or which numbers are representated by
BITS), and a binary number is composed of means of the two digits 0 and 1. These two
a sequence of bits, e.g. 1001. Like decimal digits are used in computing not only in bi-
notation the binary system is a positional nary notation but to represent the two pos-
notation: the value of a binary number de- sible values, alternatives, etc., of any
pends not only on the bits it contains but binary system or situation, including the
also on their position in the number. The two possible directions of magnetization of
positional value increases from right to left a spot on a magnetic disk or tape or a low
by powers of 2. For example, the binary or high signal fed to a logic gate. See also
number 10100 has the value bit.
1 × 24 + 0 × 23 + 1 × 22 + 0 × 21 +
0 × 20 binary tree See tree.
20
black box
bistable /bÿ-stay-băl/ See flip-flop. black box A self-contained unit in, for
example, a computer system or a commu-
bit 1. Either of the two digits 0 and 1 nications system, whose function can be
used in computing for the internal repre- understood without any knowledge or ref-
21
blind carbon copy
erence to its electronic components or cir- whole diagram can represent any level of
cuitry. The notion of a black box is useful description, from an electric circuit to an
to someone with little or no electronics over-all computer system.
training who is trying to find out, say, what
a particular peripheral device does. The blocking factor See block.
term is also used of a software module that
can be used by other programs without blog (weblog) An online journal main-
knowledge of its internal workings. tained on and publicly accessible from the
WORLD WIDE WEB. Blogs began to appear in
blind carbon copy See bcc. the mid-1990s and by the early 2000s had
become so numerous as to form a distinct
blind courtesy copy See bcc. community on the web. A blog primarily
reflects the interests and knowledge of its
BlackBerry Trademark A PDA manufac- author; however, some have become re-
tured by Research In Motion Ltd. First re- spected sources of information or influen-
leased in 1999, it became very popular tial sources of opinion. Special software
because of its wireless E-MAIL facilities. helps to create and maintain blogs, for ex-
ample allowing readers to comment on
blink To flash on and off. Various ob- their content, and blogs often use web
jects, such as the cursor and warning mes- feeds to notify users of new content. The
sages, are made to blink on the computer activity of maintaining a blog is called
screen to catch the eye. blogging, and the totality of all blogs on
the Web is called the blogosphere.
block A group of RECORDS that a com-
puter can treat as a single unit during trans-
Bluetooth Trademark A method for
fers of data to or from BACKING STORE. This
electronic devices, including computers
unit of transfer is sometimes called a phys-
and associated hardware, to exchange data
ical record. The data records in the block
over short distances by radio waves. When
are then called logical records. The number
two Bluetooth-enabled devices come
of data records, i.e. logical records, in a
within range (typically 10 meters) they ex-
block is called the blocking factor. Blocks
have either a fixed length or a variable change information to determine whether
length. they are compatible and can exchange
A stream of data recorded on magnetic data; if so they automatically form an ap-
tape is divided into blocks, and is subse- propriate link, known as a piconet. A lap-
quently read block by block (see tape unit). top and a desktop computer, for example,
Successive blocks are separated by in- will form a small network, and a cordless-
terblock gaps (IBGs), in which nothing is phone handset will begin using a base;
recorded. Recording in blocks is used for however, the laptop will not link to the
convenience in handling the tape and for handset because the two devices will deter-
error management: any errors occurring in mine that they are incompatible. All this
the recording and reading of data are dealt happens automatically, with no user action
with block by block. The equivalent required. The Bluetooth specification is
recording subdivision on a magnetic disk is controlled by the Bluetooth Special Interest
a SECTOR. Group, an association of IT, cellular tele-
phone, and other electronics companies es-
block diagram A diagram that repre- tablished in 1998.
sents graphically the interconnection be-
tween elements of a computer system. The bmp The file extension of a type of
elements may range from electric circuits to RASTER GRAPHICSdata file widely used on
major units of hardware. They are shown Windows systems.
as labeled rectangles or other geometric
figures, and are connected by lines. The BNF (Backus-Naur form) A symbolic
22
bootstrap
notation in which the SYNTAX of a pro- It is also used in the design of the electronic
gramming language can be expressed. circuits – LOGIC GATES and logical circuits –
that control the flow of signals. See also
board See circuit board; printed circuit truth table; Boolean expression.
board.
Boolean expression (logical expression)
boldface A typeface used for emphasis, An expression that is formed according to
in which the characters print darker and the laws of BOOLEAN ALGEBRA. It contains
heavier. In this dictionary the headwords variables that take the values true or false
are printed in boldface type. and that are linked by LOGICAL OPERATORS.
An example is
bomb 1. A program residing on a com- a AND (b OR NOT c)
puter unknown to the user; typically, the a, b, and c are variables or expressions
program will have been planted in order to (such as x > 1) that return the values true or
damage or destroy the system. See virus; false; AND, OR, and NOT are operators.
worm. A Boolean expression can be used to form
2. (bomb out) Of a computer or program, a FUNCTION whose value is either true or
to fail abruptly without warning to the user false depending on the combination of val-
and without giving the user a chance to re- ues assigned to the variables. This function
cover or partially recover such items as is known as a Boolean function or logical
data or files. function. A Boolean function can be repre-
sented in a TRUTH TABLE. It can also be
bookmark 1. A marker placed at a spe- transformed into a LOGIC DIAGRAM of logic
cific place in a document that enables the gates.
user to return quickly to that place at a
later date. Boolean function See Boolean expres-
2. A link set up by the user to a Web page sion.
on the Internet. The URL is stored in a
local file and the bookmark can be clicked Boolean search A search that uses
on to retrieve the Web page automatically. Boolean operators. Boolean searches are
commonly used in database searching and
Boolean algebra /boo-lee-ăn/ An exten- in Internet SEARCH ENGINES.
sion of the principles of algebra into the
field of logic. It was first put forward in boot (boot up) See bootstrap.
1847 by the British mathematician George
Boole and is now of particular importance boot disk A floppy disk or CD-ROM
in computing. Boole applied the methods containing all the necessary files for an op-
of algebra to problems in logic involving erating system to start on a computer. The
statements and conclusions that can be boot disk is inserted into the appropriate
given a truth value – either true or false – drive and the computer switched on. The
and thus have a binary nature. These true files are copied into memory and the boot
or false expressions can be combined by up sequence started. A boot disk can be
the logical operations and and or and used to start up the operating system in the
negated by the operation not to give a con- event of failure of the hard drive.
clusion whose truth or falsity can be deter-
mined from the rules of Boolean algebra. bootstrap In general, a means or tech-
Boolean algebra is basic to many as- nique enabling a system to bring itself into
pects of computing. For example it is used some desired state. The word is used in sev-
in high-level languages in conditional state- eral ways in computing. Most commonly,
ments: a bootstrap is a short program whose func-
IF NOT(a<b) AND (x=0) THEN … tion is to load another longer program into
or logical assignments and expressions: a computer. For example, when a com-
answer := (char = ‘Y’) OR (char = ‘y’) puter is first switched on its MAIN STORE
23
bot
will be empty except for those parts fabri- count involved so that the breakpoint only
cated in ROM (read-only memory). A boot- operates after it has been passed a certain
strap stored in ROM is capable of reading number of times. Breakpoints are normally
from backing store (which is NONVOLATILE) used only in DEBUGGING.
the OPERATING SYSTEM, or some part of it,
without which the computer cannot oper- bridge A device that connects two net-
ate. The computer is then said to be booted works into one large logical network so
up or booted, and a program can then be that information can flow between them.
loaded into the machine. The networks may be using different pro-
tocols, and they are joined together at the
bot A program that performs a repeti- data link layer (OSI layer 2).
tive, monitoring, or time-consuming ac-
tion. It is an abbreviation of ‘robot’. broadband /brawd-band/ 1. (wideband)
Describing communications systems sup-
bounce The return of a piece of e-mail porting a wide range of frequencies, on
because of an error in its delivery. which multiple messages can be carried by
the transmitting medium at the same time.
bounds See array. Such systems are ideal for digital transmis-
sions.
bpi Bits per inch. See density. 2. Denoting a method of INTERNET ACCESS
by telephone line that uses these techniques
bps See bits per second; bit rate. to achieve much greater data transfer rates
than are possible with traditional MODEMS.
branch 1. See jump.
The most popular form of broadband is
2. A set of instructions that are executed
ADSL.
between two BRANCH INSTRUCTIONS.
broadcasting See multicasting.
branch instruction A MACHINE IN-
STRUCTION that controls the selection of
one set of instructions from a number of al-
browse To scan files, documents, or
the Internet for items of particular or gen-
ternative sets during the execution of a pro-
gram. A branch instruction is usually eral interest. In browsing, information
regarded as being the same thing as a jump tends to be checked and gathered but not
instruction. See jump. altered.
24
bus network
jet is a trademark of Canon Inc. See also amounts of data can be stored, usually
laser printer; ink-jet printer. with relatively slow access times. Examples
are hard disk, CD, DVD, and magnetic
bubble memory An obsolete type of tape.
memory in which data in binary form is
represented by the presence of absence of bullet A character, such as a large dot or
minute magnetized regions (bubbles) small square, used to introduce the start of
within an oppositely magnetized magnetic a paragraph to add emphasis to the text.
material. Magnetic fields are used to form
the bubbles and to move them along a path bulletin board (bboard) A general facil-
through the surface of the stationary mag- ity on a computer NETWORK allowing any
netic medium. Bubble memory is a form of user of the network to leave messages that
SERIAL-ACCESS memory but the data path is can be read by all the other users. This can
constructed so that the ACCESS TIME to a be contrasted with E-MAIL messages, which
particular data item is very short. It is usu- are read only by the addressees.
ally NONVOLATILE and has high storage ca-
pacity. Bubble memory has had only bundled See unbundling.
limited application and has been super-
seded by other faster devices, such as FLASH bundling See unbundling.
MEMORY.
25
button
button 1. An area (e.g. a small box) in a 2. A movable piece on a mouse that, when
GUI FORM that, when clicked on with the pressed, will activate a particular function.
mouse or entered (if it has the focus) by
pressing the enter key, will implement a byte A unit of data or memory consist-
command. ing of 8 bits.
26
C
C A PROGRAMMING LANGUAGE developed main store into the cache; the next instruc-
in the early 1970s by Dennis Ritchie at Bell tions for execution are then likely to be in
Laboratories in the USA for systems devel- the cache as well as the one currently re-
opment, and in particular for writing the quired. This is a very successful strategy
operating system UNIX in order to make it and significantly increases execution
PORTABLE. C has the CONTROL STRUCTURES speeds. The contents of a cache may be
usually found in HIGH-LEVEL LANGUAGES modified by the program, so it will be nec-
but has features that make it suitable for essary to copy them back into main store
writing SYSTEMS SOFTWARE. Its popularity before refilling the cache with another
increased with the spread of the UNIX op- block. See also L1 cache; L2 cache.
erating system, but it is by no means con- The same principle is used to speed
fined to UNIX machines. See also C++. other data transfers. For example, a disk
cache holds copies of recently accessed disk
C++ /see-dub-’l-pluss/ A PROGRAMMING
sectors in RAM and can save many time-
LANGUAGE developed from C in the 1980s. consuming disk reads.
It retains the syntax of C while adding
comprehensive OBJECT-ORIENTED PRO- CAD /kad/ (computer-aided design) The
GRAMMING features as well as other en-
application of computer technology to the
hancements. It has become an important design of a product. CAD is used especially
language for software development.
in architecture and electronic, electrical,
mechanical, and aeronautical engineering.
C# /see-hash/ A programming language
Designs can be created by computer using
based on C++ and JAVA. It was introduced
information fed in by experts and also ac-
by Microsoft as part of the .NET software-
quired from other sources, for example
development platform and became an ISO
specifications of component sizes or build-
standard in 2003.
ing regulations. During the design process
cable television See CATV. the design itself is displayed on a screen
and can be tested and modified by the tech-
cache memory /cash/ (cache) Extremely nical designer. The computer can analyze
fast SEMICONDUCTOR MEMORY that is used various characteristics of the design, the re-
in computer systems in association with sults being fed back to the designer. The
MAIN STORE. Its ACCESS TIME is much shorter final design is normally drawn by a flatbed
than that of main store and it is therefore PLOTTER, with design specifications, etc.,
used to increase the accessibility to data re- listed on the printout.
quired by the processor. It is a temporary The CAD output, the design of, say, a
store, i.e. a BUFFER, that is continually up- PRINTED CIRCUIT BOARD, may then be
dated so that it contains the most recently passed to other systems for computer-
accessed contents of main store. aided manufacture (CAM) and computer-
When the processor requires an instruc- aided testing (CAT). The combined process
tion not currently in the cache, then a of computer-aided design and manufacture
whole block of memory (the size depends is known as CADCAM. The whole pro-
on the particular system) is copied from cedure – computer-aided design, manufac-
27
CADCAM
ture, and testing – is often referred to as calling sequence includes some means of
CADMAT. passing the necessary data to the program
or program section (see parameter), and of
CADCAM /kad-kam/ See CAD. returning control to the calling program
following execution of the program or pro-
CADMAT /kad-mat/ See CAD. gram section. When a whole program is to
be called this is done by the OPERATING SYS-
CAE (computer-aided engineering) The TEM.
use of computers in engineering design,
testing, and research. callback A network security feature
used to verify users. The user’s name and
CAI (computer-aided (or assisted) instruc- password are checked and then the net-
tion) See CAL. work disconnects and returns the call to a
predefined authorized number.
CAL (computer-assisted learning) Any
use of computers to aid or support educa- calling program See call.
tion and training. CAL can test attainment
at any point, provide faster or slower calling sequence See call.
routes through the material for people of
different aptitudes, and can maintain a capacity (storage capacity) The amount
progress record for the instructor. of information that can be held in a storage
This application is also known as CBL device. It is usually measured in BYTES or in
(computer-based learning), CAI (com- BITS. For example, the capacity of a CD is
puter-aided (or assisted) instruction), and 700 megabytes, while the capacity of a
CMI (computer-managed instruction). hard drive is several tens of gigabytes.
calculator An electronic device that can caps lock A keyboard toggle key that
perform simple arithmetic, and often other when on makes all alphabetic characters
operations, on numbers entered from a typed appear as capitals (upper case).
keyboard. Final solutions and intermediate When off all alphabetic characters are
numbers are generally presented on LCD or lower case. Other characters are unaffected
LED DISPLAYS. Present-day calculators range by the caps-lock key.
from very cheap devices that can add, sub-
tract, multiply, and divide numbers to capture To transfer received data into a
those that can perform complex mathe- file.
matical and statistical operations and may
be programmed. Add-on memory modules carbon copy See cc.
containing specialist programs – for navi-
gation, say – can be used with the more ex- card A PRINTED CIRCUIT BOARD, often of
pensive calculators, as can small printers. a fairly small size.
call An action whereby a program or a card reader A device that reads data
section of a program – a PROCEDURE, SUB- encoded on cards and converts it into bi-
ROUTINE, or FUNCTION – is brought into ef- nary code for processing by a computer.
fect. Control is transferred to the ENTRY The data is represented by magnetic pat-
POINT of this program or program section, terns in, say, a MAGNETIC STRIPE on a plas-
causing the program or program section to tic card.
be immediately executed by the computer.
The program or program section is said to caret /ka-rĕt/ The symbol ˆ on the key-
have been called. The program that issued board obtained by pressing the shift+6
the call is the calling program. keys. This symbol is used to represent ex-
A piece of code, known as the calling se- ponentiation in some programming lan-
quence, is required to perform a call. The guages, as in 3ˆ4 meaning 34, or to
28
CBL
29
cc
cc (carbon copy; courtesy copy) The di- power laser. The depressions (known as
rective in an e-mail program specifying ‘pits’) have a lower reflectivity than the flat
where to send additional copies of the text. areas (known as ‘lands’). The pattern of
Copies are usually sent to cc addressees for pits and lands correspond to 0 and 1 values
information purposes only and the recipi- of bits. CDs of this type are produced on a
ents are not expected to take any further blank disk using a high-powered laser op-
action. The mail header contains a list of erating in short bursts to ‘burn’ the pits.
all cc addressees. Copies can be produced by taking impres-
sions from a master. Such CDs are also
CCD See charge-coupled device. used as a method of distributing data for
use on computers (e.g. software). In addi-
CD (compact disk) A metal (aluminum) tion the disks can be used as extra storage
disk 120 mm in diameter on which digital available to the processor. This is known as
data is recorded by mechanical means (as CD-ROM (compact disk read-only mem-
opposed to magnetic recording). The stan- ory). CD readers on personal computers
dard type of CD is the well-known audio are generally designated as 36-speed, 52-
disk used for sound recording. The disk has speed, etc., referring to the access time in
a spiral track containing a pattern of comparison to that of a standard audio
minute depressions. It is read by a low- compact disk. CD-ROM is widely used for
1.2 mm
polycarbonate plastic
stamped CD
additional
scratch resistant protective silver, silver
polymer with lacquer alloy, or gold
label on top coating dye (recording layer)
1.2 mm
polycarbonate plastic
CD-R
1.2 mm
polycarbonate plastic
polycarbonate plastic
CD-RW
CD types
30
central processor
computer games and multimedia publica- CD-I (compact disk interactive) A stan-
tions. It is also one of the safest methods of dard for combining audio, video, and text
storing electronic information. Because the on a compact disk using optical-disk tech-
data is encoded as depressions in a metal nology. A CD-I player needs no external
surface it is less susceptible to external in- computer. It plugs directly into a TV and
fluences than data stored magnetically and stereo system and comes with a remote
is less likely to change with time. The data control to allow the user to interact with
on a standard CD is regarded as permanent software programs sold on disk, thus pro-
and cannot be changed or overwritten. A viding a games-playing capability. Many
CD can store about 700 megabytes of data. features are included such as music, special
Producing an original standard CD re- audio effects, graphics, text, animation,
quires specialized equipment using a high- and video.
power laser and is usually done by
commercial companies. Different forms of CD-R See CD.
compact disk have been introduced that
can be produced using low-power lasers CD-ROM See CD.
and relatively cheap equipment. CD-R
(compact disk recordable) is a form of CD CD-RW See CD.
in which the metal disk has a thin gold
layer covered with a photosensitive dye Celeron /sel-ĕ-ron/ See Pentium.
coating and a protective lacquer layer. The
CD writer (which usually doubles as a CD cell 1. A LOCATION in memory or a REG-
reader) contains a laser that acts in short ISTER. It is usually capable of holding a sin-
bursts to change the dye layer, reducing the gle item of information in binary form such
reflectivity of the underlying gold. These
as an integer or instruction, but may hold
regions correspond to the pits in a standard
only a single bit.
CD; the unchanged regions correspond to
2. A box on the screen in a spreadsheet
the lands. Once the data has been recorded
program or in a spreadsheet-like grid when
it cannot be changed; i.e. CD-R is an ex-
used in any other type of program, which
ample of a WORM device (write once,
contains a number, a formula, or a string
read many times). It is, however, possible
of text.
to add extra data to a CD-R later. This is
known as multisession recording. CD-R is
a cheap and reliable method of archiving cell address 1. The reference number of
data. a particular memory location that stores
CD-RW (compact disk rewritable) is a one byte of information.
further development in which the data on a 2. (cell reference) The row and column
CD can be erased and overwritten (i.e. the identifiers used to specify a cell in a spread-
CD is reusable in a similar way to a mag- sheet or in a spreadsheet-like grid.
netic disk). This type of CD has a reflective
layer consisting of a silver-based alloy. The center To align text around a horizontal
laser writing the data causes localized heat- midpoint on a line or vertical midpoint on
ing of small areas, which changes the struc- a page or other defined area in text for-
ture from a crystalline to an amorphous matting.
form. These amorphous regions have
lower reflectivity and correspond to the central processor (central processing
pits in a standard CD. The data can be unit; CPU) The principal operating part
erased by heating to a higher temperature of a computer. It consists of the ARITH-
to restore the crystalline form. This type of METIC AND LOGIC UNIT and the CONTROL
CD was formerly called CD-E (compact UNIT, i.e. the units in which program in-
disk erasable). structions are interpreted and executed.
Sometimes MAIN STORE is considered a
CD-E (compact disk erasable) See CD. component of the central processor.
31
Centronics interface
Over the years, the functions in the sequence. The entries can then be dispersed
larger or more complex computers have randomly within a storage device. One
become distributed among various units, particular entry may belong to more than
each able to handle one or more tasks quite one sequence.
independently. In these systems the term 2. A technique whereby the final action of
central processor (or central processing a program is to load its successor into main
unit) is inappropriate, and the word store. For example, a suite of demonstra-
PROCESSOR (or processing unit) is used. tion programs could be chained together
with the last one chaining to the first, thus
Centronics interface /sen-tron-iks/ An providing a continuous demonstration.
old but still widely used interface standard
designed for parallel data exchange be- chain printer See line printer.
tween computers and peripherals. Origi-
nated by Centronics Corporation for dot channel 1. A route along which infor-
matrix printers, it provides for the ability mation can be sent. It may for instance be
to send 8 bits of data (usually one charac- a telephone link between two computers or
ter) simultaneously together with control a cable connecting a terminal to a com-
and status information. puter. It may also be the route followed by
data between a user’s program and a file on
CGA (color graphics adapter) One of backing store. 2. See web feed.
the early graphics adapters introduced by
IBM in 1981 and capable of several char- character A symbol used in representing
acter and graphics modes. The quality was data and in organizing and possibly con-
quite poor with a resolution of, for exam- trolling data. A character may be one of the
ple, 320 × 200 pixels with four colors or 26 letters of the Roman alphabet, a letter
640 × 200 pixels with two colors. See also combined with a diacritical mark, one of
EGA; VGA. the digits 0–9, a punctuation mark,
bracket, plus sign, etc. Note that A and a
CGI 1. (common gateway interface) A are different characters, and that a com-
set of rules defining how a Web server puter distinguishes between the capital let-
communicates with a Web client. CGI pro- ter O and zero, 0, and between the lower
grams are usually small and perform tasks case letter l and one, 1. Modern computers
such as counting the number of hits for a are increasingly able to handle the large
Web page, accepting and filtering data number of symbols needed to represent all
from a form, or reading an image map. All the world’s writing systems. See also char-
the processing takes place on the Web acter set; byte.
server, not on the computer doing the
browsing. CGI scripts are often written in character encoding 1. The process of
Perl. transforming letters, digits, and other char-
2. See computer graphics interface. acters into an encoded form. See code.
2. The symbolic representation used for the
chad /chad/ The piece of paper, plastic, characters in a given CHARACTER SET when
etc., that is removed when a hole is encoding them. The symbols are usually
punched in a data medium. For example, the two binary digits 0 and 1. ASCII and UNI-
the punching of sprocket holes along the CODE are two widely used character encod-
edges of continuous stationery produces ing schemes.
chads.
character generator A ROM-based cir-
chaining 1. An arrangement whereby cuit or a program that converts codes for
one item in a sequence contains the means display on a screen. A valid ASCII code is
for locating the next item. For example, a converted into a set of numbers that will
FILE may be organized such that each entry cause a VDU to illuminate the correct pat-
contains the ADDRESS of the next entry in a tern of dots to display the character. This is
32
check
, ; : . ? !
much faster than the process of drawing
the character by a series of graphics com- ( ) [ ] { }
mands, and formerly made text-based ⬙ ⬘ # % & | \
VDUs the preferred systems for most tasks. + – * / > < =
However, the number of characters that
could be generated was very limited. Such
Typical special characters used in a character
speed considerations are not relevant to to- set
day’s faster computers, whose GUI operat-
ing systems use graphics commands to
generate a wide range of characters in a va- special characters are known as graphic or
riety of typefaces, styles, sizes, etc. Never- printable characters, i.e. they are symbols
theless, all computers still have a character that can be produced by printing, writing,
generator so they can communicate with a etc. In contrast the control characters each
user when the GUI is not loaded or is not produce a particular effect, such as a new
working properly. line or backspace.
Within a computer all these characters
character printer A PRINTER that pro- must be represented in binary form. There
duces a single character at a time, printing are two widely used encoding schemes for
them in the order in which they appear in a characters; these are ASCII and UNICODE.
line. The printing mechanism – known as The 128 different bit patterns used in
the print head – prints as it moves from left ASCII and the many thousand used in Uni-
to right or right to left along each line. Such code are themselves character sets. See also
printers have now largely been superseded EBCDIC.
by LASER PRINTERS and INK-JET PRINTERS. See
also daisywheel printer; thermal printer. character string See string.
33
check box
across a computer network, or following particular CPU and its supporting cir-
its computation. cuitry.
A single check digit may be used in per-
forming the check. This digit is added to chip socket A device to allow easy re-
one end of the item of information when it placement of chips (INTEGRATED CIRCUITS)
is encoded, and is derived arithmetically on a PRINTED CIRCUIT BOARD. The chip
from the digits making up the item. It can socket is soldered to the circuit board and
be used subsequently to determine whether the chip is pushed into the socket, which
the item contains an error (or errors). In has a small hole for each of the chip’s legs.
some cases a group of characters may be With larger chips care is needed to avoid
used to perform a check rather than a sin- bending the legs of the chip on insertion.
gle digit. See also parity check; modulo-n
check. CIM 1. (computer-integrated manufac-
turing) The use of computers in manufac-
check box A small, usually square, se- turing. Such topics as advanced control
lection box used for interactive control in techniques including robotics, manufactur-
graphical user interfaces. The check box ing systems design and control, systems
will contain a check or cross if it has been modeling, and analysis and management
selected and clicking on the box will re- accounting are all included.
verse the selection. There may be several 2. (common information model) A frame-
check boxes in a form for the selection of work for describing and sharing manage-
several different features at the same time. ment information that is both system and
application independent.
check digit See check; modulo-n check.
cipher, ciphertext /sÿ-fer/ See cryptog-
checkpoint A point in some processing raphy.
activity, or a place in a program, at which
a DUMP – a copy – is taken of data associ- CIR (current instruction register) See in-
ated with the active program. It is hence struction register.
the point or place from which the program
can subsequently be restarted. The use of circuit board A rigid board of insulat-
checkpoints guards against system failure ing material on which an electric circuit has
during very long program executions: if been or can be built. The term is commonly
anything goes wrong, the execution can be used to refer to a PRINTED CIRCUIT BOARD.
continued from the last checkpoint once Circuit boards come in a variety of sizes,
repairs have been made. The frequency of some of which are standardized.
checkpoints depends on the reliability of
the system. circular shift See shift.
checksum /chek-sum/ See modulo-n CISC /sisk/ (complex instruction set com-
check. puter) A CPU designed with a full set of
assembly calls. Systems based on this phi-
chip 1. A small section of a single crystal losophy can make software simpler and
of semiconductor, usually silicon, that thus smaller, but are generally slower in the
forms the substrate (i.e. base) on which an execution of each instruction.
INTEGRATED CIRCUIT is fabricated.
2. Informal name for integrated circuit. class A fundamental structure in OBJECT-
ORIENTED PROGRAMMING. A class contains
chipset /chip-set/ A group of MICROCHIPS all the methods (i.e. FUNCTIONS and PROCE-
performing one or more related functions DURES) and data necessary to represent
and designed to work as a single unit. some object. A games program, for exam-
Often a chipset will fit on a single chip. The ple, might contain a class called ‘Auto’ that
term commonly refers to a set containing a models the characteristics of an automo-
34
client-server
bile, with such methods as ‘accelerate’, click trail The record of a Web surfer’s
‘brake’, ‘turn’, etc. The accelerate method route through the Web sites visited. Web
might use data items in the class, called browsers generally maintain such a record
PROPERTIES, that define such characteristics as a convenience to the user. However, the
as maximum speed and maximum rate of possibility of this data being analyzed by
acceleration to determine how long it takes third parties has raised privacy concerns,
the automobile to reach a given speed. Be- especially if it is obtained illicitly through
cause multiple instances of the Auto class MALWARE.
can be declared, each automobile can be
given its own characteristics by assigning client 1. A desktop computer or work-
different values to this data. The rest of the station attached to a network and relying
program interacts with a class entirely on one or more SERVERS for resources such
through its public methods and properties; as files, devices, and other services.
the internal workings of the class, which 2. A computer system or process that re-
might include private methods and proper- quests a service of another computer sys-
ties, are a BLACK BOX. The use of classes can tem or process. In client–server computing,
considerably simplify complex program- the ‘front-end’ client program that the user
ming tasks. runs contacts and obtains data from a
server software program on another com-
clear 1. To set the contents of a register, puter, often at long distances. Each client
counter, or storage location to zero. program is designed to work with one or
more specific kinds of server programs and
2. To erase all the characters from a VDU
each server requires a specific kind of
screen, leaving it blank.
client. For example, a workstation request-
ing the contents of a file from a file server
CLI See command line interface.
is a client of the file server.
click To press and release a mouse but-
client-server (client/server) Describing
ton with the CURSOR pointing at an ICON or
a program whose operation is divided be-
other object on the screen. In general click-
tween two computers on a network. These
ing selects the object, but in some cases,
are the CLIENT, which interacts with the
such as BUTTONS and hyperlinks on Web user by receiving commands and present-
pages, an action is initiated. Compare dou- ing results; and the SERVER, which manages
ble click. the application’s data and does most of the
computation or data processing in re-
clickstream In Web advertising, the se- sponse to requests from the client. Often
quence of ‘clicks’ or pages requested as a the server is a dedicated machine that
visitor explores a Web site. serves multiple clients or even multiple
client-server programs of a similar type. A
click through (click-through; click- single server running a DATABASE MANAGE-
through) The action of directing a Web MENT SYSTEM, for example, can handle
browser to a particular Web site by click- multiple clients running a range of pro-
ing the mouse on a link from another site grams that utilize its databases. The princi-
or in an e-mail. When the link is an adver- pal advantages of client-server applications
tisement, it is usually paid for at a fixed are: reduced cost, because the computa-
rate per click through: i.e. the payment de- tional workload is concentrated in the
pends directly on how successful the adver- server and so the clients need not be pow-
tisement is at attracting people to the site. erful and expensive machines; increased se-
In E-MAIL MARKETING the click-through curity and robustness, because the server
ratio measures the success of a marketing manages all access to shared data; and re-
e-mail by dividing the number of click duced network traffic, because the clients
throughs it produces by the total number send requests to the server rather than ma-
of copies sent out. nipulating the data directly.
35
clip art
clip art Simple artwork in digital form by a different company, usually with
designed for use in publications or Web cheaper components.
pages. The use of clip art can save time and
make art available for nonartists. Clip art closed source See open source.
includes both subject-related illustrations
and visual elements such as lines, bullets, cluster 1. A group of terminals, mag-
and text separators. It can be purchased netic disks, processors, whole computers,
and is also available free for personal use etc., that are configured to work together
from many Web sites. and which perform as a single unit in inter-
actions with the rest of the computer or
clipboard A temporary storage area in- network. Clustering improves both system
side a computer. It is used to copy or move speed and fault tolerance.
data from one area of a document to an- 2. The base unit of disk-space allocation in
other, or from one application to another. a FILE SYSTEM. A cluster consists of one or
See cut; paste. more SECTORS.
36
com
continuous path along which electrical sig- 5. To represent data or a program in a sym-
nals (usually of high frequency) can be con- bolic form.
veyed from one point to another in a
system. codec /koh-dek/ 1. (coder–decoder) An
integrated circuit or chip that performs
Cobol /koh-bol/ A high-level PROGRAM- analog-to-digital conversion and vice
MING LANGUAGE whose name is derived versa. The most common use for such a de-
from the words common business oriented vice is in a modem.
language. It was developed specifically for 2. (compression–decompression) An algo-
administrative and financial use in the late rithm or specialized computer program or
1950s and early 1960s. It was adopted by hardware that reduces the size of large
the US Department of Defense, and is still files.
the most widely used commercial data-pro- 3. A combination of these functions, par-
cessing language. There have been several ticularly used for creating and compressing
international standards as the language has digital data for video or audio signals from
developed. analog data being received in real time and
subsequently expanding it for playback.
code 1. A rule for transforming data or MPEG is a popular codec for computer
other information from one symbolic form video.
into another. In the codes employed within
a computer, the symbols used are generally cold start See restart.
restricted to the two BITS 0 and 1. The
process of transformation is called encod- collating sequence An ordering of
characters in a character set used within a
ing, and the symbolic representation used
computer. It is employed, for example, in
or achieved is known as an encoding. The
sorting into alphabetic or alphanumeric
process of reconverting the coded informa-
order.
tion into its original form is called decod-
ing. Like encoding the decoding process
collision In an Ethernet network, the re-
follows a strict rule. Both processes are
sult of two devices on the same network at-
thus algorithmic in nature. They are car-
tempting to transmit data at exactly the
ried out by an encoder or a decoder. These
same time. The network should detect the
may be in the form of a piece of hardware collision and discard both pieces of data.
or a piece of software.
Different kinds of codes are used in color graphics adapter See CGA.
computing for different purposes. For ex-
ample, when characters – letters, numbers, color monitor A monitor that displays
etc. – are fed into a computer they are colors. The highest resolution can be ob-
transformed into a CHARACTER ENCODING, tained using a monitor in which dots of
such as UNICODE, so that they can be ma- three different phosphors, one each for red,
nipulated by the computer equipment. blue, and green, coating the inside of the
Again, ERROR-DETECTING and ERROR-COR- screen are caused to glow by electronic sig-
RECTING CODES are used to improve the re- nals representing different intensities of the
liability of data when it is transferred to or three colors. By this method the three basic
from disk or when it is transmitted. colors can be blended into millions of dif-
2. Any piece of program text written in a ferent colors. A monitor using this tech-
programming language. nique is known as an RGB monitor.
3. The particular language or type of lan-
guage in which a piece of program text is com /kom/ 1. In MS-DOS, the file exten-
written, e.g. machine code, C++ code, sion of a command program file. This is a
source code, etc. type of executable binary file limited to one
4. The encrypted form of a message. See 64-kilobyte segment and is now seldom
cryptography. used except to provide backward compati-
37
COM
bility. COMMAND.COM is the main pro- truth table (as is done in the diagram at
gram of commands for the MS-DOS oper- logic circuit). The output of the circuit at a
ating system. particular time can thus be determined by
2. A top-level Internet domain name, gen- the combination of inputs.
erally describing addresses operating in a
commercial environment. combo box A type of CONTROL on a
GUI FORM. It combines the functions of a
COM /kom/ 1. (communication port) A LIST BOX and a TEXT BOX: the user can either
port on the central unit for a cable connec- select its content from a list of options or
tion to a serial device such as a modem. type it in if something not on the list is re-
2. (component object model) A Microsoft quired.
standard designed to allow implementa-
tion of shared software components across command See command language.
diverse programs running on Microsoft
Windows platforms by specifying the sys- command language A kind of pro-
tem their INTERFACES must use. ActiveX gramming language by means of which a
controls and DirectX are examples of user can communicate with the OPERATING
COM objects. SYSTEM of a computer. Statements in such a
3. (computer output on microfilm) Com- language are called commands. They are
puter output recorded in miniaturized requests from someone using a terminal or
form on microfilm, either on a reel of film a microcomputer for the performance of
or on card-sized sheets of film known as some operation or the execution of some
microfiche (or fiche). The term COM also program, examples being ‘list’, ‘sort’,
applies to the techniques used to produce ‘delete’, ‘load’, ‘run’. JOB CONTROL LAN-
this form of output. The information is GUAGES are a type of command language; in
considerably reduced in size before the turn, command languages are a type of
recording process, enabling a large amount SCRIPTING LANGUAGE.
of information to be stored. Special optical
viewers must be used to enlarge the infor- command line interface (CLI) A user
mation on the microfilm so that people can interface to an operating system or appli-
read it. New film or fiche must be pro- cation program in which the user responds
duced if additions or corrections are re- to a visual prompt by typing in a command
quired. on a line, receives a response back from the
COM has been extensively used since system, and then enters another command,
the 1960s, for example, in libraries to cat- and so on. The Command Prompt in a
alog books by author and subject. Micro- Windows operating system is an example
film and microfiche also have applications of a command line interface. Today most
in areas that do not require a computer, users operate with a GRAPHICAL USER INTER-
e.g. in the recording of books, newspapers, FACE (GUI) such as those offered by Win-
and other documents. dows, Apple Macintosh, and others.
38
compilation error
39
compilation time
SYNTAX ERROR, detected when a program is translated into MACHINE CODE, so that the
compiled. See error. program can subsequently be executed by
a computer. It is at compile time that in-
compilation time The time taken to formation about the program, such as a
perform the compilation of a program. See compiler listing or a SYMBOL TABLE, may be
compiler; compile time. produced to aid later DEBUGGING. See also
compiler; run time.
compiler A program that takes as input
a program written in a HIGH-LEVEL LAN- complement The outcome of a NOT OP-
GUAGE, such as C++ or Fortran, and trans- ERATION on a digit, integer, truth value,
lates it into MACHINE CODE, or occasionally proposition, or formula. The complement
another form of object code. Each STATE- of the binary digit 0 is 1 while the comple-
MENT in the high-level language is con- ment of the binary digit 1 is 0. The com-
verted into many MACHINE INSTRUCTIONS, plement of the truth value true is false and
sometimes hundreds. The input to the vice versa. The complement of the proposi-
compiler is called the source program and tion ‘this object is spherical’ is ‘this object
the output is called an OBJECT MODULE. The is not spherical’.
object module is usually stored on mag- In general, the complement of a digit, d,
netic disk separately from the source pro- can be found from the formula
gram. The translation process is known as (B – 1 – d)
compilation, and the program that has where B is the BASE of the number system
been translated is said to have been com- (equal to 10 in decimal notation, 2 in bi-
piled. The entire program must be com- nary notation). The complement of the
piled before it can be executed. Complex decimal digit 6 is thus 3.
programs are usually written in several The complement of a whole number,
modules, each of which is compiled sepa- i.e. an integer, is obtained by replacing
rately. The object modules are then com- each digit in the integer by its complement.
bined by a LINK EDITOR to produce the In binary notation this gives the one’s com-
OBJECT PROGRAM. plement. (In decimal notation it gives the
A compiler not only translates program nine’s complement.) The one’s comple-
statements but includes LINKS for PROCE- ment of 0110 is thus 1001. In binary nota-
DURES and FUNCTIONS from the system li- tion the two’s complement is formed by the
brary and allocates areas in main store. addition of 1 to the one’s complement. The
The object program can subsequently be two’s complement of 0110 is thus
loaded into and run by a computer. The 1001 + 1, i.e. 1010. To determine the orig-
compiler can also produce a LISTING of the inal integer from its one’s or two’s comple-
source program, if required, and reports ment, the same process is repeated: the
the errors (mainly SYNTAX ERRORS) found complement is taken (for the one’s comple-
in the source program during compilation; ment) or the complement is taken and 1 is
it normally indicates the position and na- added (for the two’s complement).
ture of these errors. One’s complement and two’s comple-
Each high-level language that can be ment are both used in computing. A one’s
used on a particular computer requires its complement is produced by a NOT opera-
own compiler or INTERPRETER. Program tion on the binary number concerned, con-
run times are much faster when a compiler verting 1s to 0s and 0s to 1s; a two’s
is used rather than an interpreter. Once complement is produced by a further addi-
compiled, the same program can be run tion of 1. Two’s complement arithmetic is
any number of times. In contrast, an inter- simpler than one’s complement arithmetic
preter must be used each time the program and is used both in the representation of
is run. See also assembler. negative numbers and in subtraction. Each
positive number is represented in its usual
compile time The period of time during form in binary notation except that it al-
which a program in a high-level language is ways has at least one leading 0. Each nega-
40
computer
41
computer-aided design
tion is to be read from the input devices or nipulate the information, for example to
written to the output devices. straighten lines, move or erase specified
Computers range widely in per- areas, expand or contract details. Images
formance, size, and cost. They are often may be two-dimensional, i.e. flat, or may
classified as MICROCOMPUTERS, MINICOM- appear three-dimensional. A series of im-
PUTERS, or MAINFRAMES. All three groups ages can be combined in ANIMATION.
may be used as general-purpose machines, There are many different formats for
capable of solving a wide variety of prob- storing images but they fall into two main
lems. Alternatively they may be designed classes. In raster graphics the picture is
for a special purpose or for a limited range stored as a series of dots (or pixels). The in-
of problems. formation in the computer file is a stream
of data indicating the color of a dot. Im-
computer-aided design See CAD. ages of this type are sometimes known as
bitmaps. This format is used for high-qual-
computer-aided engineering See CAE. ity artwork and for photographs. Dia-
grams are more conveniently stored using
computer-aided testing See CAT. vector graphics, in which the information
is stored as mathematical instructions. For
computer architecture See architec- example, it is possible to specify a circle by
ture. its center, its radius, and the thickness of
the line forming the circumference. More
computer-assisted learning (computer- complicated curves are usually drawn
assisted instruction) See CAL.
using BEZIER CURVES. Vector images are eas-
ier to change and take up less storage space
computer-based learning See CAL.
than raster images.
computer game Any game recorded on
CD or DVD or accessed from such online
computer graphics interface (CGI) A
standard used for software applying to
sources as the Internet that can be used on
a computer. Computer games are played computer graphics devices such as printers
by manipulating a mouse, joystick, or and plotters.
games pad, or by using the keyboard or
other device in response to the graphics on computer instruction See machine in-
the screen. struction.
42
constant
computer network See network. dicate whether it is the only disk in the
computer, or, if not, whether it is the main
computer-orientated language See or a subsidiary disk. A program may need
low-level language. to know the display’s resolution, how
many different colors it supports, what
computer science The study of com- sort of printer is available, how many disks
puters and their application. This is a are available, and so on. Once the program
broad discipline covering the theory and has been configured, the initial settings will
practice of hardware and software design. be stored and used the next time the pro-
gram is run. All kinds of programs, from
computer system (computing system) word processing programs on microcom-
A self-contained set of computing equip-
puters to mainframe operating systems,
ment consisting of a COMPUTER, or possibly
need to be configured. See also reconfigu-
several computers, together with associ-
ration.
ated SOFTWARE. Computer systems are de-
signed to fulfil particular requirements.
connector 1. A device for connecting
computer word See word. electrical components. Most connectors
are available as male (plug) and female
concatenation /kŏn-kat-ĕ-nay-shŏn/ An (socket) versions and can be used to con-
operation by which two STRINGS are joined nect cables together or cables to devices.
together in a specified order. The result is 2. In software, a logical construction, such
itself a string whose length equals the sum as AND or OR, used for example to con-
of the lengths of the two strings. If one nect conditions in a database query.
string, A, is given by ‘com’ and another 3. In flowcharting, a circular symbol used
string, B, by ‘puter’, then the concatenation to indicate a break, as to another section or
of A and B yields ‘computer’. The concate- page.
nation of A and B is denoted by, for exam-
ple, connect time The time a user is con-
A+B A&B A || B nected to a remote system. This time is
depending on the programming language often used to determine user charges.
used.
console /kon-sohl/ A part of a computer
conditional Taken in some but not all used by an operator to communicate with
circumstances. See control structure; jump. the computer and monitor its operation.
Nowadays the console is usually, and for
conditional jump (conditional microcomputers always, a VDU and a KEY-
branch) See jump.
BOARD considered as a single input/output
device, or possibly a TERMINAL. In a system
conferencing /kon-fĕ-rĕn-sing/ See tele-
with many such devices attached, the con-
conferencing.
sole is the one that interacts with the com-
configuration /kŏn-fig-ŭ-ray-shŏn, -fig- puter when it is first switched on, before
yŭ-/ The particular pieces of hardware the operating system has been loaded and
making up a computer system, and the way normal multiuser access enabled. For con-
in which they are interconnected. See also venience and security it is normally situ-
reconfiguration; configure. ated near the computer’s hardware.
configure To select various optional set- constant A value or an item of data that
tings of a hardware device or a software is fixed or that cannot change. It may, for
program so that it will run to best advan- example, be a numerical constant, like π,
tage on a particular computer. A new disk, whose value must be specified, or it may be
for example, might need a switch set to in- a LITERAL.
43
content-addressable store
44
CORBA
gram can be executed it must be translated Web site so that it can remember some-
into a sequence of MACHINE INSTRUCTIONS. thing about the user at a later date. Ac-
The control unit receives the machine in- cording to the browser being used cookies
structions in the order in which they are to can be refused or can be stored for a short
be executed. It interprets each instruction or long time. In general, cookies help Web
and causes it to be executed by sending sites to serve users better, but their poten-
command signals to the ARITHMETIC AND tial and actual misuse is a current cause for
LOGIC UNIT (ALU) and other appropriate concern.
parts of the computer. The data required
for the execution is moved between MAIN cooperative multitasking See multi-
STORE, ALU, and other portions of the ma- tasking.
chine in accordance with the sequence of
instructions and under the direction of the coprocessor /koh-pros-ess-er/ A micro-
control unit. processing element designed to extend the
The control unit contains REGISTERS, capabilities of the main MICROPROCESSOR in
COUNTERS, and other elements enabling it a microcomputer. For example, it may
to perform its functions. In its simplest have better mathematical processing capa-
form it has an INSTRUCTION ADDRESS REGIS- bilities, including high-speed floating-point
TER, an INSTRUCTION REGISTER, and a regis-
arithmetic and calculation of trigonometric
ter to decode the operation part of the functions. Another design could provide
instruction. It can then operate in a two- superior graphics display capabilities (see
step fetch-execute cycle. In the fetch step graphics accelerator). A coprocessor ex-
the instruction is obtained from main store
tends the set of instructions available to the
and loaded in the instruction register; the
programmer. When the main processor re-
decoder can then determine the nature of
ceives an instruction that it does not sup-
the instruction. In the execute step the in-
port, it can transfer control to a
dicated operation, or operations, are car-
coprocessor that does. More than one co-
ried out, including the necessary references
processor can be used in a system if the
to main store to obtain or store data. In
main processor has been suitably designed.
some cases no reference to memory is re-
quired, as when a JUMP INSTRUCTION oc-
curs. In other cases an additional step is copy To READ data from a source, leav-
required, as in INDIRECT ADDRESSING when ing the source data unchanged, and to
WRITE the same data elsewhere. The physi-
two (or more) memory references are
needed. cal form of the data in this destination may
Control units nowadays can be very differ from that in the source, as occurs
much more complex than this, containing when data is copied from magnetic disk to
additional registers and other functional main store.
units that can, for instance, prepare the
next instruction for execution before the copy protection Methods used by soft-
current one is complete (see pipeline pro- ware companies to prevent unauthorized
cessing). At present the functions of most duplication of their programs and so pre-
control units are accomplished by means of vent software piracy.
MICROPROGRAMS.
CORBA /kor-bă/ (common object request
conversational mode See interactive. broker architecture) A system that al-
lows for the sharing of objects, such as
cookie A small file of information, usu- files, applications, and pieces of programs,
ally containing personalized details, sent over the Internet, regardless of what pro-
by a Web server to a Web browser and sent gramming language they are written in or
back to the server whenever the browser what operating system they are running
makes further requests. In simple terms in- on. CORBA was developed by an industry
formation is put on a user’s hard disk by a consortium known as the Object Manage-
45
core store
ment Group (OMG). See also DCOM; in- the 16-bit microprocessor chips in the Intel
terface. 8086 family was called CP/M-86, but it
was never widely used due to the popular-
core store A form of magnetic storage ity of MS-DOS. It is now obsolete.
widely used for MAIN STORE in computers
built between the mid-1950s and mid- cps /see-pee-ess/ (characters per second)
1970s. It has been displaced in present-day A measure of the rate of output of, say, a
machines by SEMICONDUCTOR MEMORY – character printer, or sometimes a measure
RAM and ROM. Core store consisted essen- of the rate at which data is processed or
tially of tiny rings of ferrite (the brown ma- transferred.
terial used to coat magnetic tapes), which
could be magnetized in either of two ways CPU (central processing unit) See cen-
and were thus each capable of storing one tral processor.
BIT of information.
CR See carriage return.
corrective maintenance MAINTENANCE
that is performed after a fault has been cracker See hacker.
found – in hardware or software – in order
to correct that fault. Compare preventative crash 1. (system crash) A system failure
maintenance. that requires at least operator intervention
and often some maintenance before system
corrupt To alter data stored in a com- running can resume. The system is said to
puter system, usually accidentally, and have crashed.
hence introduce errors. The process is 2. (program crash) A failure in an individ-
known as data corruption. It results in a
ual program that causes it to exit prema-
loss of data INTEGRITY.
turely, to be aborted by the operating
system, or to HANG so that the user must
counter An electronic device whose
abort it. In all cases any changes to data
‘state’ represents a number, and that on re-
files that have not been saved will be lost.
ceipt of an appropriate signal causes the
3. See head crash.
number represented to be increased (or de-
creased) by unity. Only a fixed group of
numbers can be represented. Once all these crawler (spider) A program that visits
numbers have occurred, one after the Web sites, particularly ones that are new or
other, the sequence starts again. The updated, and collects information from
counter is usually able to bring the number pages to enter in the index of a search en-
represented to a specified value, such as gine.
zero. In a computer system a counter can
be used as a timer if the signal received is a CRC (cyclic redundancy check) An al-
clock pulse (see clock), the counter nor- gorithm used to derive a check number
mally counting down from a selected value from data being transmitted. The CRC
and raising an INTERRUPT when it has fin- number is transmitted with the data. By re-
ished. calculating it at the receiving end and com-
paring it to the original, errors in
courtesy copy See cc. transmission can be detected. CRC meth-
ods are also used in other circumstances,
CP/M /see-pee-em/ Trademark An OPER- for example in the production of CDs so
ATING SYSTEM produced by Digital Re- that they are virtually error free.
search in the late 1970s. It was intended for
use on microcomputers with floppy disks crop 1. To insert marks on a document
and the 8-bit Intel 8080 microprocessor, or to be printed showing where the paper can
either of the later Intel 8085 or Zilog Z80 be trimmed to produce the correct page
8-bit chips. A version of CP/M for use with size.
46
cybernetics
2. To reduce the size of an image by cutting current instruction register (CIR) See
off one or more margins. Most graphics instruction register.
applications allow images to be cropped.
cursor A symbol on a VDU screen that
cross assembler An ASSEMBLER that indicates the ‘active’ position, for example
runs on one computer, producing an object the position at which the next character to
module to run on a different computer. It is be entered will appear. The cursor’s shape
usually used to generate software for EM- usually varies with the content: for exam-
BEDDED COMPUTER SYSTEMS.
ple, the active position in a word-process-
ing document is often indicated by a thin
cross compiler A COMPILER that runs on flashing vertical line, whereas a WINDOW
one computer, producing an object module displaying the contents of a directory as
ICONS might use an arrow.
to run on a different computer. It is usually
used to generate software for EMBEDDED
cursor keys Keys on a KEYBOARD that
COMPUTER SYSTEMS.
can be used to move the CURSOR to a new
position on a display screen. They include
cross-post To copy a message from one the ARROW KEYS.
communications channel, such as a news-
group or e-mail system, to another. cut To delete selected text from a docu-
ment displayed on a screen. The text is re-
CRT (cathode-ray tube) See display. tained on the CLIPBOARD until another piece
of text is cut or the computer is closed
cryptography The protection of a mes- down. It can then be pasted – i.e. inserted
sage so as to make it unintelligible to any- into the document at a different place or in-
body not authorized to receive it. The serted into another document.
sender of a message renders it into an un-
intelligible form by processing it. This pro- cut sheet feed A technique for feeding
cessing is known as encryption. Many paper into a printer, the mechanism used
techniques are known for the conversion of being called a cut sheet feeder. The feeder
the original message, known as plain text, picks up a single sheet from a pile of sheets
into its encrypted form, known as cipher in a hopper and feeds it into the printer. In
(or cypher), ciphertext, or code. Comput- more sophisticated feeders there may be a
ers usually accomplish this by applying an number of hoppers, for example one for
encryption algorithm to the plain text to headed top sheets, one for plain second
produce a pseudorandom cipher, the en- sheets, and one for envelopes. A multipage
coding being governed by a user-specified letter complete with envelope may thus be
printed. See also tractor feed; friction feed.
KEY. The original message is recovered by
processing the encrypted message. When
cybercafe /sÿ-ber-ka-fay/ 1. A place,
this is done by an authorized recipient
usually a coffee shop or restaurant, where
holding the secret key to the encryption, Internet access is made available to the gen-
the processing is known as decryption. Re- eral public for a charge.
covery of the message is expected to be im- 2. A virtual cafe on the Internet where
possible without prior knowledge of the chats can take place and messages can be
key. posted using BBSs.
CSLIP /see-slip/ (compressed serial line In- cybercash /sÿ-ber-kash/ See e-money.
ternet protocol) A version of SLIP in
which the Internet address information is cybernetics /sÿ-ber-net-iks/ The study of
compressed making transmission quicker. thought processes and control systems,
which draws an analogy between brains
CSS See cascading style sheets. and electronic circuits. It is based on theo-
47
cyberspace
ries of communication and control and the a storage device and the fetch-execute cycle
extent to which they can be comparatively of a CONTROL UNIT.
applied to living and nonliving systems. 2. (cycle time) The minimum period of time
required to complete such a sequence of
cyberspace /sÿ-ber-spayss/ A term origi- events, i.e. the time between the start of
nated by the author William Gibson in his one sequence and the start of the next. The
novel Neuromancer and now used to de-
term is often used specifically to mean
scribe the whole gamut of information re-
STORAGE CYCLE, i.e. the minimum time re-
sources obtainable through the Internet.
Cyberspace is a shared virtual universe quired between successive accesses to a
whose inhabitants, contents, and spaces storage device.
comprise data that is seen, heard, and felt
and through which a virtual-reality user cycle time See cycle.
can move. See also virtual reality.
cyclic redundancy check See CRC.
cycle 1. A sequence of events that is re-
peated regularly and in the same order. A cylinder See disk pack.
piece of hardware may operate in such a
way, examples being the STORAGE CYCLE of cypher /sÿ-fer/ See cryptography.
48
D
paper
hammer
ribbon
print wheel
U
movement of
printing mechanism
q
t y
embossed characters
on spoke ends
Daisywheel printer
49
DASD
used. The quality of the print was high, and order. The organization of the records in a
daisywheel printers were used for printing database, and the methods used to access
letters, documents, etc. the data, are more sophisticated than those
of a data file with no associated DBMS.
DASD (direct access storage device) A Databases are usually more complex
storage device from which data can be ac- than data files, and are often collections of
cessed in any order rather than sequen- data previously held in many separate files.
tially. Magnetic disk drives, CDs, and They are not necessarily large and complex
DVDs are examples of DASDs, whereas however: DBMS software is available on
tape units are not. small as well as large computers. Storing
data as conventional files can limit the
DAT /dat, dee-ay-tee/ See digital audio data’s use to one particular application. A
tape. database provides data that is available to
all users of the system, and may be shared
data /day-tă, dat-ă, dah-tă/ In comput- by a number of different applications.
ing, the basic facts – numbers, digits, Often, a dedicated computer called a data-
words, characters – that are fed into a com- base server provides database services to
puter system (in the required form) to be all users on a network.
stored and processed for some purpose. In A database together with its DBMS and
this case data is regarded as the INPUT to a hardware is often referred to as a database
computer system as opposed to the OUTPUT system. A data item in a database is typi-
of the system, i.e. the results obtained from cally associated with many other items,
processing. It is possible, however, for the
some of which may be in the same record.
results of one process to serve as the data –
Data is usually retrieved by giving values of
the input – to another process.
specified items in order that the database
The word ‘data’ is also used as distinct
system should respond with the values of
from program instructions. In this case
specified associated items. For example, a
data refers to all the OPERANDS that a pro-
system might retrieve the registration num-
gram handles, i.e. the numbers or quanti-
bers of all cars of a given color, make, and
ties upon which ARITHMETIC and LOGIC
OPERATIONS are performed.
year of registration. See also database lan-
guage.
DATA See Basic.
database language Any of a group of
databank /day-tă-bank, dat-ă-, dah-tă-/ languages used for setting up and commu-
A system that offers facilities to a commu- nicating with DATABASES. The instructions
nity of users for the deposit and with- given in a database language are executed
drawal of data on a particular topic, such by the DATABASE MANAGEMENT SYSTEM.
as trade statistics or share prices. The user Each language consists of at least one data
community is usually widespread and the description language (DDL) and at least
databank itself may be a public facility. Ac- one data manipulation language (DML).
cess to a databank may, for instance, be by The DDL is used to define the structure of
some form of NETWORK, such as the Inter- the RECORDS within the database files and
net. The data to be accessed may be orga- the relationships between them. It can be
nized as a DATABASE or as one or more FILES. used to specify what type of information is
to be stored – numbers, characters, dates,
database /day-tă-bayss, dat-ă-, dah-tă-/ etc. – as well as which values are to be used
A set of DATA FILES – i.e. an organized col- for indexing. The DML is then used to
lection of related data – that is defined, ac- copy the data into the database files and
cessed, and managed by a set of programs manipulate and retrieve it in various ways.
known as a DATABASE MANAGEMENT SYSTEM That part of the DML used for retrieving
(DBMS). The RECORDS in a database are the data may be called the query language.
addressable and can be accessed in any An important database language is SQL.
50
data integrity
database management system (DBMS; of data via a keyboard, etc. This reduces
database system) A collection of pro- the work involved and the chance of error.
grams that handle and control all accesses For example, the letters Y and N can be
to a DATABASE and that maintain the in- used for ‘yes’ and ‘no’.
tegrity of the database, i.e. the correctness
of the stored data. It achieves this by carry- data collection The process by which
ing out instructions couched in a DATABASE data from several locations is collected to-
LANGUAGE while checking that their effects gether before it is fed into a computer.
do not violate the integrity rules; any that
do are aborted. A DBMS thus has features data corruption See corrupt.
in common with OPERATING SYSTEMS and
COMPILERS. The DBMS allows simultane- data description language (DDL) See
ous access to the database by a number of database language.
users. A good DBMS is characterized by
the ease and speed with which complex data dictionary A set of descriptions of
searches and retrievals are carried out and the data components of some computer-
the flexibility and power of the way data- based system. It is normally held in the
bases are specified. A single DBMS is often form of a FILE or DATABASE. It gives infor-
capable of managing more than one data- mation about the nature of the data –
base. meaning, relationships with other data,
format, etc. – and its use. A database, for
database system 1. See database man- example, will have a data dictionary. A
agement system. data dictionary is an important tool in the
2. A DATABASE MANAGEMENT SYSTEM to- effective planning of a computer-based sys-
gether with its associated hardware (stor- tem, and in the overall control, storage,
age devices, dedicated computers, etc.) and and use of data in the operational system.
the database or databases that it manages.
data element (data item) An individual
data bus See bus. unit of data.
data cleaning (data vetting) The process data file A FILE containing data, i.e. the
of checking RAW DATA for completeness, numbers, text, etc., upon which operations
consistency, and validity. Any bad charac- are performed by a computer. A data file is
ters, out-of-range values, and inconsisten- often organized as a set of RECORDS. Com-
cies are either removed or brought to the pare program file; database.
attention of someone for a decision.
data flowchart (system flowchart) See
data coding The use of standard abbre- flowchart.
viations or simplified representations in the
recording of data on documents, the input data integrity See integrity.
51
data link
data link The physical medium – tele- conducted mainly in business, industrial,
phone line, optical fiber cable, etc. – by and government organizations whereby
which two locations are connected for the data is collected, stored, and processed on
purpose of transmitting and receiving data, a routine basis in order to produce infor-
together with the agreed procedures (i.e. mation, regularly or on request. The oper-
the PROTOCOL) by which the data is to be ations almost always (but not necessarily)
exchanged and any associated devices or involve computers. A DP system usually
programs. handles large quantities of data organized
in a complex way. Typical applications in-
data logging 1. The recording of all clude production of payslips, accounting,
data passing through a particular point in market research and sales forecasting,
a computer system. The point chosen can, stock control, and the handling of orders.
for example, be in a pathway to or from a
VDU. The record – or log – that is pro- data protection The protection of data
duced can be used to reconstruct a situa- handled in a computer. The term is usually
tion when, say, a fault occurs in the system applied to confidential personal data, i.e.
or an unexpected result is obtained. data concerning a living person who can be
2. The regular sampling of a number of identified from that information, and pos-
quantities, such as temperature, flow rate, sibly including some opinion expressed
and pressure, by a device known as a data about that person. Credit ratings produced
logger. The information is stored within by banks, donations received by charities,
the logger and periodically transmitted to a transactions by mail-order firms, and per-
computer for analysis.
sonal records kept by government agencies
and the police all involve data concerning
data manipulation language (DML)
the individual; for ease of access, ease of
See database language.
updating, saving of space, and other rea-
sons, this kind of information is now usu-
data mining The analysis of large quan-
ally stored in computers.
tities of data to discover previously un-
Legislation exists in many countries to
known patterns. The data may be held in a
database or a DATA WAREHOUSE, and many protect personal data when it is stored in
database management systems provide computers. The aim is to control the po-
such facilities as OLAP that assist data min- tential for misuse of such information. Per-
ing. The use of store loyalty cards, for ex- sonal data could, for example, be extracted
ample, means that a record of each from one stored record and correlated with
customer’s purchases can be built up over data concerning the same person from an-
a period of time; data mining can then other file. The combination of information
identify any patterns, such as favored that could result is considered an infringe-
brands, amount usually spent on each visit ment of privacy.
to the store, etc., which can be used to gen-
erate marketing literature and even special data retrieval The process by which
offers tailored to each customer. data is selected and extracted from a FILE, a
group of files, a DATABASE, or some other
data preparation The conversion of area of memory. See also information re-
data into a coded form that can be read by trieval.
a machine and hence fed into a computer.
The necessity for data preparation has data structure Any of several forms in
been largely removed with the advent of which a collection of related data items can
DIRECT DATA ENTRY. The operation of a be organized and held in a computer. Ex-
keypunch to encode data on punched cards amples include ARRAYS, RECORDS, FILES,
was an example. STRINGS, and TREES. Various operations can
be performed on the data structure as a
data processing (DP) The operations whole and on its individual elements, the
52
debugging
choice depending on the type of data struc- oped by Ashton Tate. The first version was
ture and the programming language. dBase II, which in the early 1980s became
the first widely used DBMS on microcom-
data tablet See digitizing pad. puters. It lost its market dominance in the
1990s and is no longer widely used.
data transfer The movement of data
from one point to another. These points DBMS See database management system.
may, for instance, be storage locations
within a computer system or at either end DCE (data communication equipment)
of a long-distance TRANSMISSION LINE. An intermediary device that enables data
terminal equipment (DTE) to access a com-
data transfer rate The rate at which munications line, as defined in RS232 and
data can be moved between two points, for X25 specifications. One example is a
example from a magnetic disk to main modem, which represents the provider side
store or from one location on a NETWORK of a data communication interface.
to another.
DCOM (distributed component object
data transmission See transmission. model) A version of Microsoft’s compo-
nent object model (COM) specification. It
data type Any one of the kinds of data deals with the communication of compo-
(e.g. integers, real numbers, strings) that nents over Windows-based networks when
can be identified by a VARIABLE, an ARRAY, different parts of an application are split
or some other more complex data object in and run across different machines on the
a programming language. The choice of network.
data type defines the set of values that the
variable, etc., is allowed to take and the op- DDE 1. (dynamic data exchange) A
erations that can be performed on it. technique used by Windows and some
other operating systems that enables two
data vetting See data cleaning. or more applications running simultane-
ously to link data and commands together.
data warehouse A sophisticated data- If data is changed in one application, then
base system holding a large amount of data the same data will automatically be up-
that can, in fact, be made up of several dated in the other application. DDE has
databases spread over several computers at been superseded on Windows systems by
different sites. Fast data searching and fil- OLE and COM.
tering techniques are used to supply infor- 2. (direct data entry) The input of data and
mation to the user, who is unaware of the updating of files directly in an on-line sys-
complexity of the system. In general data tem, usually by typing the data at a key-
warehouses hold a historical record of an board or sometimes by using a data-
organization’s data, which is augmented capture device.
periodically with copies of the current state
of the ‘live’ data. They are optimized for DDR SDRAM See SDRAM.
OLAP rather than OLTP operations.
DDS (digital data storage) See digital
daughterboard /daw-ter-bord, -bohrd/ audio tape.
A circuit board that can be plugged into
another board, thus increasing its function- debugger See debug tool.
ality.
debugging The identification and re-
dB See decibel. moval of BUGS – i.e. localized errors – from
a program or system. The bugs in a pro-
dBase /dee-bayss/ Trademark A rela- gram may, for example, be SYNTAX ERRORS
tional database management system devel- uncovered by the compiler, RUN-TIME ER-
53
debug tool
54
descending order
55
desktop
set of data in order starting with the largest DFS See distributed file system.
character value, as in Z to A or 50 to 1.
DHCP (Dynamic Host Configuration
desktop The general screen background Protocol) A protocol that allows a com-
on which windows appear and icons can be puter on a network to obtain various con-
placed by the user in Apple Macintosh and figuration parameters from a server. This
Windows systems. server (a DHCP server) can ensure that
these parameters allow the computer to
desktop computer A compact micro- operate properly in the particular network
computer in which the main components, environment and that it does not interfere
such as the CPU, keyboard, and monitor, with other machines on the network. The
can be placed on top of a desk but are too most important use of DHCP is to allocate
large to be portable. IP ADDRESSES dynamically from an organi-
zation’s available pool. The alternative is
desktop publishing (DTP) The use of a to configure each machine manually to use
computer system to produce professional- a specific IP address, which is not only less
standard page layout, including the choice efficient (more IP addresses are required)
of fonts and type size and the inclusion of but is also prone to such errors as giving
pictures. Illustrations can be created in two machines the same address.
graphics software programs and artist’s
images, photographs, and other pictures DHTML See dynamic HTML.
can be scanned, converted to a suitable for-
mat, then displayed, scaled, and cropped.
diagnostic routine Part of a program,
DTP software usually produces its output
or a sequence of instructions called by a
in a suitable form for printing, such as
program, that is entered as a result of the
PostScript files.
detection of some condition causing an
ERROR. A diagnostic routine may analyze
desktop video The use of a personal
the cause of the error, or provide informa-
computer to display video images from a
tion that can be used for this purpose. It
video tape, a DVD, a file stored on the
computer or downloaded from the Inter- might attempt to isolate the cause of the
net, or directly from a video camera. error to a particular piece of hardware or
software.
destination The drive, folder, or direc-
tory to which a file is being sent, or the dialog box A small box that is displayed
memory address or location in a document on the screen and is an essential part of
or program to which data is being sent. GUI programming. It contains a message
that needs a reply from the user. This may
developer A company or individual that be as simple as clicking with the mouse on
produces software. This involves the analy- a Yes/No selection, or entering a few char-
sis of requirements, defining the applica- acters of data as required.
tion structure, programming, testing,
modification, documentation, implemen- dial-up A temporary connection be-
tation, and issue. tween machines, established over a stan-
dard telephone line.
device A peripheral or other active part
of the electronics of a computer system. dictionary See data dictionary.
device driver A short program that pro- Difference Engine A mechanical com-
vides the interface between a computer and puting device designed by Charles Babbage
a device such as a printer, disk drive, or in 1823 to produce mathematical tables
mouse and controls the detailed operation but never completed. Some of the ideas
of the device. See also plug-and-play. used in the Difference Engine were incor-
56
digital versatile disk
porated by Babbage in the design of his ables. Usually only binary variables are in-
ANALYTICAL ENGINE. volved (i.e. quantities that can take either
of two values), and the term binary logic
digit /dij-it/ Any of the numerals (or pos- can then be used. The components consist
sibly letters) used in a particular NUMBER of LOGIC GATES and LOGIC CIRCUITS.
SYSTEM. Decimal notation uses 10 numer- 2. The methods and principles underlying
als: this implementation of logic operations.
0, 1, 2, 3, 4, 5, 6, 7, 8, 9
Binary notation uses the numerals 0 and 1. digital recording The storage of infor-
Octal notation uses the numerals 0 to 7. mation such as text, graphics, video, or
Hexadecimal notation uses 16 digits: 10 sound in binary coded form (a string of 1s
numerals (0–9) and 6 letters (A–F). and 0s) that can be physically represented
on a particular medium. The term is used
digital audio tape (DAT) A magnetic especially for such recordings of video and
audio-tape storage medium for recording sound, to contrast them with recordings
digitally coded audio information. The se- that use analog technologies.
ries of DDS (digital data storage) formats
(DDS, DDS2, DDS3, DDS4, and DDS5) digital signal An electrical SIGNAL whose
builds on DAT and adapts it for computer voltage at any particular time will be at any
storage. Tapes using the DDS standards one of a group of discrete (distinct) levels.
are used for backing up computer systems. The voltage therefore does not vary contin-
They are 4 mm wide and varying lengths uously (unlike an ANALOG SIGNAL). In gen-
from 90 m upwards offer different capaci-
eral there are just two discrete levels: the
ties, with each standard providing a greater
voltage jumps back and forth between the
recording density than its predecessor.
two levels over a period of time; such a sig-
nal is often called a binary signal. (A digi-
digital computer A computer that ac-
tal signal is therefore usually but not
cepts and performs operations on discrete
necessarily a binary signal.) The LOGIC
data, i.e. data represented in the form of
GATES and LOGIC CIRCUITS used in comput-
combinations of digits, letters, or other
CHARACTERS. Compared with an ANALOG
ers handle two levels of voltage, and thus
COMPUTER, a digital computer is much
their inputs and outputs are binary signals;
more versatile and hence much more the high level of voltage is usually used to
widely used. It is thus generally referred to represent binary 1, with the low level rep-
simply as a computer. See computer. resenting binary 0.
57
digital video disk
digital video disk See DVD. erated. The digitizing process is activated
by one or more switches or buttons on the
digital video interface (DVI) A hard- pen. The thickness, color, and/or opacity
ware-based technique for compressing and of lines is governed by the pressure applied.
decompressing full-motion video, audio,
and computer data on a CD. The technique DIL switch /dil/ (dual in-line switch) A
is also referred to as digital video interac- device similar in form to a DIP, but instead
tive. DVI is a hardware-only CODEC. of an integrated circuit the package con-
tains a row of small switches making or
digitize /dij-i-tÿz/ To transcribe any con- breaking the circuit between opposite pairs
tinuously varying (analog) source of data, of legs. DIL switches are used for setting
such as sound, into a digital (binary coded) the default state of some printers, termi-
form so that it can be directly processed by nals, etc.
a computer. The term is also applied to the
scanning of images or OCR of text. DIM /dim/ See Basic.
digitized signal The representation of a dimension See array.
continuously varying signal in a digital
form. The digitized signal has values that DIMM /dim/ (dual in-line memory mod-
are identical to those of the continuously ule) A memory board on which memory
varying (i.e. analog) signal but only at dis- chips are mounted and separate connector
crete (separate) instants of time. As a re- pins are on both sides of the circuit board.
sult, the voltage or current changes in steps
This increases the amount of memory that
between discrete (distinct) values. For ex-
can be plugged into a single connector and
ample, if the voltage of an analog signal is
also increases the size of the data path for
measured at discrete intervals, then the se-
faster data transfers. See also SIMM.
quence of measured values is a digitized
signal; the analog signal is said to have
dingbat /ding-bat/ 1. A small graphical
been sampled so as to produce the digitized
element used for decorative purposes.
signal.
2. A font that is composed of a set of sym-
digitizer A device to change information bols rather than alphanumeric characters.
into a digital form.
DIP /dip/ (dual in-line package) An INTE-
GRATED CIRCUIT contained in a rectangular
digitizing pad (digitizing tablet; data
tablet; graphics tablet) A flat surface that plastic or ceramic package with a row of
can be placed on a desk and is used to- metal legs down each of the long sides (see
gether with a penlike device for the input of diagram). The legs can either be soldered
data to a computer graphics system. into holes in a PRINTED CIRCUIT BOARD or in-
(Larger surfaces are also available and are serted into a CHIP SOCKET. A SIP has only
known as digitizing tables or boards.) The one row of pins. See illustration overleaf.
position of the pen on the digitizing pad
can be accurately and rapidly located by direct access RANDOM ACCESS to a stor-
any of a variety of methods; it is measured age device, such as a magnetic disk or CD.
in terms of the x and y coordinates of the
point of contact, i.e. the horizontal and direct-access file See file.
vertical distances of the pen from one cor-
ner of the digitizing pad. The position is direct addressing (absolute addressing)
thus in a digital form that can be fed into An ADDRESSING MODE in which the address
the computer. When the pen is moved by specified in a MACHINE INSTRUCTION is the
hand over the surface, closely spaced posi- actual address to be used, i.e. the MACHINE
tions can be measured. A DIGITIZED SIGNAL ADDRESS of the location to be accessed is
representing the path of the pen is thus gen- given explicitly. The address specified is
58
disk
package containing
integrated circuit
then called a direct address. See also ad- ries, and there are one or more shared di-
dress. rectories. Units of removable media (CDs,
DVDs, floppy disks, etc.) each have their
direct data entry See DDE. own directory. Directories are used by the
operating system to locate files when given
direct memory access (DMA) A their names, and by computer users to keep
method of transferring data between main track of what files are available. See also
store and other storage devices without in- folder.
volving the CENTRAL PROCESSOR. Without
DMA, every byte or word of data has to be DirectX /dă-rekt-eks/ A Windows appli-
individually transferred from main store to cation program interface (API) for creating
the central processor, and then written to and manipulating graphic images and mul-
the destination device (or vice versa). With timedia effects. It is used in games and in
DMA, the central processor initiates the active Web pages. Games specifications
transfer by informing an I/O processor of will often require a certain level of DirectX
the source and destination of the data and to be implemented before the game can be
the total number of bytes or words to be played.
transferred. The I/O processor then per-
forms the transfer, and informs the central disable To switch off a device or prevent
processor (probably by means of an INTER- the operation of a particular function of a
RUPT) when the transfer is complete. The device or a particular feature of a program.
central processor can continue to operate Compare enable.
during the DMA.
disassembler A program that attempts
directory A FILE in a computer system to translate machine code back into ASSEM-
containing a list of filenames, their loca- BLY LANGUAGE, i.e. it performs the reverse
tions on BACKING STORE, and their size, as function of an ASSEMBLER. It is used as an
well as other information such as creation aid to DEBUGGING. It is only the one-to-one
date, author, date of last access, and FILE relationship between machine instructions
PROTECTION code. A computer system may and assembly language instructions that
have many directories, usually organized in makes this process possible. It does not
a HIERARCHICAL FILE SYSTEM where directo- work with high-level languages.
ries can contain other directories (subdirec-
tories) as well as data files. On a discrete data See computer.
MULTIACCESS SYSTEM each user normally
has a directory, and possibly subdirecto- disk See magnetic disk; optical disk.
59
disk array
disk array A hardware device consisting vice. In such a controlled environment pre-
of several HARD DISKS controlled by a disk cision engineering is possible that gives fast
array controller, which handles all interac- access times and very high recording densi-
tion with the rest of the computer system. ties. A single disk 3.5 inches in diameter
The array is thus a BLACK BOX and the log- can now hold many gigabytes of data.
ical structure it presents to the computer However, hard disks are fragile and must
need not match its physical structure; for be protected from physical damage.
example it can appear to be one large disk. In floppy disk drives the disks are re-
An important use of disk arrays is RAID, a movable and do not require such careful
service that is usually implemented by the protection. Disk plus protective envelope is
disk array controller. fed into the disk drive by hand through a
slot, and is automatically mounted on the
disk cache See cache. rotation mechanism. Rotation speeds are
considerably lower than for hard disk dri-
disk cartridge An obsolete type of HARD ves. The read/write head or heads operate
DISK consisting of a single disk permanently through slots in the envelope. Floppy disks
housed inside a protective plastic cover. A were formerly used extensively on micro-
disk cartridge was used in a specially de- computers but have now largely been su-
signed DISK DRIVE, from which it could be perseded by recordable CDs and DVDs.
removed and replaced by another car- In a disk drive there is usually one
tridge. It was thus an EXCHANGEABLE DISK read/write head per recording surface, ei-
STORE. ther touching it (in the case of a floppy
disk) or very close to it (in the case of a
disk controller A printed circuit board hard disk). The head is normally mounted
containing the electronics to control the on an arm that moves radially (i.e. towards
detailed operation of magnetic disks and and away from the disk center). To read or
carry out requests from the OPERATING SYS- write data, the head is accurately posi-
TEM to read or write data. A single disk tioned by the disk drive over the required
controller can control one or more disks. track, and then waits until the right sector
rotates into place underneath it. When the
disk crash See head crash. head is to write data, it receives an electri-
cal signal coded with the data and converts
disk drive (disk unit) A peripheral de- it into magnetized patterns along the spec-
vice that has a mechanism for rotating one ified sector of a specified track. (The head
or more MAGNETIC DISKS at constant high is an electromagnetic device.) The data is
speed, and devices known as read/write encoded in one of the appropriate disk
heads (plus associated electronics) for writ- FORMATS. For reading data, the head senses
ing and reading data on the spinning the magnetized patterns in the specified
disk(s). Data is recorded on both sides of a sector and produces a corresponding elec-
disk, along concentric tracks in the mag- trical signal.
netic coating. Items of data are stored and A large disk will probably be used to
retrieved by the process of RANDOM ACCESS. store many FILES. The allocation of storage
Two basic types of disk drive are avail- space on the disk for each file is handled
able. One type uses HARD DISKS and is de- automatically by the operating system.
signed for high performance and large
storage CAPACITIES. The other type is de- diskette /diss-ket/ See floppy disk.
signed for low purchase price and uses in-
expensive storage media in the form of disk format See format; magnetic disk.
FLOPPY DISKS.
In disk drives using hard disks, the disks disk pack An obsolete form of disk stor-
are fixed in position and, with the age used in a specially designed DISK DRIVE,
read/write heads and supporting machin- from which it could be removed and re-
ery, are hermetically sealed inside the de- placed by another pack of the same type. It
60
distributed processing
protective disk spindle read/write
mechanism
recording
surface
of disks
read/write
protective disk heads
was thus an EXCHANGEABLE DISK STORE. It while the term disk drive is restricted to the
was an assembly of identical MAGNETIC rotational mechanism.
DISKS permanently mounted on a single
spindle. The disks were usually 14 inch display 1. To make information visible
(350 mm) hard disks. Their number varied on a screen.
from 5 to 12 per pack. A similar-sized disk 2. The device that enables information –
used for protective rather than recording textual or pictorial – to be seen but not per-
purposes was fitted to the top and bottom manently recorded. The most widely used
of the assembly; in a 6-disk pack there were technology involves a cathode-ray tube
thus 10 recording surfaces. When not (CRT): in most cases the image is formed
mounted on a disk drive the assembly was and changed as a beam of electrons contin-
kept in plastic covers to protect it from ually traces a pattern of horizontal lines (a
damage and dust. The bottom cover was raster) on the screen, as occurs in a domes-
removed just before mounting the pack; tic TV; in some CRT devices the image is
the top cover could only be removed when formed by ‘drawing’ individual lines on the
the pack was mounted. Once the pack was screen, of any length and at any angle.
mounted, the read/write mechanism The CRT display is rapidly being dis-
moved automatically into position and the placed by the flat panel display, a thin
pack was set into rotation. A read/write panel with a FLAT SCREEN that is both more
head was provided for each recording sur- compact and more rugged. Many of these
face. The heads were carried on a single devices use ACTIVE MATRIX DISPLAYS; an-
comblike mechanism that was moved radi- other popular technology is the plasma
ally between the disks. The heads thus all panel (or gas panel) display, where electri-
moved together, and could be positioned cal discharges through a gas form the
over the same track number on each of the image on the panel. See also LCD; LED dis-
recording surfaces. This set of tracks could play.
therefore be written to and read without
moving the heads; the tracks formed and distributed file system A file manage-
were referred to as a cylinder. The ACCESS ment system in which files may be kept on
TIME to related records could thus be mini- several computers connected by a LAN or
mized by writing them in a cylinder. WAN and a number of users may have
shared access to these files.
disk unit A disk drive. The term disk
unit is sometimes used to mean the rota- distributed processing The processing
tional mechanism plus read/write heads of data in a system in which a number of
and associated electronics as a whole, independent but interconnected computers
61
distributed system
can cooperate. The system itself is known DNS See domain name system.
as a distributed system, an example of
which is a LOCAL AREA NETWORK. The doc /dok/ The file extension that identi-
processors involved may be situated at dif- fies a word processing document. It is used
ferent places and are connected by commu- as the default by Microsoft Word for doc-
nication lines. They normally have their ument files.
own peripherals – console, disks, printers –
so that ‘local’ data can be processed and docking station A set of equipment
‘local’ decisions made. Data with a wider consisting of a monitor, a keyboard, and a
application can be exchanged over the power supply into which a laptop can be
communication lines. slid, effectively combining to form a desk-
top computer.
distributed system See distributed pro-
cessing. document Any piece of data created by
a program and saved as a file. The term is
dithering A technique for mixing the used especially of text formatted in some
pixels of available colors to simulate miss- way. Sometimes document information
ing colors in a picture. For example, it is such as the author, date of last change,
used to simulate shades of gray in mono- number of words, etc., is included.
chrome (black and white) pictures. There
are several ways of dithering. The tech- documentation The written informa-
nique is also used in anti-aliasing in order tion describing the operation, structure,
and use of hardware or software. It may be
to make jagged lines appear smoother on
in the form of printed manuals. Alterna-
the screen.
tively it may be held on BACKING STORE and
be accessible from a computer CONSOLE or
DLL (dynamic linked library) A Mi-
TERMINAL; this is called on-line documenta-
crosoft Windows standard for sharing
tion. Software documentation normally in-
code between programs. A DLL file (usu-
cludes a tutorial guide to new users, a
ally given the extension dll) consists of a
reference section, an explanation of error
collection of PROCEDURES and FUNCTIONS.
messages, an installation guide, and per-
Rather than copies of these being incorpo- haps a small detachable summary card.
rated in every program that uses them, a Hardware documentation is similar but in-
single copy is shared between all programs cludes a technical specification and a trou-
and only loaded into MAIN STORE when re- ble-shooting and maintenance-engineers’
quired during execution. manual. Good documentation is crucial to
the success of any computer system.
DLT (Digital Linear Tape) Trademark
A type of TAPE CARTRIDGE that was the tape document object model See DOM.
backup medium of choice for microcom-
puter networks from the mid-1990s. A sin- document reader An INPUT DEVICE that
gle DLT cartridge can hold up to 40 Gb of reads data directly from a paper document
uncompressed data; usually data compres- and feeds it, in coded form, into a com-
sion allows up to 80 Gb to be stored. Its puter. The document reader has to recog-
successor, SDLT (Super DLT), can store nize characters on the document, or to
up to 300 Gb (600 Gb compressed) with sense marks on it. It operates by OCR (opti-
future capacity increases planned. The cal character recognition), MICR (magnetic
DLT format was developed in the 1980s by ink character recognition), or OMR (optical
Digital Equipment Corporation and is now mark reading). The documents can gener-
owned by Quantum Corporation. See also ally be read not only by the input device
LTO. but also by people, and may involve more
than one reading process, for instance
DMA See direct memory access. OCR and OMR.
62
double click
domain name The human-oriented ver- dot matrix printer A PRINTER that cre-
sion of the address of a network connec- ates each character as an array or matrix of
tion on a TCP/IP network, particularly the dots. The dots are usually formed by trans-
Internet. Domain names are hierarchical ferring ink by mechanical impact. The de-
with parts separated by dots. For example, vice is almost always a CHARACTER
www.nasa.gov is the domain name of the PRINTER, i.e. it prints a character at a time.
US space agency. The highest subdivision It typically has a print head containing 9,
of the domain name is called the domain. 18, or 24 thin rods, known as needles. The
This may identify the type of organization needles are selectively operated as the head
(e.g. .gov for government, .com for com- moves along the line, thus building up each
mercial companies, .edu for educational es- character as required. In many printers the
tablishments, or .org for noncommercial number of characters per inch (cpi) can be
organizations). It may also identify the varied. Generally between 100–1000 char-
country in which the organization is lo- acters per second can be printed, with
cated (e.g. .uk for United Kingdom or .fr slower speeds giving better quality output.
for France). Each domain name has an as- Dot matrix printers can print a large se-
sociated unique IP ADDRESS, to which it is lection of shapes and styles of letters and
translated by DNS. digits, and may also print Arabic charac-
ters or the idiograms of oriental languages.
domain name system (DNS) A cooper- In addition to characters it is possible to
ative Internet service that provides a data- produce diagrams, graphs, etc.
base of host and domain names. DNS is
used to match domain names to their asso- dot pitch The measure in millimeters
ciated IP ADDRESSES. When a request is (the smaller the number the better) of the
made to a particular Web site it is sent to sharpness of a monitor display. Most CRT
the nearest DNS server, which, if it cannot monitors will have a dot pitch of between
translate the name, will ask another DNS .25 and .28. The difference between a dot
server, and so on. On finding a match the (as in dot pitch) and a pixel is that a pixel
IP address is returned to the browser, is mapped to dots on the screen. When a
which then tries to connect to that IP num- monitor is set to lower resolutions, one
ber. pixel encompasses many dots.
63
double-length word
with the CURSOR pointing at an ICON or draft quality A printing mode available
other object on the screen. In general dou- on most printers where quality of output is
ble-clicking initiates the default action for sacrificed for increased printing speed.
that object: programs are executed, data
files are opened in their associated applica- drag To hold the mouse button and move
tion, etc. Compare click. the mouse pointer to a new position. This
can be used to highlight text in a document.
double-length word See precision.
drag and drop To move or copy an ob-
double precision See precision. ject using the mouse by pointing at the ob-
ject, holding down the mouse button,
downlink The transmission link from a moving the mouse pointer (and the object)
communications satellite to a ground sta- to the new location, and then letting go of
tion. Compare uplink. the mouse button.
64
dump
defines the structure of a set of similar doc- dual in-line memory module See
uments. It uses a special syntax to detail DIMM.
such matters as which ELEMENTS are al-
lowed in the document, in what order they dual in-line package See DIP.
must appear, and how many of each type
can be used together. HTML, for example, is dumb terminal See terminal.
defined in SGML by a DTD detailing the
structure of Web pages. Documents that dump To copy or make a copy of some
are correctly structured are said to con- or all of the contents of a storage device at
form to their DTD, and DTDs can be used a particular instant in order to safeguard
to process or display any conforming doc- against loss of data, to check for errors in a
ument using appropriate software. In program, or for some other purpose. The
XML, DTDs are being superseded by data copied is called a dump. For example,
SCHEMAS. The abbreviation DTD is some- in a system handling large numbers of
times used for ‘document type declaration’ users’ files stored on magnetic disks, the
– a statement at the start of the document contents of the disks may be dumped peri-
assigning the document type definition. odically on magnetic tape. This provides a
reference copy of the data in the event of,
DTP See desktop publishing. say, accidental overwriting or damage of
the disks. Such a dump, which can be used
DTV See desktop video. to restore the data automatically, acts as a
BACKUP. Again, when a system CRASH oc-
dual-core processor A CHIP that con- curs, a human-readable version of the con-
tains two MICROPROCESSORS. A computer tents of main store and possibly backing
equipped with a dual-core processor can store might be produced. This can be stud-
perform MULTIPROCESSING. Functionally ied to try and determine the immediate
equivalent to two single-core chips, a dual- cause of the crash. Dumping can also be
core processor is faster, uses less power, performed at some selected point in the ex-
and takes up less space. Dual-core proces- ecution of a computer program, usually
sors began to appear in mainstream PCs in during TESTING or DEBUGGING of the pro-
2005 and are expected to be the forerun- gram, or may occur when a program
ners of multi-core processors. crashes or is aborted.
DSL TYPES
65
duplex transmission
1 1 4.7 Gb 2 hours
1 2 8.5 Gb 4 hours
2 1 9.4 Gb 4.5 hours
2 2 17.0 Gb more than 8 hours
66
E
Easter egg An unexpected message, pic- are bipolar in nature and are all character-
ture, extra function, etc., hidden in a pro- ized by very high SWITCHING SPEEDS, but
gram and designed to appear when a have high power requirements and a low
combination of circumstances occur, for PACKING DENSITY compared with CMOS. See
example pressing a particular set of keys. also TTL.
Easter eggs are sometimes put in applica-
tion programs for amusement by program- e-commerce (e-business) Business activ-
mers. ities such as advertising, buying and sell-
ing, and the servicing of customers
EBCDIC /eb-see-dik/ (extended binary conducted through the Internet.
coded decimal interchange code) A CODE
developed by IBM and used for the inter- edge connector Part of a PRINTED CIR-
change of data between equipment manu- CUIT BOARD (PCB) where a number of the
factured by or associated with IBM. metallic conducting tracks meet the edge of
EBCDIC encoding produces coded charac- the board, at right angles, to form the male
ters of 8 bits, and hence provides 28, i.e. half of a plug and socket. The tracks are
256, different bit patterns. These 256 char- broadened, thickened, and usually gold-
acters form a CHARACTER SET, consisting of
plated to provide a good electrical contact;
letters, digits, special characters such as
the PCB itself provides the necessary
punctuation marks, and control charac-
strength and rigidity. A single edge connec-
ters.
tor may have a hundred or more individual
connections, half on each side of the board.
echo To reflect transmitted data back to
The female half of the connector, the
its point of origin. (As in ordinary speech
the word is also used as a noun.) For ex- socket, consists of a number of sprung
ample, characters typed on a keyboard will metal contacts, embedded in plastic, which
not appear on the display unless they are make contact with the corresponding
echoed. See also echo check. tracks on the PCB when the two are
pressed together.
echo check A way of establishing the
accuracy achieved during the transfer of EDI (electronic data exchange) A stan-
data between, say, devices in a computer dard format for exchanging business data
system or computers in a NETWORK. When using computer systems; for example, sales
the data is received it is stored and is also information, orders, and invoices.
sent back – echoed – to its point of origin.
The returned data is compared with the edit To create or change text, graphics,
original data and errors can then be de- etc., in a file by deleting, moving, copying,
tected. See also check. adding, formatting, etc. See also word pro-
cessing.
ECL (emitter-coupled logic) A family of
LOGIC CIRCUITS that are all fabricated with editor See text editor; link editor.
a similar structure by the same INTE-
GRATED-CIRCUIT techniques. The circuits EDO RAM (extended data out dynamic
67
EDP
random access memory) A type of fast EIDE is preferred to SCSI in many areas.
DYNAMIC RAM used in the 1990s. Today virtually all PCs have two EIDE
controllers built into the motherboard and
EDP Electronic data processing. See so can support up to four EIDE devices.
data processing.
EISA /ee-să/ (extended industry standard
EDSAC /ed-sak/ Electronic Delay Stor- architecture) A 32-bit data path com-
age Automatic Calculator. See first genera- puter expansion bus. This maintained ISA
tion computers. standards but gave additional features. It is
now obsolete.
EDVAC /ed-vak/ Electronic Discrete
Variable Automatic Computer. See first electric cable See coaxial cable; twisted
generation computers. pair; ribbon cable.
68
emulation
for goods and services through the Inter- embedded Describing a command, code,
net. See also e-commerce. file, etc., that is inserted into another piece
of data while remaining distinct from it in
element In an SGML or XML document, a some way. For example, when a spread-
‘tagged’ section that indicates a specific sheet is embedded in a word-processing
type of content. In this book, for example, document, it is printed as if it were part of
the headword of each entry is contained in the word-processing document but is still
an ‘hw’ element, which for this entry is edited using a spreadsheet rather than a
keyed <hw>element</hw>. Elements can word-processing application.
contain text, other elements, or both.
embedded computer system Any sys-
elevator The small movable box in a tem that uses a computer as a component
scroll bar that, when dragged with the dedicated to a particular task. Examples
mouse, alters the part of the text or image are programmable washing machines, ar-
that is displayed on the screen. cade video games, and satellite navigation
systems. It is not possible for the operator
e-mail (electronic mail) Messages sent of a washing machine to reprogram it to
between users of computer systems, the play games, whereas a general-purpose
computer systems being used to transport computer system can perform a series of
and hold the messages. Sender and recipi- unrelated tasks under the control of the op-
ent(s) need not be on the same computer to erator.
communicate, or be ON-LINE at the same
time. The mail-handling program will have e-money (electronic money; cybercash)
Money exchanged through the Internet.
facilities for typing in messages, selecting
one or more recipients, and checking for
incoming mail. It will also be able to reply
emoticon /i-moh-ti-kon/ A sequence of
text-based characters often used in e-mails
to messages, forward copies of incoming
and online chat. When viewed sideways
messages to other people, and store copies
they are supposed to represent facial ex-
of sent and received messages for future
pressions to give the reader a sense of how
reference or action. A single message may
a writer is feeling. For example :-) (known
possibly combine text, graphics, and other
as a smiley) is used to show the writer is
forms of information and can contain one happy or has just written something in-
or more data files as attachments. It is also tended to be humorous, while :-( shows
possible to send e-mail using certain types that the writer is sad.
of cellular telephone. See illustration over-
leaf. empty string (null string) A STRING with
no characters. Not all programming lan-
e-mail marketing The use of E-MAILS as guages cater for empty strings, and the con-
a direct-marketing tool. Its main advantage cept, although useful, should be treated
is that is it very cheap since a single e-mail with care.
can be sent to thousands of recipients. Each
e-mail therefore does not have to attract emulation The imitation of all or part
much extra business to justify its cost. It is of one computer system by another com-
easy to analyze an e-mail’s effectiveness by puter system such that the imitating system
measuring such responses as CLICK executes the same programs, accepting
THROUGHS. However, many recipients find identical data and producing identical re-
such e-mails highly irritating, especially sults (but not necessarily in the same way),
when they are unsolicited (see spam), and as the system imitated. A device or pro-
software has been developed to filter them gram used in producing an emulation is
out. The use of e-mail marketing tech- called an emulator. A particular emulation
niques for PHISHING and other types of could be used as a replacement for all or
fraud has given it a bad reputation. part of the system being emulated, and fur-
69
user creates
a message
which may contact ISP
have various
attachments
acknowledge
CLIENT EM AIL SOFTWARE
I have message
to send
SM TP SERVER at ISP
send it now
or
too busy, send it later
send message
please confirm
receipt
confirmation
router
other gateway
recipients
CLIENT EM AIL SOFTWARE
DNSs
SMTP router
SERVER
recipient
logs onto
POP the POP
SERVER server and
asks for
any mail
E-mail flowchart
70
entry point
thermore could be an improved version. contrasted with the FRONT END, which in-
For example, a new computer may emulate teracts with the user. For example, a data-
an obsolete one so that programs written base engine is the part of a database
for the old one will run without modifica- application that manages the database it-
tion. self; it might be used with a variety of front
In contrast with a SIMULATION, an emu- ends, each tailored to the needs of a specific
lation is usually a realistic imitation; a sim- application. (Often, the engine is a DBMS
ulation may be no more than an abstract running on a server, while each client com-
model. puter runs its own front end.) Similarly,
Web search engines communicate with
enable To switch on a device or to select users directly via an HTML front end but
and activate a particular function of a de- can also be used to provide the search fa-
vice or a particular feature of a program. cility for other Web sites.
Compare disable.
enhanced IDE See EIDE.
encapsulated PostScript See EPS.
ENIAC /en-ee-ak/ Electronic Numerical
encapsulation 1. In object oriented pro- Integrator and Calculator. See first genera-
gramming, the inclusion within a program tion computers.
object of all the resources needed for the
object to function and interact with the rest enter key See return key.
of the program, which treats the object as
a BLACK BOX. entity In an SGML or XML document, a
2. In telecommunication, the inclusion of means of representing of an object that is
one data structure within another so that not simple text. Entities have the form
the first data structure is hidden for the &entity; and are often used to represent
time being. For example, in layered proto- non-ASCII characters; for example, the en-
cols each layer adds header information to tity α often represents a Greek
the protocol data from the previous layer. alpha. Entities thus allow SGML and XML
documents to use only ASCII characters,
encoder A means by which data is con- which makes them extremely portable. In
verted into a coded form. It may be a hard- the above example the correct character
ware device, such as a KEYBOARD, or a piece for a Greek alpha would only have to be
of software. See also code. substituted for the entity when the docu-
ment was typeset or displayed on screen in
encoding See code. a WYSIWYG form. Entities also have
other uses: for example they can be simple
encryption /in-krip-shŏn/ See cryptog- shortcuts that expand into lengthy se-
raphy. quences of text (e.g. &sgml; could be a
keyboarding shortcut for ‘Standard Gener-
END /ee-en-dee/ See Basic. alized Markup Language’), or they can
represent other files that SGML- or XML-
end user The person for whom a hard- aware software will insert into the docu-
ware or software product is designed as op- ment in place of the entity.
posed to the developer, installer, and
servicer of the product. entry point (entry) The point to which
control is passed when a program or a sec-
Energy Star A symbol attached to sys- tion of a program – a PROCEDURE, SUBROU-
tems and components with low power con- TINE, or FUNCTION – is called. The entry
sumption. point is the first instruction in the program
or program section to be executed (or more
engine (back end) The part of an appli- specifically is the ADDRESS or LABEL of this
cation that manages and controls data, instruction). There may be more than one
71
ENUM
entry point in a program or program sec- tion of the size of the output (the bounding
tion. box). This extra information allows EPS
files to be treated as graphics files by DTP
ENUM (Electronic Numbering) A pro- software. Their content is incorporated in
tocol, defined in RFC 2916 (2000), that al- a normal PostScript file for printing, type-
lows telephone numbers to be mapped to setting, etc.
Internet URLS. It is used in VOIP.
EQ gate See equivalence gate.
environment 1. The combination of
hardware and software in a computer. equivalence A LOGIC OPERATION com-
2. A collection of system- and user-speci- bining two statements or formulae, A and
fied settings that can be used to define the B, in such a way that the outcome is false
context in which a program operates. Each when either A or B is true but not both;
setting is held in a separate named environ- otherwise the outcome is true. The TRUTH
ment variable, whose value can be exam- TABLE is shown in the diagram, where the
ined by the program. For example, the two truth values are represented by 0
current USERNAME might be held in an envi- (false) and 1 (true). The operation can be
ronment variable called ‘USERNAME’, al- written as A ≡ B.
lowing the program to run differently
depending on who is using it.
72
error message
error A discrepancy between the value structed by an encoder using a specific rule.
or condition that is calculated, observed, A decoder at the other end of the channel
or measured and the value or condition detects any departure from this construc-
that is known to be accurate or correct, or tion, i.e. any errors, and automatically cor-
that has been specified as being accurate or rects them – with a high probability of
correct, or that was simply expected. In success. It is possible for errors to be cor-
computing, errors can be caused by some rected by an ERROR-DETECTING CODE, by a
failure in the hardware of the system or by request for the retransmission of the af-
some condition occurring in the software. fected data. Error-correcting codes are
Errors may arise, for example, as data is more complex and hence more costly to
being transferred to or from magnetic disk, implement than error-detecting codes but
i.e. as data is being written or read. They reduce the necessity for retransmission.
may also arise during the transfer of data
between two computers over a network. error-detecting code Any of a variety
Errors may be caused by a flaw in the stor- of CODES used to check data for certain
age medium, by a fault in a device, or by kinds of ERRORS that may have occurred,
NOISE in a communication channel. A bi- say, during transfer of the data or due to a
nary 1 may then be incorrectly written, writing or reading process. The informa-
read, or transmitted as a binary 0, or a bi- tion is encoded in such a way that a de-
nary 0 as a binary 1, producing an error coder can subsequently detect – with a high
in the data. The detection, handling, and probability of success – whether an error
possible correction of such errors is an im- has arisen. A fixed number of bits, k, is
portant aspect of computing. See also error- normally taken into the encoder at a time,
detecting code; error-correcting code. and then output as a codeword consisting
In programming, errors may arise be- of a greater number of bits, n. Each code-
cause the program fails to obey the SYNTAX word in a particular code is constructed
of the programming language, i.e. the pro- following a specific rule. The decoder will
gram breaks one of the rules defining how subsequently take in n bits at a time and
characters and sequences of characters can output k bits. On each codeword the de-
be combined. These are known as SYNTAX coder performs a CHECK for errors. If an
ERRORS and are normally detected when error is detected then generally the sending
the program is compiled. Other program- device is requested to retransmit the af-
ming errors may not become apparent fected set of bits. See also acknowledg-
until the program is executed. These are ment; error-correcting code.
called RUN-TIME (or execution) errors, and
may arise from, say, attempted division by error diagnostics Information that is
zero or from OVERFLOW. Logical errors are presented following the detection of some
another type, occurring because the logic condition causing an ERROR, for example
or design of the program is incorrect; a after an attempt to execute instructions in
JUMP to the wrong instruction is an exam- a program that are invalid or that operate
ple. Logical errors can arise if the problem on illegal data. The information is mainly
that the program is designed to solve is not to assist in identifying the cause of the
fully understood, so that the wrong ALGO- error. Error diagnostics can be produced,
RITHM is selected or written. Yet again, er- for instance, by the COMPILER, and the in-
rors may arise from the ROUNDING or formation would then generally concern
TRUNCATION of numbers, producing round- SYNTAX ERRORS. Other kinds of errors are
ing or truncation errors. detected at run time. In this case the error
diagnostics may be produced by a RUN-
error-correcting code Any of a variety TIME SYSTEM. See also debugging.
of CODES used to correct ERRORS that occur,
for example, during the transfer of data. error message A message that reports
The data is sent along a channel as a coded the occurrence of an ERROR. There may be
signal, with each element of the code con- some attempt to diagnose the cause of the
73
error rate
error. The message appears on a VDU sions are Excel 2003 (Windows) and Excel
screen or is stored in an error file. 2004 (Macintosh). Excel’s early success
was important in establishing Microsoft as
error rate A measure of the proportion an applications developer and it is now the
of ERRORS occurring in data transfers to or most widely used spreadsheet application.
from a magnetic disk or magnetic tape or Excel forms part of Microsoft OFFICE.
in data transmissions along a communica-
tion line. The error rate of, say, a magnetic exchangeable disk store (demountable
tape system is usually given as the number disk) An obsolete disk storage medium –
of bits or bytes of data transferred per either a DISK PACK or a DISK CARTRIDGE –
error, e.g. 1 error in 1014 bits (often ex- that could be removed from a DISK DRIVE
pressed as an error rate of 10–14). The error and replaced by another of the same type.
rate of a communication line may be ex- The store could be mounted on other disk
pressed as the number of bits transmitted drives of appropriate design.
per error, e.g. 1 error per 105 bits. Since er-
rors tend to come in bursts during trans- exclusive-OR gate (XOR gate) See
mission, the error rate is sometimes given nonequivalence gate.
as the percentage of error-free seconds.
execute To carry out a MACHINE IN-
escape key A keyboard key that when STRUCTION or a computer PROGRAM. The
depressed sends the escape character to the process is described as execution. A pro-
computer. This is often used in applica- gram executes from the time that it is
tions to negate the previously initiated ac- started until either it completes its task and
tion, to go back one level in the menu returns control to the OPERATING SYSTEM, it
structure, or to exit the program. encounters an error whose severity pre-
vents it from continuing, or it is aborted by
e-text (electronic text) A text-based the user. See also fetch-execute cycle; con-
work, such as a journal or book, that is trol unit.
available for reading on-line or that can be
downloaded and read off-line. execution See execute.
74
extranet
75
F
failure Termination of the normal oper- exercise or may involve the construction of
ation of some portion of a computer sys- an experimental or prototype version. It
tem, or of the whole system. A failure is the may concentrate on specific areas or deci-
result of a FAULT. The mean time between sions where the feasibility is questionable
failures (MTBF) is often used as a measure or the potential risk is greatest.
of the RELIABILITY of a computer system.
feed See Web feed.
FAQ /fak/ (frequently asked question) A
list of questions about the basic points of feedback The use of part of a system
some subject, together with answers. FAQs output as an input to the same system,
are often included in hardware and soft- sometimes intentional and at other times
ware documentation, and on the Internet unwanted. This term usually applies to
they are provided in specialized news- analog rather than digital computing. It is
groups to help new subscribers or to give used as a form of correction opposing a
general information on Web pages. change and is then called negative feed-
back. Positive feedback reinforces changes
FAT /eff-ay-tee/ See file allocation table. and can cause instability and other prob-
lems.
father file See master file.
Ferranti Mark I /fĕ-ran-tee/ See first
fault An accidental condition that pre- generation computers.
vents some portion of a computer system
from performing its prescribed function. It fetch To locate and load a MACHINE IN-
usually causes a FAILURE, although it is pos- STRUCTION or item of data from main store.
sible for a system to be FAULT-TOLERANT. See also fetch-execute cycle; control unit.
76
file
plastic buffer
coating
transmitter
optical
cladding
material
cross section of a single optical fiber total internal reflection in an optical fiber
Fiber optics
achieved by making the refractive index of tains a fixed or variable number of letters,
the optical fiber vary across its diameter: it digits, or other characters.
is lowest on the outside of the fiber and in- 2. A portion of a MACHINE INSTRUCTION. It
creases either smoothly or in one or more contains a coded representation of, say, the
steps towards the center. (The refractive operation to be performed or the address
index of a medium is a measure of the ex- of the operand or operands.
tent to which light bends on entering that
medium.) The light is thus made to snake FIFO /fÿ-foh/ (first in first out) See
or to zigzag down the center of the fiber. queue.
A simple fiber optics transmission sys-
tem consists of a transmitter with a light fifth generation computers The types
source (such as a laser), a receiver with a of computer currently under development.
light detector, and an intervening optical The features are still conjectural but point
fiber. An electrical signal carrying the data toward the ‘intelligent’ machine. See also
to be transmitted is fed to the transmitter first, second, third, fourth generation com-
puters.
and causes the light emitted by the source
to be modulated, i.e. encoded with the data
fifth generation language A type of
in some way (see modulation); normally
programming language that expresses a
the light is emitted as a sequence of pulses.
problem to be solved in terms of the logical
The light signal enters and travels along the relationships between and the constraints
optical fiber, and at the other end is de- on the elements of the problem. Such lan-
tected and converted back into the original guages remove entirely the need for the
electrical signal. user to specify an ALGORITHM to solve the
In fiber optics transmission a sheaf of problem, and so carry further the abstrac-
optical fibers is used so that many signals tion already visible in FOURTH GENERATION
can be carried at the same time. Optical LANGUAGES.
fibers have become cost-competitive with
COAXIAL CABLE, with the advantages of file A named collection of information
having a much greater data-carrying ca- held temporarily or permanently on mag-
pacity, much lower signal loss, immunity netic disk, CD, DVD, tape, etc. The file-
to electrical interference, and no signal in- name is stored in a DIRECTORY and is used
teraction between the fibers. to identify the file to the OPERATING SYSTEM
of the computer. The details of how a file is
fiche /feesh/ (microfiche) See COM. stored are handled by the FILE SYSTEM. Files
may hold data, programs, text, or any
field 1. A portion of a RECORD. It con- other information, and can therefore be de-
77
file allocation table
scribed as DATA FILES, PROGRAM FILES, etc. file attribute A property of a file, such
They can also be broadly classified as MAS- as whether the file is hidden, read only, or
TER FILES, TRANSACTION FILES, or REFERENCE archived.
FILES. Some files are WORK FILES and have
only a very brief existence. file extension See extension.
A file is sequentially accessed if the
bytes or RECORDS are read or written one file format A method of organizing data
after another starting at the beginning. A in a FILE. There are many file formats, some
file organized to support sequential access of which (e.g. TEXT FILES) are understood
is known as a sequential file. Sequential or- by many programs. However, sophisti-
ganization is used for any files for which cated applications tend to use their own
there is no requirement to access a particu- formats. A file may or may not contain an
lar byte or record directly, but that are nor- indication of its format.
mally accessed by starting at the first byte
or record and moving through the file one file maintenance The maintenance of
byte or record at a time. This is the case for both the integrity of FILES, i.e. the correct-
many data files. ness of all the data values, and an efficient
A file is randomly accessed if any byte internal organization of files. It is per-
or record can be accessed directly without formed by software. See also file manage-
having to pass over previous bytes or ment.
records in the file. A file organized to sup-
port random access is known as a random- file management The overall manage-
access file (or direct-access file). In such ment of FILES, including their allocation to
cases a program needs a method of deter- space on BACKING STORE, control over file
mining which byte or record to access to ACCESS, the writing of BACKUP copies, the
find the data it needs. This information movement of files to the FILE ARCHIVE, and
must be already known to or computable the maintenance of DIRECTORIES. It is per-
by the program or held within the file itself. formed by software, basic file management
Three common random access methods are being done by the OPERATING SYSTEM of the
indexed (see indexed file), indexed sequen- computer. See also file maintenance.
tial (see indexed sequential file), and HASH-
ING. filename All files have names but differ-
ent operating systems impose different
file allocation table (FAT) The area on rules on how names are constructed. Most
a hard drive that uses the FAT FILE SYSTEM operating systems ban the use of certain
or its derivatives where details about the characters, for example a colon, and im-
location of files are kept. These include pose a limit on the length. Also many sys-
which CLUSTERS hold each file’s data, the tems including Windows and UNIX allow
order of clusters within the files, what free filenames to have an EXTENSION that con-
space is available, and where any flawed sists of one or more characters following
segments occur. The FAT file system was the proper name.
used by MS-DOS and is still supported by
Windows. file protection The protection of FILES
from mistaken or unauthorized ACCESS of
file archive A collection of magnetic information or, in the case of PROGRAM
disks, CDs, DVDs, or tapes on which are FILES, from mistaken or unauthorized exe-
held rarely used FILES, which can be re- cution. File protection may be concerned
stored to the computer system if required. with the security of the contents of files, or
Files may be explicitly placed in the with the protection of the media on which
archive, or may be automatically moved files are held. The former is implemented
there after they have remained unused for by software, the latter by operating proce-
a given period. dures.
78
finger
file server A computer on a network the receiver must agree to the transfer and
that acts as a storage device, on which any inform the sender of its eventual success or
user can store files. File servers are often failure. In MULTIPROCESSING SYSTEMS file
dedicated, i.e. they perform no other func- transfer is normally achieved by BACK-
tion. GROUND PROCESSING. See also file transfer
protocol.
file sharing The sharing of public or pri-
vate computer data or space in a network file transfer protocol (FTP) A protocol
with various levels of access privilege. File that allows a user on one host to access
sharing can happen on a small LAN and (read and possibly write to), and transfer
has been a feature of large mainframe com- files to and from, another host over the In-
puters for many years. It allows users to ternet. It is the simplest way to exchange
read, write, modify, erase, copy, and print files on the Internet and can be used with a
files according to their level of access priv- simple command line interface from MS-
ilege. DOS, with a Windows GUI program, or
through most Web browsers. A password
filestore /fÿl-stor, -stohr/ The portion of may be needed to gain access to some files.
disk BACKING STORE used for storing per- See also anonymous FTP; HTTP; SMTP.
manent FILES.
file updating Inserting, deleting, or
file system 1. The method by which the amending values in a file – usually a DATA
storage of FILES on magnetic disk, CD, FILE – without changing the organization of
DVD, etc., is organized. Files must be the contents. File updating can be achieved
stored in ways that make efficient use of in two ways. In one, commonly used in
the available space (e.g. files might not be DATA PROCESSING, the updating process is
physically contiguous), that provide quick carried out automatically by program. In
access times (e.g. filenames might be in- the other, a file is displayed on an INTERAC-
dexed), and that take account of the char- TIVE device and the operator can then see to
acteristics of a particular storage device amend it, using an appropriate editing pro-
(e.g. CDs work differently to magnetic gram.
disks). There are several ways this can be
done and each operating system supports filter A section of code or a program
one or more. For example, MS-DOS used the that takes input data and transforms it in
FAT (file allocation table) file system, some way. The output might be written to
whereas Windows XP uses NTFS (NT file a file or passed to another program or sec-
system) or FAT for fixed disks, FAT only tion of code.
for floppy disks, and CDFS (CD file sys-
tem) for CDs. find See search.
2. The part of an operating system respon-
sible for maintaining a particular file sys- Finder In Apple Macintosh computers,
tem and hiding its organizational details the desktop and file management system. It
from users, so that they can treat all files is also responsible for looking after the
simply as named collections of informa- clipboard and scrapbook and all desktop
tion. icons and windows.
file transfer The movement of entire finger A piece of software that displays
FILES from one computer system to an- information about a particular user, or in-
other, typically across a NETWORK. In order deed all users, logged on to a local or re-
to transfer a file successfully there must be mote system. It typically shows the full
cooperating programs running on each sys- name, last name, last login time, idle time,
tem: the sender has to be able to specify the terminal line, and terminal location (if ap-
filename, ultimate destination, and autho- propriate). The most common use is to see
rization for the transfer to the receiver, and if a person has an account at a particular
79
Firefox
Internet site. Many sites do not allow in- tating MAGNETIC DRUMS. Input/output was
coming finger requests. achieved mainly with PUNCHED CARDS,
punched PAPER TAPE, and later on with
Firefox Trademark An OPEN SOURCE MAGNETIC TAPE; simple PRINTERS were
WEB BROWSER produced by the Mozilla available. The use of valves meant that first
Foundation. First released in 2004 and fa- generation computers were very large in
vored by technical users, it quickly became size, generated considerable heat and thus
the second most widely used Web browser required cooling, and did not operate very
after INTERNET EXPLORER. fast or very reliably. They were still able to
perform impressive computations. Most of
firewall Software or a combination of them embodied the concept of the STORED
hardware and software that monitors and PROGRAM.
controls the data that passes across a net- The following are important first gener-
work’s connection to another network. It ation computers.
is often used to protect a private network ENIAC: design begun 1943 by J. W.
or intranet that also has connections to the Mauchly and J. P. Eckert at the University
Internet from unauthorized access by users of Pennsylvania; operational 1946, but did
of other networks. The firewall is often in- not use a stored program and was regarded
stalled in a dedicated computer separate as a general-purpose electronic calculator.
from the rest of the network so that in- EDVAC: design begun 1945 by J. von
coming users cannot directly access private Neumann at the University of Pennsylva-
network resources. Software-only firewalls nia, the proposal being the first written
can be installed on standalone PCs, and are documentation of a computer with a
now considered an essential safety precau- stored program; not operational until
tion when using the Internet. 1952.
Manchester Mark I: design begun 1946 by
FireWire (IEEE 1394) Trademark A T. Kilburn and F. C. Williams at the Uni-
standard INTERFACE used to connect a com- versity of Manchester; regarded as the first
puter to up to 63 external devices. Devel- operational stored program computer,
oped by Apple Computer, Inc., it became with realistic problem-solving first
an IEEE standard in 1995. It supports PLUG achieved in April 1949.
AND PLAY and offers high data-transfer EDSAC: design begun 1946 by M. V.
speeds, either 400 Mbps or 800 Mbps de- Wilkes at Cambridge University; started
pending on the version. FireWire has long operating in May 1949 as the first com-
been the de facto standard method of plete operational stored program com-
transferring data from digital camcorders puter.
to computers. All Macintosh and an in- Ferranti Mark I: commercial version of the
creasing number of Windows computers Manchester Mark I, produced by Ferranti
support FireWire. Ltd., Manchester, and delivered in 1951; it
was the world’s first commercially avail-
firmware /ferm-wair/ Programs that do able computer. Ferranti’s computer inter-
not have to be loaded into a computer but ests were sold, 1963, to ICT, which
are permanently available in main store. merged, 1968, with English Electric Leo
They are held in ROM (read-only memory) Marconi to form ICL (International Com-
and thus are fixed in content even when the puters Ltd.).
power supply is removed. UNIVAC I: product of the US company
Eckert-Mauchly Corp. and commercially
first generation computers The earli- available in 1951, shortly after the Ferranti
est computers to be produced, in the 1940s Mark I. Eckert-Mauchly was formed in
and early 1950s, based on the technology 1947 following development of ENIAC
of the time. Electronic valves were used in and later became Sperry-Univac.
the circuitry and storage devices were usu- LEO I: design begun 1947 by T. R.
ally delay lines, electrostatic devices, or ro- Thompson and J. Pinkerton for J. Lyons &
80
flash memory
Co. (a large UK catering firm) and opera- FLOATING-POINT NOTATION are unaccept-
tional by 1953; enhanced version (LEO II) able (e.g. in monetary calculations). How-
marketed by Leo Computers Ltd., formed ever, the values involved must be neither
1954. Leo Computers merged with English very large nor very small. The range of val-
Electric, 1963, and subsequently formed ues that a computer can handle efficiently
part of ICL. is limited by WORD length: each item of in-
See also second, third, fourth, fifth gen- formation must fit into a word containing
eration computers. a fixed number of bits. A 32-bit word, for
example, can have 232 possible values,
fixed disk, fixed disk drive See disk which can be used, say, to represent whole
drive. unsigned numbers from 0 to
4 294 967 295, or 0 to 429 496.7295 if an
accuracy of one ten-thousandth is re-
quired. Although more than one word can
1 2 3 4 5 6 7 8 9 10 be used to represent a number and hence
give a wider range of values, this increases
J O N E S 1 0 3 processing time. Where a wide range is
S M I T H 8 7 needed, floating-point notation must be
used.
B R O W N 1 2 3
flag A device, item of data, etc., that can
be set to a particular value to indicate that
some condition or state has been attained
Fixed format by a piece of equipment or by a program.
For example, an overflow flag is a single bit
fixed format Data arranged so that par- that is set to 1 when OVERFLOW occurs in a
ticular items always occur in particular po- register during an arithmetic operation; an
sitions. In the example shown in the interrupt flag is set when an INTERRUPT oc-
diagram, the name always starts in the first curs.
position, while the numbers are in posi- The word sentinel is used as a synonym
tions 8, 9, and 10. Because the where- for flag but is usually used in the context of
abouts of each item of information is input and output. For example, an end-of-
strictly defined, there is no need for gaps data sentinel indicates that all the data has
between the items. The format of the ex- been read. A flag or sentinel, once set, is
ample could be ‘A name in columns 1 to 5, subsequently used as a basis for a condi-
an age in positions 8 and 9, and the num- tional JUMP and similar decision processes.
ber in the family in position 10.’ Compare
free format. flame A strong opinion, criticism, or
derogatory comment about something or
fixed-point notation A representation someone in an e-mail, chat, or usergroup
of REAL NUMBERS that is in everyday use posting.
and in which a number is expressed in the
form of a set of digits with the RADIX POINT flash drive See flash memory.
fixed in position so that the digits have
their correct positional value (see number flash memory (flash RAM) A non-
system). Examples of fixed-point notation volatile memory that can be erased and re-
in decimal and binary are programmed in units called memory
4927.54, 101011.11, 0.0000001 blocks. It is a variation of EEPROM, which
Fixed-point notation is used in computing is erased and rewritten at byte level and is
in situations, such as commercial data pro- thus slower than flash memory updating.
cessing, where huge quantities of numbers Flash memory can be used to hold the
are handled, making speed important, or BIOS in PCs but is not suitable for RAM.
where the rounding errors possible with A popular use is in the flash drive, a tiny
81
flash RAM
device that acts as a disk drive holding nary 1 and binary 0. Flip-flops can there-
many megabytes of data. Flash drives usu- fore store one BIT of information, and are
ally connect to a computer through a USB widely used as storage elements, e.g. in
port and can thus be easily unplugged. ROM and RAM and in REGISTERS. Various
Flash memory is also used in digital cellu- types have been developed, differing
lar phones, digital cameras, cards for note- slightly in function.
book computers, and many other devices.
floating-point notation A representa-
flash RAM See flash memory. tion of REAL NUMBERS that allows both very
large and very small numbers to be ex-
flatbed plotter See plotter. pressed in a convenient form. Examples are
1.3 × 106, –2.5 × 103, 0.65 × 10–3
flatbed scanner A type of scanner in Such numbers are encountered in science,
which the page to be scanned is stationary math, and engineering and cannot be han-
behind a glass window while the head dled efficiently in a computer in FIXED-
moves past the page, similar to a photo- POINT NOTATION.
copier. In the flatbed scanner a series of A floating-point number is represented
mirrors keeps the image picked up by the by a pair of quantities called the mantissa
moving head directed onto a lens that feeds and the exponent, and has the general form
the image to a bank of sensors. The advan- m × be
tage of flatbed scanners is that they are rel- b is the base of the number system, equal to
atively inexpensive and thick documents, 10 in decimal notation; the mantissa, m,
such as books, can be scanned. can be positive or negative; the exponent,
e, is the power to which the base is raised,
flat panel display See display. and is a positive or negative integer. In pro-
gramming languages, whose character sets
flat screen A type of SCREEN in the form
do not include superscripts, different con-
of a thin flat panel rather than the protrud-
ventions are used. A common one is to use
ing surface usually associated with the
an ‘E’ to stand for ‘10 to the power of’, e.g.
cathode-ray tube used in many display de-
2.7 × 10–3
vices. Plasma panel DISPLAYS, ACTIVE MA-
TRIX DISPLAYS, and LCDS have flat screens.
is written
2.7 E –3
flexidisk /fleks-ă-disk/ See floppy disk. Inside a computer a floating-point num-
ber is stored and manipulated as a se-
flicker The phenomenon whereby a dis- quence of bits. Typically, the first bit is a
play screen appears to change brightness. sign bit, indicating whether the mantissa is
Flicker can be the result of several factors positive or negative; this is followed by a
including a low refresh rate, ambient light- fixed number of bits giving the sign and
ing, or low-persistence phosphors on the magnitude of the exponent; this in turn is
screen. It is somewhat subjective and most followed by a fixed number of bits repre-
people perceive no screen flicker if the re- senting the magnitude of the mantissa. The
fresh rate is 72 Hz or higher. radix point of the mantissa is usually taken
to be before the leftmost bit, i.e. the man-
flip-flop (bistable) An electronic circuit tissa is a fraction. There may be, say, 7 bits
that has two stable states, either of which for the exponent and 40 for the mantissa.
can be maintained until the circuit is made This gives a wide range of numbers but
to switch, or ‘flip’, to the other state. A means that on small machines each float-
change in output state requires a suitable ing-point number needs several WORDS of
combination of input signal(s) and existing store, which degrades performance. Also,
output state; in effect, the receipt of a pulse floating-point numbers cannot represent
will reverse the state. The two states are all possible real numbers within their range
used in computer circuits to represent bi- exactly; this leads to rounding errors.
82
flowchart
metal shutter
hub
disk
read-only
window
floating-point number A number ex- protect hole in one corner of the en-
pressed in FLOATING-POINT NOTATION. velope.
floppy disk (diskette; flexidisk) A rela- flops (floating-point operations per sec-
tively small flexible MAGNETIC DISK, en- ond) A commonly used measure of com-
closed within a stiff protective envelope. puter performance, made in terms of the
Floppy disks were formerly used in small number of arithmetic operations that can
computers as on-line BACKING STORE and be performed on FLOATING-POINT NUMBERS
off-line storage devices. They have now in a period of one second. It is usually
largely been superseded by writeable CDs, expressed in megaflops (a million flops),
but are still used for storing or transferring gigaflops (a billion flops), teraflops (a tril-
small quantities of data. They have much lion flops), or petaflops (a thousand trillion
smaller storage CAPACITIES and much flops).
longer ACCESS TIMES than HARD DISKS, but
are considerably cheaper and need less floptical disk /flop-tă-kăl/ A type of
careful handling. magnetic disk the same size as a 3½″ floppy
The most common type of floppy disk that used a laser beam to help position the
is 3½ inches in diameter. The older 5¼ read/write heads more accurately than was
inch disks are now rarely seen. As with possible with an ordinary floppy. There
other magnetic disks, data is recorded in was a series of small, precise concentric
concentric tracks in the magnetic coating. tracks stamped in its surface coating. These
The tracks themselves are divided into SEC- tracks were where the data was written and
TORS. Most floppy disks have a capacity of were thinner and more numerous than on
about 1.4 megabytes. a conventionally formatted floppy. The
The floppy disk remains inside a rigid combination of more precision and more
plastic envelope when it is fed into a tracks allowed a floptical to hold up to 20
DISK DRIVE for reading and writing pur- megabytes of data. Flopticals are now ob-
poses. The disk rotates inside the en- solete.
velope, becoming rigid as it spins. A sliding
metal shutter covers a slot in the en- flowchart 1. (program flowchart) A di-
velope, which allows the read/write head agrammatic representation of the structure
to make contact with the disk surface. of a program, showing the actions per-
The disk drive can be prevented from formed by the program and the flow of
writing to the disk by means of the write- control. It consists of a set of boxes of var-
83
flowchart
symbol use
description
for any input or output of data, however achieved
medium
content for data held in any on-line file, the medium usually
medium being disk
program
for processing of data
function
84
form
computer system, showing both the opera- font A set of printable or displayable
tions involved and the flow of data in the characters. The design for a set of fonts is
system. It does not indicate how the pro- called the typeface and the variations of
cessing operations are carried out – this in- this design (bold, italic, etc.) form the type-
formation would appear in a program face family. Many fonts have been de-
flowchart. The number of flow lines enter- signed and can be purchased for general
ing and exiting each symbol depends on the use. Three main kinds of font are in use on
situation. computers. TRUETYPE fonts are the com-
monest on personal computers. The other
flow control Procedures used to control kind are POSTSCRIPT fonts (also known as
the DATA TRANSFER rate, i.e. to limit the rate Type 1 fonts), which are preferred for out-
at which data is transferred to the rate at put of PostScript printing files used in com-
which it can be received – either at its final mercial printing. OpenType fonts are a
destination or at a series of points along its development of TrueType fonts that can
route. incorporate PostScript font data.
There are three flow-control methods in
common use. With XON/XOFF the re- footprint The area of floor or desk
ceiver can temporarily halt the sender by space occupied by a device, or the shape
transmitting an XOFF (transmitter off, and size of panel opening into which it
ASCII code 19), and an XON (transmitter
would fit.
on, ASCII code 17) when once more ready
to receive; this method requires full DUPLEX FOR See Basic.
TRANSMISSION. With ETX/ACK the sender
terminates a block of data with an ETX
foreground processing In a MULTI-
TASKING system, the processing of a com-
(end transmission, ASCII code 3), and the
puter program in a way that allows
receiver responds with an ACK (acknowl-
interaction with the user through the key-
edge, ASCII code 6) only when it is ready
board. Only one program per user can run
for more data; this method can be used in
in this way at any one time, and these are
HALF-DUPLEX communications but relies on
called foreground programs; all other pro-
the sender not sending too much at a time. grams are background programs. Fore-
Thirdly there is the use of control signals, ground programs are usually granted a
where the receiver uses a separate channel high priority and can thus preempt the re-
to indicate to the sender whether it is ready sources – main store, processor, etc. – of
for data. the computer from background programs,
which have a lower priority. Users can usu-
focus The state of the active item in a ally change the foreground program at will
GUI FORM. Only one item per form can by selecting one of the background pro-
have the focus at any one time and it is this grams, which immediately becomes the
item that receives input from the keyboard foreground program.
when the form is active. The focus can be
moved by clicking on a new item with the for loop See loop.
mouse or by pressing the tab key (move to
next item) or shift+tab (move to previous form 1. A page of printer paper. It may
item) on the keyboard. be a single sheet or may be sheets joined to-
gether in the form of CONTINUOUS STA-
folder 1. An alternative term for DIREC- TIONERY. In either case it may be multipart
TORY on GUI systems. so that copies can be produced. Individual
2. A component of a storage facility found sheets may be preprinted with headings or
in some programs. For example, an e-mail other information, possibly with lines or
client will often provide a system of user- boxes.
configurable folders to store received and 2. A display that contains blank fields,
sent messages. which can be filled in using a keyboard and
85
format
mouse. The data can then be transferred the earliest HIGH-LEVEL LANGUAGES gener-
automatically to a database. ally available for scientific and technical
3. A WINDOW whose content is so laid out applications, and has been under continu-
as to display a particular type of informa- ous development. Fortran is still widely
tion to the user, and possibly to receive used for scientific work, even though its
input. DIALOG BOXES are a type of form. CONTROL STRUCTURES are relatively crude
compared with more modern languages.
format 1. The structure that is used in Each new release has more sophisticated
the arrangement of data. It is determined features.
before use, and in some cases is set by in-
dustrial or international standards. There forum An on-line discussion group.
are various kinds. Instruction format de- Some Web sites, especially Web PORTALS,
fines the arrangement of the parts of a MA- provide a large selection of forums in
CHINE INSTRUCTION. Printer format defines which participants can exchange open
the layout on paper of the output from a messages. Newsgroups and conferences are
PRINTER, for example the areas of a page also types of forums.
where printing will occur and the spacing
between words and between lines. Tape forward compatibility See upward
format and disk format define the arrange- compatibility.
ments in which data can be recorded on
MAGNETIC TAPE or MAGNETIC DISK. fourth generation computers Com-
2. To prepare a blank storage medium, puters currently in use and designed in the
such as a disk, to accept data. period from about 1970. Since the design
3. To put data into a predetermined struc- of computers is a continuous process by
ture. different groups in several countries, it is
4. To organize or change the appearance difficult to establish when a generation of
and arrangement of text and images on a computers starts and finishes. Conceptu-
screen or printed page. ally the most important criterion that can
be used to separate fourth generation from
form feed A special character (ASCII THIRD GENERATION COMPUTERS is that they
decimal 12), sometimes abbreviated to FF, have been designed to work efficiently with
that by convention causes a printer to ad- the current generation of HIGH-LEVEL LAN-
vance one page length or to the top of the GUAGES and are intended to be easier to
next page. program by their end-user. From a hard-
ware point of view they are characterized
forms tractor See tractor feed. by being constructed largely from INTE-
GRATED CIRCUITS and have very large
Forth /forth/ A PROGRAMMING LANGUAGE (multi-megabyte) main store fabricated
designed originally for the control of scien- from SEMICONDUCTOR MEMORY.
tific instruments from microcomputers The availability of an ever-increasing
with very small memories. It is a LOW-LEVEL range of integrated circuits has produced a
LANGUAGE that makes great use of the rapid fall in the cost of hardware while
STACK, and is very compact. Although the software costs have continuously in-
current memory sizes of microcomputers creased. This has led to the existence of
have made the small size of Forth largely ir- many computers designed for special tasks
relevant, it is still used as a substitute for such as communications, automatic con-
ASSEMBLER language among some micro- trol, and military systems, which in the
computer users. past would have been general-purpose ma-
chines adapted to their task by software. It
Fortran /for-tran/ A PROGRAMMING LAN- has also led to a dramatic increase in home
GUAGE whose name is derived from for- computing and Internet use.
mula translation. It was first released by See also first, second, fifth generation
IBM in the mid-1950s. Fortran was one of computers.
86
fractal
87
fragmentation
the information that could produce the The format here could be ‘A name fol-
whole figure. In this sense, fractals are said lowed by an age followed by the number in
to be ‘self-similar’. The study of fractals the family.’ Compare fixed format.
has applications in chaos theory and in cer-
tain scientific fields (e.g. the growth of free space 1. Space not being used on a
crystals). It is also important in computer storage device.
graphics, both as a method of generating 2. Space provided free of charge for a user
striking abstract images and, because of to set up a Web page.
the self-similarity property, as a method of
compressing large graphics files. freeware /free-wair/ Copyrighted soft-
See also Mandelbrot set. ware that is offered at no cost but may
have limitations on its use, most commonly
fragmentation /frag-men-tay-shŏn/ The that it cannot be incorporated in any pro-
breaking up of disk FILES into a number of gram or package that is sold for profit and
different sections scattered across a disk. It that the author’s copyright must always be
occurs when files of different sizes are fre- acknowledged. See shareware.
quently deleted and written. Each new file
is written in the first available gap left by frequency division multiplexing (FDM)
the deletion of a previous file; if the new See multiplexer.
file is too big for the gap then the rest is
written in the next gap, and so on. The re- friction feed A technique for feeding
sult is that a number of movements of the paper into a PRINTER. Each sheet is fed in by
read/write heads of the DISK DRIVE might be hand and is gripped by a rotating roller, as
required to access each fragmented file, in a typewriter. The paper cannot be posi-
thus reducing the efficiency of the file sys- tioned as accurately as it is with TRACTOR
tem. Programs exist for defragmenting FEED. See also cut sheet feed.
disks.
front end The part of an application
frame 1. In data transmission, a unit of that interacts with the user, contrasted
data complete with addressing and the nec- with the ENGINE, which manages and con-
essary protocol control information that is trols the application’s data. In client/server
to be transmitted between network points. applications the front end usually runs on
The transmission is usually serial and con- the client while the engine runs on the
tains header and trailer fields that ‘frame’ server.
the data.
2. A separate section of a Web page that front-end processor A small computer
acts as an independent unit (e.g. text can be that is used to relieve a mainframe com-
independently scrolled). Frames can be de- puter of some of the tasks associated with
fined by tags in HTML. input/output. It receives data from a num-
ber of input/output devices (such as the ter-
free format Data arranged so that, al- minals in a MULTIACCESS SYSTEM), organizes
though the items are in a specified order, it, and transmits it to the more powerful
the actual positions of the items are not de- computer for processing. It also handles
fined. In free format it is necessary to have the output from the mainframe to the ter-
a separator such as some spaces and/or a minals. In a NETWORK, a front-end proces-
comma between the items so that it is pos- sor may handle the control of
sible to tell when one ends and the next be- communication lines, code conversion,
gins, for example error control, etc. It is more powerful than
JONES,10 3 a MULTIPLEXER in that it can perform simple
SMITH,8 7 editing, echoing, routing, etc.
BROWN,12 3
ROBINSON,25, 1 FTP See file transfer protocol.
88
FYI
be called. telligence.
A collection of different functions are
provided as part of a programming lan- FYI (for your information) Publications
guage, and any of these can be used by the that convey general information about top-
programmer without an understanding of ics related to the Internet. They are a sub-
the details of its internal working. Most series of RFCs that are not technical
languages also permit user-defined func- standards or descriptions of protocols.
89
G
G The symbol used for GIGA-, or some- way of classifying computer systems on the
times for gigabyte. basis of advances in technology, especially
electronic technology, and latterly also on
game port The port on a PC into which advances in software. The generation of
devices such as joysticks or games con- computers currently in use is the FOURTH
trollers are connected. GENERATION (which succeeded the THIRD,
SECOND, and FIRST GENERATIONS), with
games console An interface designed FIFTH GENERATION COMPUTERS currently
especially for computer games. Movement under development. Since the design of
and speed are manipulated using buttons computers has been a continuous process
or levers on the console. Some controllers for decades – by a wide variety of people in
are specialized, such as those with steering different countries, faced with different
wheels and accelerator and brake pedals problems – it is difficult (and not very prof-
for racing games. Force feedback joysticks itable) to establish when generations start
are now available to add realism. and finish.
garbage Stored data that is no longer generator A program that accepts the
valid or no longer wanted. Removal of this definition of an operation that is to be ac-
superfluous data from store is known as complished and automatically constructs a
garbage collection. See also GIGO. program for the purpose. An example is
the report generator, which can be used in
garbage in garbage out See GIGO. commercial data processing to extract in-
formation from files. The input to such a
gas panel display See display. program is a description of the file struc-
ture and a specification of the information
gate See logic gate. required and the way in which it should be
presented to the user. The report generator
gateway A network connection device can then construct a program to read the
similar to a BRIDGE except that it is usually file, extract the desired information, and
visible to the user. It translates data be- output it in the desired format.
tween one type of network and another.
GHz (gigahertz) A measure of frequency
gender bender A component for joining equal to 109 hertz.
two male or two female connectors to-
gether. GIF /jiff, giff/ (graphics interchange for-
mat) A graphics file format developed by
general-purpose computer A com- CompuServe, which is widely used for
puter that can be used for any function for transmitting raster image files on the Inter-
which it can be conveniently programmed net. The files are encoded in binary and use
and can thus perform a series of unrelated LZW compression. GIF files are often
tasks. smaller than the same file would be if
stored in JPEG format but are limited to 256
generations of computers An informal different colors and so photographic im-
90
GOTO statement
ages are not stored so well. GIF files can ENGINE operated by Google, Inc. First de-
contain simple animations. These files have veloped in 1996–98 by Larry Page and
the file extension gif. See also PNG. Sergey Brin while they were graduate stu-
dents at Stanford University, it is based on
giga- /gig-ă-/ A prefix indicating a multiple their theory that the relative importance of
of a thousand million Web pages containing a match of the
(i.e. 109) search term can best be determined by an-
or, loosely, a multiple of 230 alyzing the number of links to each page
(i.e. 1 073 741 824). from the other matching pages; the results
In science and technology decimal notation are then presented in descending order of
is usually used, and powers of 10 are thus this importance. This theory was con-
encountered – 103 106, 109, etc. The sym- firmed in practice and Google acquired a
bol G is used for giga-, as in GV for giga- reputation for quickly and reliably return-
volt. Binary notation is generally used in ing relevant results – a notorious problem
computing, and so the power of 2 nearest with other search engines at the time. By
to 109 has assumed the meaning of giga-. early 2004 the Google engine handled over
The prefix is most frequently encoun- 80% of all Web searches. The verb ‘to
tered in computing in the context of stor- Google’, meaning to perform a Web search
age CAPACITY. With magnetic disks, mag- using Google, has entered the English lan-
netic tape, and main store, the capacity is guage.
normally reckoned in terms of the number
of BYTES that can be stored, and the giga- Gopher A distributed information ser-
byte is used to mean 230 bytes; gigabyte is vice (named after the mascot of the Univer-
usually abbreviated to G byte, GB, or just sity of Minnesota) that makes hierarchical
to G; thus a 1.2 G byte disk can hold collections of information available across
1.2885 × 109 bytes. Giga- is part of a se- the Internet. Each node is either a menu, a
quence search node, or a leaf node. A simple pro-
kilo-, mega-, giga-, tera-, peta-, … tocol is used that allows a single Gopher
of increasing powers of 103 (or of 210). client access to information from any ac-
cessible Gopher service. However, its use
GIGO /gÿ-goh/ Garbage in garbage out, declined with the spread of the more flexi-
signifying that a program working on in- ble World Wide Web.
correct data produces incorrect results.
GOSUB /goh-sub/ See Basic.
global Applicable to a whole document,
file, or program rather than a particular GOTO statement /goh-too/ An uncon-
section. See local. ditional JUMP INSTRUCTION in a high-level
language. It causes the normal flow of con-
global variable See scope. trol to be broken by specifying explicitly
the next statement to be executed, usually
golfball printer An obsolete type of im- by means of a statement label, as in
pact character PRINTER whose printing goto error_handler
mechanism was first used in electric type- Excessive use of GOTOs can make pro-
writers. The characters were moulded on a grams difficult to follow and their use is
spherical print head. The head was rotated therefore discouraged. An extended form,
(in two directions) in order to bring the re- the computed GOTO, appears in Fortran
quired character opposite the printing po- and is used to go to one of several state-
sition, and was then struck against the ments, the choice dependent on the value
inked ribbon and paper. The head was in- of an integer variable. Many versions of
terchangeable, enabling different character Basic have a similar facility; however, the
sets to be used. same effect can be achieved more elegantly
by the CASE or SWITCH STATEMENTS of other
Google Trademark An Internet SEARCH languages.
91
GPU
GPU See graphics processing unit. graphics processing unit (GPU; graphics
coprocessor) A CHIP or CHIPSET that per-
grandfather file See master file. forms data manipulations related to graph-
ical output on a VDU. Such operations
graph A network of lines and vertices, could be accomplished in the computer’s
or a data structure that emulates this. A CPU by appropriate software, but using spe-
TREE is a kind of graph. cialized hardware is much faster and mod-
ern uses of computer graphics – in
graphical user interface (GUI) The particular 3-D animations – require this
type of user interface provided with Win- extra speed. A GPU is the core component
dows and Macintosh operating systems of a GRAPHICS ACCELERATOR.
(and also others), which provides the user
with an intuitive graphical menu-driven graphics tablet See digitizing pad.
system incorporating such things as win-
dows, icons, pull-down menus, scroll bars, graph plotter See plotter.
and pointing devices.
gray scale The representation of colors
graphic character A symbol that can be in a graphics file or on a monochrome out-
produced by printing, typing, or handwrit- put by shades of gray ranging from black
ing. It may be an ALPHANUMERIC CHARAC- to white. The number of bits per pixel de-
TER, a SPACE CHARACTER, a SPECIAL
termines the number of shades of gray that
CHARACTER, or any other symbol. Compare
can be achieved.
control character.
greeking The substitution of meaning-
graphics See computer graphics.
less line patterns or bars of gray to repre-
sent characters or images whose size makes
graphics accelerator A chipset at-
them too small to be displayed at the cur-
tached to a video board that contains a
rent screen resolution. The spacing and
GRAPHICS PROCESSING UNIT. This enables it
arrangement of the text is visible but the
to update the video display much more
quickly than the CPU can and also com- text itself is not legible.
pute special effects common to 2-D and 3-
D images. Another benefit is that the CPU groupware Applications designed to be
is freed up to execute other commands useful to a team of people. Groupware ser-
while the graphics accelerator is handling vices cover such activities as the sharing of
graphics information. Graphics accelera- calendars, e-mail handling, shared infor-
tion has become a necessity since the ad- mation access, collective writing, and elec-
vent of multimedia and sophisticated tronic meetings. Examples of groupware
games, in which very large images need to applications are Microsoft Exchange and
change quickly in response to user input. Lotus Notes.
The older graphics accelerators tend to run
using the PCI bus but more power can be guest A log-in account on a network
obtained with accelerators designed for an that can be accessed without a password.
accelerated graphics port (see AGP). For example, on some subscription Web
sites it is possible to sample the content by
graphics card (graphics board) See ex- entering as a guest. The ACCESS PRIVILEGES
pansion card. of a guest account are generally very lim-
ited.
graphics display A VDU with graphics
capability, i.e. one that can display pic- GUI /goo-ee/ See graphical user inter-
tures, graphs, and charts as well as text. face.
92
H
93
handle
amounts of text, but unsuitable for repro- vances have greatly increased the recording
ducing high-quality images. DENSITY of hard disks and storage capaci-
ties of many gigabytes on a 3.5 inch disk
handle 1. An outline box that appears in are now common.
many applications when a graphical object
is selected. It can be used to resize the hard return The character that forces
image. The same mechanism can be used to the following text to be printed on a new
alter the boundaries of the area selected line in a document. In word-processing
without changing its contents. documents a hard return indicates the end
2. A name chosen to identify a user on an of a paragraph. See also soft return.
on-line service. It may be the user’s real
name, or a nickname. hard-sectored disk See sector.
94
heuristic
hashing /hash-ing/ A technique used for with the surface of a hard disk as it rotates
organizing TABLES or FILES of information in a DISK DRIVE. (Normally the head flies
to permit rapid SEARCHING or TABLE LOOK- just above the surface.) A crash is often
UP. It is particularly useful for random-ac- caused by the head passing over a particle
cess FILES and for tables to which items are of dust on the surface. The contact destroys
added in an unpredictable manner, such as the track so affected – and the data stored
the SYMBOL TABLE of a compiler. Each item in that track. Particles of surface material
of information to be placed in the table or produced by the contact rapidly spread
file has a unique key. A special hashing al- around and cause other tracks to be de-
gorithm uses the key to allocate a position stroyed. The damaged disk usually has to
to each item, the collection of items being be thrown away. Disks should be BACKED
distributed fairly evenly over the table or UP at regular intervals so that in the event
file. The same algorithm is used to yield the of a crash a duplicate is available.
starting point for a search for the key.
header 1. Text or an image that is re-
hash total /hash/ See control total. peated at the top of pages.
2. In digital communications, information
Hayes compatible /hayz/ Describing a at the beginning of a packet of data con-
modem or system that has the same set of taining details about the sender, the re-
commands as a Hayes Microcomputer ceiver, the length of the packet, and,
Products modem. Although the company sometimes, the protocol being used.
no longer exists, Hayes compatibility re- 3. The part of an e-mail that precedes the
mains a de facto standard for microproces- main message and contains the name of the
sor modems. sender, the addressee and other recipients,
the date and time, etc.
HDTV (high-definition television) A
television with higher resolution than one help On-line documentation containing
based on the NTSC standard. HDTV sets information about a particular application
have a wider screen and roughly twice the or operating system.
normal resolution. The picture quality is
similar to 35-mm movies with sound qual- helper application A PLUG-IN that runs
ity similar to that from a CD. HDTV signal in a browser’s window so that streaming
transmissions are digital rather than ana- video, sound, and other processes can be
log, and need to be compressed, using the used in the Web page. The helper applica-
MPEG-2 file format and compression tion is launched when the browser down-
standard, before transmission through the loads a file it cannot process. Many
narrow TV lines. The signals are then de- modern Web browsers no longer need
compressed in the television receiver. helper applications and can cope with the
more common multimedia file formats
HDX See half-duplex transmission. themselves.
head The part of a peripheral device that hertz /herts/ The SI unit of frequency. A
is in contact with the data medium, or very frequency of 1 hertz is obtained if one cycle
close to it, and that is responsible for writ- of an event is repeated every second.
ing data on the medium or for reading or
erasing data. For example, DISK DRIVES heuristic /hyû-riss-tik/ A method of
have magnetic heads; a magnetic head may solving a problem that, unlike an ALGO-
be a read head, a write head, an erase head, RITHM, cannot be guaranteed always to
or a read/write head, depending on its find a good solution or to run in an accept-
function. able time. Heuristics are used where there
is no satisfactory algorithm available and
head crash (disk crash) The accidental with the acceptance that, while they will
and disastrous contact of a read/write head usually work better than an algorithm, in
95
hex
some cases they will work very badly or hidden file A file that has the hidden at-
even fail. Steps must be taken to ensure tribute set and is not normally visible to
that such cases cannot arise in practice, or users; i.e. is not shown on directory listings
can be detected if they do and an alterna- or as an ICON when its FOLDER is opened in
tive heuristic used. Systems based on a window. Files are usually hidden to pro-
heuristics often monitor their own perfor- tect them from deletion or modification
mance and adjust their heuristics accord- and tend to contain critical system code or
ingly. Heuristic methods have been data. Most file-management utilities allow
employed, for example, to control the the user the option to view hidden files.
routing of data through computer NET-
WORKS: the actual performance of a net- hierarchical database A type of data-
work is used to influence subsequent base that is structured as a TREE: all
decisions on routing. Heuristic techniques records, or nodes, have one parent node
have also been applied to problems in AR- (except for the root node) and possibly one
TIFICIAL INTELLIGENCE and to the detection or more child nodes. No other kind of link
of VIRUSES and SPAM. between nodes is allowed. Hierarchical
databases are less powerful than RELA-
hex /heks/ See hexadecimal notation. TIONAL DATABASES. See also network data-
base.
hexadecimal notation /heks-ă-dess-ă-
măl/ (hex) A number system that uses 16 hierarchical file system A FILESTORE
digits and thus has BASE 16. The 16 digits where files are organized in directories
are represented by (folders) that contain other directories or
0, 1, 2, … 9, A, B, C, D, E, F folders as well as files. The highest level di-
It is a positional notation (see number sys- rectory is called the root and the directories
tem), positional values increasing from passed through to reach a file is the path.
right to left by powers of 16. Hex is a con- Introduced by the Unix operating system in
venient shorthand by which people (rather the 1970s, hierarchical file systems are
than computers) can handle binary num- now the norm.
bers. Each hex digit corresponds to a group
of 4 binary digits, or bits, as shown in the hierarchy of operators The ORDER OF
table. Conversion of binary to hex is done PRECEDENCEof OPERATORS in a particular
by marking off groups of 4 bits in the bi- programming language.
nary number (starting from the right) and
replacing each group by its hex equivalent. high-definition television See HDTV.
Conversion of hex to binary is done by re-
placing each hex digit by its equivalent bi- high-level language A type of PRO-
nary group. See also binary notation. GRAMMING LANGUAGE whose features re-
flect the requirements of the programmer
rather than those of the computer. It
HEXADECIMAL AND BINARY
achieves this by being designed for the so-
EQUIVALENTS
lution of problems in one or more areas of
application. High-level languages are thus
Binary Hex Binary Hex
described as applications-orientated (or
0000 0 1000 8 problem-orientated). They are easier for
0001 1 1001 9 the programmer to use than LOW-LEVEL
0010 2 1010 A LANGUAGES, being closer to natural lan-
0011 3 1011 B guage and to the language of mathematics.
0100 4 1100 C The programmer needs little knowledge
0101 5 1101 D of the computer on which the program is to
0110 6 1110 E run: unlike a low-level language a high-
0111 7 1111 F level language does not reflect the facilities
provided by computer hardware. Before
96
housekeeping
being run on a particular computer, how- the pointer to the beginning of the line or
ever, programs written in a high-level lan- to the top of a menu, or to some recognized
guage must be translated into a form that home position in an application.
can be accepted by that computer, i.e. they
must be converted into MACHINE CODE. home page 1. The entry page of a hy-
This is achieved by means of a special pro- pertext system, usually acting as the intro-
gram – either a COMPILER or an INTER- ductory page on a Web site. The address
PRETER. Each statement in the original for a Web site is usually the home page ad-
program – called the source program – is dress.
translated into many machine instructions. 2. An individual’s personal Web page.
A source program can be translated into
different machine codes using different host (host computer) 1. A computer run-
compilers or interpreters. As a result, pro- ning a SERVER, or that provides services in
grams written in high-level languages can some other way to computers on its net-
be moved from one computer to another. work. Hosts can range from microcomput-
The first high-level languages were re- ers to mainframes. For example, the host
leased in the late 1950s. The number and for a Web site is the computer whose Web
variety now available is large and still server software manages the site and
growing. Examples include BASIC, COBOL, makes it available on the WWW.
FORTRAN, PASCAL, ADA, PROLOG, JAVA, and 2. In mainframe computing, the host is the
C++. Some high-level languages are aimed mainframe and may have intelligent or
at specific areas of application. Cobol, for dumb workstations attached to it that use
instance, was developed for commercial it as a host provider of services.
data processing while Fortran is widely
used for scientific computation. Java was host computer See host.
originally designed for programs down-
loaded from the Internet. Other languages, hot key A user- or application-defined
such as Basic, Pascal, C++, and Ada, are key combination that executes a command
designed to be of more general use. or causes the operating system to switch to
another program. For example, in Win-
highlight To change the appearance of dows applications Ctrl-S often saves the
displayed text, giving it emphasis. Data file currently being edited.
that has been SELECTED is usually high-
lighted, by displaying it in reverse video. hotlist A list of frequently accessed Web
sites. See also bookmark.
history A list of actions recently per-
formed by a user. For example, many ap- hot swapping (hot plugging) The re-
plications have a feature that shows the last placement of a hard drive, CD-ROM drive,
few files opened, so that the user is able to or other device without the need for
open a previously used file quickly. Web switching off the computer or reconfigura-
browsers also display a history of the pre- tion or resetting of system parameters.
vious Web sites visited.
housekeeping Actions performed within
hit A success in matching data. a program or within a computer system in
order to maintain internal orderliness. In
HMD (head mounted display) The the case of a program it may, for example,
headgear worn in VIRTUAL REALITY applica- involve the management of storage space
tions. by freeing areas no longer required (see
garbage). In the context of an entire com-
home computer A computer specifi- puter system, housekeeping involves back-
cally aimed at the home computing market. ing up the filestore, deleting files that are
no longer required or whose expiry dates
home key The keyboard key that moves have passed, running appropriate pro-
97
HTML
98
I
IA-64 A 64-bit processor and instruc- signed Names and Numbers) A private
tion set based on explicitly parallel instruc- non-profitmaking corporation that since
tion computing (EPIC) design philosophy, 1998 has administered Internet IP AD-
produced by Intel. In 64-bit mode these DRESSES, DOMAIN NAMES, and some techni-
processors can calculate two bundles on up cal matters.
to three instructions at a time. However in
32-bit mode they are slower because de- icon A pictorial symbol or small picture
coders must first translate 32-bit instruc- used in a GRAPHICAL USER INTERFACE to rep-
tion sets into 64-bit instruction sets. The resent an available resource (a file, a
ITANIUM was the first in Intel’s line of these printer, etc.). Double-clicking an icon or
processors. dragging it onto another icon will initiate
an appropriate action.
IAP (Internet access provider) See ISP.
IDE (integrated drive electronics; intelli-
IAR See instruction address register. gent drive electronics) An interface for
mass storage devices that is built into the
IAS (immediate access store) See main disk or CD-ROM. This replaced the sepa-
store. rate drive-specific disk controller card that
had formerly been required. The result was
IBG (interblock gap) See block. greater flexibility in the choice of disk dri-
ves because all makes of IDE disk drive
IBM International Business Machines; communicated with the rest of the com-
formerly the largest computer manufac- puter in a standard way that concealed the
turer in the world, formed in 1911. The details of their internal workings. Devel-
company’s main business was originally as oped in the 1980s, IDE has been super-
a producer of punched-card tabulating ma- seded by EIDE.
chines, and then later in mainframe com-
puting and office equipment manufacture. identifier A STRING of one or more let-
Its first microcomputer, called the IBM PC, ters, digits, or other characters selected by
was built in the early 1980s and set stan- a programmer to identify some element in
dards that were widely adopted. a program. An identifier can be a NAME or
a LABEL. The kind of element that can be
IBM-compatible (PC-compatible) De- identified depends on the programming
scribing a personal computer that was a language. It could, for example, be a VARI-
clone of the IBM PC and its successors. The ABLE, an ARRAY, a PROCEDURE, or a STATE-
PC’s open architecture was further devel- MENT. When the program is compiled or
oped and improved and was subsequently assembled, prior to execution, identifiers
known as industry-standard architecture are converted to machine addresses using a
(ISA). SYMBOL TABLE. See also literal; reserved
word.
IC See integrated circuit.
idle time Time during which the central
ICANN (Internet Corporation for As- processor of a computer system is perform-
99
IDSL
if then else statement A simple condi- image Any picture or diagram stored ei-
tional CONTROL STRUCTURE that appears in ther as a set of pixels (raster image) or as a
most programming languages and allows series of instructions for reproducing the
selection between two alternatives; the picture (vector image).
choice is dependent on whether a given
condition is true or not. An example is image compression The use of data
if it rains this afternoon compression techniques on graphics files to
then we go to the cinema reduce the amount of storage required. See
else we go for a walk also compression.
Most languages also provide an if then
statement to allow conditional execution image map An image on a Web page
of a single statement or group of state- that contains links to other parts of the
ments, for example page, other pages, or to files. On older Web
if you have a dog with you browsers these maps were executed
then put it on a lead through CGI scripts. However, with newer
100
index
browsers the maps are executed by the immediate addressing The process by
browser itself. which the data needed as the OPERAND for
an operation is actually held in the associ-
image processing (picture processing) ated MACHINE INSTRUCTION. Although the
The analysis and manipulation of informa- data contained there is usually restricted in
tion contained in images. The original sub- size, immediate addressing provides a con-
ject can be an actual object or scene, or a venient and quick way of loading small
photograph, drawing, etc. A numerical numbers into a REGISTER or ACCUMULATOR.
version of this subject is obtained electron- See also addressing mode.
ically, for example by scanning: the origi-
nal is effectively divided into a large implementation 1. The working ver-
number of tiny individual portions, and a sion of a given design of a system.
set of numbers is produced, corresponding 2. The processes involved in developing a
to the measured brightness and color of working version of a system from a given
these portions. The original is thus con- design.
verted into a two-dimensional ARRAY of 3. The way in which some part of a system
data. The individual portions are known as is made to fulfil its function.
pixels; the greater the number of pixels per
image, the greater the detail available. The import To bring some external object,
original is said to have been digitized. such as a picture or another file, into the
This numerical version of the original is current system or application. The format
of the information being imported must be
usually stored in a computer and can be
supported by the system receiving the data.
manipulated in various ways to highlight
different aspects of the original. For exam-
inactive window In a system capable of
ple, a specified range of brightness can be
displaying multiple windows, a loaded
increased, or two slightly different images
window that is not currently being used. It
can be compared or superimposed. The
may be totally or partially hidden by other
final form of the image can be produced on
windows and becomes active (receives the
a screen, printer, etc., and information de- focus) when selected by the user.
rived from the image can appear in graphs
or tables and can be further analyzed. inclusive OR A Boolean operator that
returns the value of true if either or both of
imagesetter /im-ij-set-er/ A typesetting the operands is true.
device that produces a very high resolution
output (paper or film). Imagesetters are ex- incremental backup A backup system
pensive and are used by specialized compa- in which the only data backed up is that
nies or bureaus. Most users provide changed since the last backup.
PostScript or PDF files, which have been
proofed by printing to PostScript laser index 1. A data structure consisting of a
printers, and are then processed by the im- set of POINTERS that can be used to locate
agesetter. records in another data structure (often a
DATA FILE) based on the value of a KEY
IMAP See POP3. within the records. An index allows
records where the key has a particular
IMHO In my humble opinion; one of value or range of values to be accessed di-
many abbreviations in common use on the rectly, without having to search for them
Internet, for example in newsgroups and e- sequentially through the entire structure. A
mails. It is used to soften a criticism in small index might be a simple list of key
order to avoid a FLAME. values, with pointers to the records in
which they occur, arranged in ascending
immediate access store (IAS) See main order. However, large indexes have more
store. complex structures to minimize lookup
101
indexed addressing
times. Common uses of indexes are in data- by a variety of methods, depending on the
bases and in INDEXED FILES and INDEXED SE- type of computer; it may for example be
QUENTIAL FILES. loaded from the ACCUMULATOR.
2. The value held in an INDEX REGISTER.
3. (subscript) A value, usually an integer or indirect addressing An ADDRESSING
integer expression, that is used in selecting MODE in which the address specified in a
a particular element in an ARRAY. MACHINE INSTRUCTION identifies a storage
location that itself holds an address. The
indexed addressing (indexing) An AD- address part of the instruction is thus an
DRESSING MODE in which the address given address of an address. It is called an indi-
in a MACHINE INSTRUCTION is modified by rect address. The second location identified
the contents of one or more INDEX REGIS- generally holds the required item of data,
TERS; the contents of the index register(s) but may however hold a further address.
are usually added to the address. The result
identifies a storage location. The address infection The corruption of a computer
specified in the machine instruction is system by a VIRUS.
called an indexed address. It may be an ac-
tual address or an address of an address infinite loop A LOOP in a program from
(see direct addressing; indirect addressing). which, under certain circumstances, there
is no exit except by terminating the pro-
indexed file A DATA FILE in which gram. This can be done accidentally, per-
records can be accessed by means of an haps by a while loop whose condition can
INDEX. This is stored in a separate portion
never be false, or deliberately when a pro-
of the file. An indexed file has characteris-
gram is to repeat a sequence of actions
tics of a random-access file. See also file; in-
until stopped by some external event such
dexed sequential file.
as an interrupt or termination by the oper-
ating system.
indexed sequential file A FILE whose
records are organized in such a way that
they are written sequentially and can be
infix notation /in-fiks/ A form of nota-
tion in which an OPERATOR is placed be-
read sequentially or by means of an index
(see file; indexed file). The index can be tween its OPERANDS, as in the arithmetic
used to read data randomly or to skip over expression
unwanted records in a sequential read. The a+b*c
records are accessed by means of an in- If no brackets are used the expression is
dexed sequential access method (ISAM). evaluated in a computer according to the
An indexed sequential file is similar to a ORDER OF PRECEDENCE of the programming
102
initialize
103
ink-jet printer
input device Any device that transfers install /in-stawl/ 1. To store and config-
data, programs, or signals into a computer ure software permanently so that it can be
system. Examples include KEYBOARDS, run on a particular computer and make use
graphical devices such as a MOUSE or DIGI- of the various peripherals attached to that
TIZING PAD, VOICE INPUT DEVICES, SCANNERS computer. Software and games can be ob-
and DOCUMENT READERS, SAMPLING devices tained on CDs or DVDs and installed from
that convert analog data, such as video and these media, or they can be downloaded
sound, to digital form, and independent and installed from the Internet.
digital devices, such as digital cameras and 2. To add hardware to a system.
camcorders, whose data can easily be
transferred to a computer. An input device installation program (setup program)
need not be operated by a person – it could A program provided with an operating sys-
104
integer arithmetic
tem or application program that facilitates the next instruction from the INSTRUCTION
the process of installation. The user is ADDRESS REGISTER, fetches this instruction
guided through a question-and-answer ses- from main store (via the MEMORY DATA
sion that will enable the software to be set REGISTER), and places it in the instruction
up and run effectively. register. The control unit can then interpret
the instruction in the instruction register
instruction See machine instruction. and cause the instruction to be executed.
See also microprogram.
instruction address register (IAR) A
REGISTER, i.e. a temporary location, in the instruction set All the instructions that
CONTROL UNIT from whose contents the AD- are available in a particular MACHINE CODE
DRESS of the next machine instruction is de- or ASSEMBLY LANGUAGE, i.e. all the instruc-
rived. It thus stores in turn the addresses of tions that a particular computer is capable
the instructions that the computer has to of performing. See machine instruction.
carry out. Each time the control unit
fetches an instruction, it automatically in- INT /int/ See Basic.
creases the contents of the IAR by one. If
the sequence of the instructions is altered integer /in-tĕ-jer/ Any positive or nega-
by an unconditional JUMP INSTRUCTION, the tive whole number, or zero. Integers be-
contents of the IAR are changed to the ad- long to the set … –4, –3, –2, –1, 0, +1, +2,
dress indicated by the jump instruction; in +3, +4, …
the case of a CONDITIONAL JUMP instruction, Negative integers carry a minus sign to dis-
the contents of the IAR are changed only if tinguish them from positive integers. Posi-
the condition is met. tive integers do not necessarily carry a plus
Since the contents of the IAR are incre-
sign. Any integer without a sign is said to
mented so as to point to the start of the
be unsigned.
next instruction, the IAR is a COUNTER and
Unsigned integers are stored in a com-
hence is also called an instruction counter
puter in BINARY NOTATION, i.e. as a se-
or a program counter.
quence of bits making up a WORD; the
sequence is the binary equivalent of the in-
instruction counter See instruction ad-
teger. In a machine with a 16-bit word, 216
dress register.
(i.e. 65 536) possible values can be stored,
instruction cycle See fetch-execute ranging from 0000000000000000 to
cycle. 1111111111111111. These represent the
integers from 0 to 216 – 1, i.e. from 0 to
instruction format The layout of a MA- 65 535. Signed integers are usually stored
CHINE INSTRUCTION, showing its con- in two’s complement (see complement),
stituent parts. and a 16-bit word then stores values from
–215 to +(215 – 1), i.e. from –32 768 to
instruction overlap See control unit. +32 767. See also integer arithmetic; real
number.
instruction prefetch /pree-fech/ See
control unit. integer arithmetic Arithmetic involving
only INTEGERS. The addition, subtraction,
instruction register (IR; current instruc- and multiplication of integers always pro-
tion register, CIR) A REGISTER (i.e. a tem- duces an integer as a result. In integer arith-
porary location) usually in the CONTROL metic, an integer division also always
UNIT, used for holding the MACHINE IN- produces an integer result, unlike normal
STRUCTION that is currently being per- arithmetic. This is achieved by ignoring the
formed or is about to be performed. Once remainder. Hence in integer division 7 di-
the control unit has finished execution of vided by 2 equals 3 (rather than 3.5). Other
one instruction, it obtains the address of examples are shown in the table overleaf.
105
integer division
106
interface
In 2005 Pentium chips were produced with clock speeds of over 3.5 GHz, with about
170 million transistors, and a 64 bit data width.
107
interface card
which was developed jointly by several now the Internet offers such services as e-
computing companies. See also RS232C mail, World Wide Web, IRC, Telnet, FTP,
interface. and news channels and is used for the ex-
In software, an interface of a module is change of data, news, conversation, and
a set of FUNCTIONS, PROCEDURES, and PROP- commerce. It is still decentralized in that no
ERTIES that can be used by the rest of the one person, organization, or country con-
program. Interfaces are important in soft- trols it or what is available through it.
ware development, for they allow different
modules of a program to be developed by Internet access The capability of a user
different teams, or even different compa- to connect to the Internet. A home user and
nies, that do not need to know the details many smaller organizations generally con-
of each other’s work. If all interaction be- nect through a MODEM or ADSL TRANS-
tween the modules is via agreed interfaces, CEIVER connected to a personal computer
the modules can be treated as BLACK BOXES. and, via a telephone line, to an ISP. Larger
Standards that include systems for specify- users may connect through a dedicated
ing interfaces, such as COM, CORBA, and line. Communications can also be estab-
.NET, allow pre-built third-party compo- lished using TV set-top boxes and cell-
nents to be integrated into a program. phones, although these tend to offer a
restricted service.
interface card (interface board) See ex-
pansion card. Internet address See domain name.
interlacing /in-ter-layss-ing/ 1. A tech- Internet Engineering Task Force See
nique for raster scanning in which all odd
IETF.
numbered scan lines are updated in one
sweep of the electron beam and all even
Internet Explorer (IE) A Web browser
numbered in the next sweep. This means
produced by Microsoft and introduced in
that the number of lines displayed and the
1995. It is, controversially, integrated into
information carried by the display signal
WINDOWS.
are halved. Images produced by interlacing
are not as clear as those produced by pro-
gressive scanning and, since the reaction Internet presence provider (IPP) A
time is slow, images such as animation and company that provides disk space, high
video may flicker or appear streaked. speed Internet connection, and possibly
2. The technique of displaying a graphics Web design and other services for compa-
image so that alternating rows are dis- nies or individuals requiring a presence on
played in separate passes. The effect is that the Internet.
the image is displayed quickly and then the
details are filled in gradually. It is often Internet protocol See IP.
used for images on Web pages.
Internet protocol security (IPSec) A
internal modem A MODEM that resides standard for security at the network or
on an expansion board that fits into one of packet-processing layer of a TCP/IP net-
the expansion slots on a personal com- work system. (Previously security has been
puter. Compare external modem. applied at the application layer of the com-
munications model.) IPSec is useful for vir-
Internet /in-ter-net/ (the Net) The com- tual private networks (VPNs) and for
plete worldwide system of networks and remote access through dial-up connections
gateways that communicate using the to a private network.
TCP/IP group of protocols. It was set up as
a decentralized network called ARPANET Internet relay chat (IRC) An Internet
by the US Department of Defense in 1969. service provided by an IRC server, en-
Later other networks were connected and abling a user to CHAT with other users.
108
IP address
Internet service provider (ISP) See ISP. there are different procedures – interrupt
service routines – to be executed for each
Internet Voice See VoIP. kind of interrupt, and (c) when the correct
action has been taken, control is returned
internetworking The process and sci- to the task that was being processed when
ence of connecting individual LANs to cre- the interrupt occurred.
ate WANs and connecting WANs to form Interrupts can be caused by an external
even larger WANs. This is an extremely event, such as a key being depressed on a
complex technology and is accomplished keyboard. They can also be caused by in-
using routers, bridges, and gateways to ternal conditions in the processor, such as
deal with the differing protocols used the detection of an attempt to divide by
across the various networks. zero. This kind of interrupt is called a hard-
ware interrupt. A software interrupt works
InterNIC /in-ter-nik/ (Internet network in- in the same way but is triggered by a MA-
formation center) The organization for- CHINE INSTRUCTION in a program. Each in-
merly responsible for registering and terrupt has a priority: higher-priority
maintaining top-level Internet domain interrupts may interrupt the servicing of
names. Its functions have been taken over lower-priority interrupts but not the other
by ICANN. way round.
109
Ipng
this means. However, the growth of the In- ISAM /ÿ-sam/ (indexed sequential access
ternet is making this number inadequate. method) See indexed sequential file.
This problem is confronted by IPV6, which
supports 128-bit IP addresses. ISDN (integrated services digital network)
A telephone service that combines voice
Ipng (Internet protocol next generation) and digital network services in a single
An old name for IPV6. medium. Most ISDN lines offered by tele-
phone companies give two 64 kbps lines at
iPod Trademark A range of MP3 PLAYERS once, where one can be used for voice and
manufactured by Apple Computer, Inc. one for data, or both for data, which dou-
First released in 2001, the iPod has come to bles the transmission rate. For computing
purposes, such as Internet access, ISDN is
dominate the MP3-player market.
being superseded by ADSL.
IPTV The delivery of television over the
ISO The International Organization for
Internet. This is not yet practical because it
Standardization, the body that establishes
requires a higher BANDWIDTH than is pro-
international standards in industry, etc.,
vided by current BROADBAND connections, including those used in computing. Exam-
but it is expected to become important in ples include standards for many HIGH-
the next few years. LEVEL LANGUAGES, CHARACTER SETS and
CHARACTER ENCODINGS, aspects of net-
IPv6 (Internet protocol version 6) The working, and data formats for CDs and
latest level of the Internet protocol. Future other storage devices. While compliance
growth of the Internet is provided for by with such standards encourages interoper-
the lengthening of IP addresses from 32 ability, many manufacturers’ products de-
bits to 128 bits, enabling many more ad- viate from the relevant standard(s) for
dresses to be assigned. However, the adop- commercial reasons. See also ANSI.
tion of IPv6 has so far been slow because of
the problems and expense of integration ISO-7 A CHARACTER ENCODING scheme
with the existing, IPv4-based, Internet. See developed by ISO and hence internationally
also IP; IP address. agreed. ISO-7 uses 7 bits for each charac-
ter, and so provides 27, i.e. 128, different
IR See instruction register. bit patterns. Certain bit patterns are re-
served for national use, allowing countries
IRC See Internet relay chat. to include the symbols used for their cur-
rency, the accents used above and below
IRQ (interrupt request) A hardware line letters, etc. The US version is ASCII, which is
over which a device can send an inter- a code widely used in computing.
rupt signal to a computer. Sometimes DIP
ISP (Internet service provider) A com-
switches are used to set a device’s IRQ
pany that provides access to the Internet,
number before attaching to a computer.
usually for a charge. A software package,
Conflicts can occur when adding EXPAN-
username, password, and (if appropriate)
SION CARDS, but PLUG-AND-PLAY technology
access phone number are provided, and ac-
has largely resolved this problem. cess can be obtained through a modem,
ADSL transceiver, or by direct connection
ISA /ÿ-să/ (industry standard architecture) from a company’s network. An ISP is also
A standard bus architecture that was orig- called an Internet access provider (IAP).
inally associated with the IBM AT mother-
board. It allowed for 16 bits to flow IT /ÿ-tee/ See information technology.
between the motherboard and an expan-
sion card and its associated devices. It is Itanium /ÿ-tay-nee-ŭm/ The first mem-
now obsolete. See also EISA. ber of Intel’s IA-64 family of processors. It
110
iterative
111
J
112
jump
devices and any computer that can run duced and often used as a trail to aid re-
Java will be able to access this code. covery in the event of problems occurring.
It is used to contain details of network
JIT compiler See just-in-time compiler. transactions, such as file usage or message
transfer information.
job A set of programs and the data to be
manipulated by these programs, regarded joystick A device for generating signals
by a computer system as a unit of work. that can control movement on a display. It
The complete description of a job, required is a shaft a few inches in height that is
for its execution, is written in a JOB CON- mounted in an upright position in a base
TROL LANGUAGE. and can be pushed or pulled by the opera-
tor in any direction. Generally, the device
job control language (JCL) A special also has control buttons. Joysticks are
kind of programming language by means most commonly used in computer games.
of which a user can communicate with the
OPERATING SYSTEM of a computer. It is em- JPEG /jay-peg/ (joint photographic experts
ployed most often in connection with group) A compression standard common
BATCH PROCESSING format for image files. It is especially used
A JCL is used to write the sequence of for photographic images. Files in this for-
commands that will control the running of mat often have the extension jpg.
a JOB on a particular computer: an initial
command identifies the job while subse- Julia set A set of points in the complex
quent commands are each an instruction plane defined by iteration of a complex
for the operating system to take some ac- number. One takes the expression z2 + c,
tion on behalf of the user. For example, where z and c are complex numbers, and
there are usually commands to run a spec- calculates it for a given value of z and takes
ified program, find input data at a specified the result as a new starting value of z. This
source, and output results at some speci- process can be repeated indefinitely and
fied destination. The objects manipulated three possibilities occur, depending on the
in a JCL are thus not VARIABLES as in a nor- initial value for z. One is that the value
mal programming language but can, for tends to zero with successive iterations.
example, be complete programs or the Another is that the value diverges to infin-
input or output streams for these pro- ity. There is, however, a set of initial values
grams. A JCL command can be obeyed as of z for which successive iterations give val-
soon as it is keyed. Alternatively a com- ues that stay in the set. This set of values is
plete set of JCL commands can be stored a Julia set and it can be represented by
and executed later. See also command lan- points in the complex plane. The Julia set is
guage. the boundary between values that have an
attractor at zero and values that have an at-
job queue/stream The lining up of jobs tractor at infinity. The actual form of the
for a computer device such as a printer. Julia set depends on the value of the con-
stant complex number c. Thus, if c = 0, the
Josephson junction /joh-zĕf-sŏn/ The iteration is z → z2. In this case the Julia set
junction between a thin layer of insulating is a circle with radius 1. There is an infinite
material and a superconducting material number of Julia sets showing a wide range
across which a superconducting current of complex patterns. See also Mandelbrot
can flow in the absence of an applied volt- set.
age. Josephson memory, a cryogenic mem-
ory, uses Josephson cells that contain jump (branch) A departure from the
Josephson junctions, and is inherently ex- normal sequential execution of program
tremely fast but expensive to run. instructions. It is used when writing a pro-
gram in ASSEMBLY LANGUAGE or MACHINE
journal A log that is computer-pro- CODE. Instead of performing instruction 1,
113
jumper
then instructions 2, 3, 4, and so on, there word can be used if the space in a line is too
could be a jump from, say, instruction 3 to great. WORD PROCESSING systems allow the
instruction 19. This departure from nor- user to specify both the right and the left
mal program flow is specified by using a margin, between which text is automati-
special instruction in a program, known as cally justified as it is typed in; hyphenation
a jump instruction (or branch instruction). is offered as an option.
A jump can be either unconditional or 2. To shift the contents of a REGISTER, if
conditional. An unconditional jump takes necessary, so that the bit at a specified end
place whenever the instruction specifying it (right-hand or left-hand) of the data en-
is executed (see instruction address regis- tered in the register is at a specified posi-
ter). A conditional jump takes place only if tion (usually the right- or left-hand end) of
specified conditions are met, for example if the register. The contents are then said to
the contents of the accumulator equal zero be right-justified or left-justified, as the
or if the value of some variable exceeds a case may be.
stated number.
Jump instructions offer the only way of just-in-time compiler (JIT compiler) A
controlling the flow of events when using compiler that operates when a program is
assembly language or machine code. This run. Such programs are normally compiled
can lead to programming errors. Most in two stages. The source code is first
high-level languages offer improved facili- processed by a normal compiler, which
ties in the form of a variety of CONTROL checks for syntax errors, performs opti-
STRUCTURES. See also GOTO statement.
mizations, etc.; however, this compiler
does not generate MACHINE CODE but rather
jumper A small piece of plastic-encased
a machine-independent code, known as in-
metal that closes a circuit so that electricity
termediate language or bytecode. When
can flow through. The jumper is placed
the program is run, the JIT compiler trans-
across two metal pins sticking out of a cir-
lates this code into machine code for each
cuit board. Sometimes jumpers are used to
configure expansion boards; by placing program section when it is first used. (This
them over different sets of pins the board’s differs from an INTERPRETER, which trans-
parameters can be changed. lates each program section every time it is
used.) Unlike a traditional compiler a JIT
jump instruction (branch instruction) compiler is aware of its operating environ-
See jump. ment and can vary the machine code it gen-
erates accordingly; for example it can often
justify 1. To control the positions of determine the exact version of the comput-
printed or typed characters so that regular er’s central processor and take advantage
vertical margins to a page or column of of any special MACHINE INSTRUCTIONS that
text are obtained; the lines are then said to version provides to produce optimal code.
be justified. In some cases only one margin The slight overhead of compiling during
is aligned; the lines are then said to be left- program execution has been shown not to
justified when the left-hand margin is regu- affect the speed of many applications to
lar or right-justified when the right-hand any significant degree; nevertheless there
margin is regular. When both margins are remain programs where this overhead is
to be aligned, the spaces between words significant, and for these traditional com-
(and sometimes between letters) is adjusted pilation remains the appropriate method.
until each line fills the distance between the
two specified margins. Hyphenation of a JVM See Java virtual machine.
114
K
function keys
Print Scroll Pause
Esc F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12 Screen Num Caps Scroll
Lock
SysRq Break Lock Lock Lock
: ~ 6
+
Caps A S D F G H J K L @ 4 5
Lock
; ‘ #
| Z X C V B N M < > ? 1 2 3
\ , . / End Pg Dn
Enter
0 •
Ctrl Alt Alt Gr Ctrl
Ins Del
A typical keyboard
115
key frame
consists of an array of labeled keys that are keystrokes it takes to perform common op-
operated by pressure applied by the fin- erations.
gers. In current devices the operation of a
particular key generates a coded electrical key to disk An obsolete system of data
signal that can be fed directly into a com- entry in which the data entered by a num-
puter. The characters that have been keyed ber of keyboard operators was recorded on
may be displayed on a VDU screen. a magnetic disk under the control of a
A computer keyboard consists of the small computer. The data from each key-
standard typewriter layout – the QW- board was routed by the computer to the
ERTY keyboard – plus some additional appropriate file on the disk. The data en-
keys (see diagram). These can include a coded on the disk was verified, often by
control key, function keys, cursor keys, comparing it with data entered by a second
and a numeric keypad. The control key op- operator working from the same source.
erates in the same way as a shift key but al- The disk could then be transferred to the
lows CONTROL CHARACTERS to be sent to computer on which the data was to be
the computer; the function keys send not processed.
one but a whole sequence of characters to
the computer at a time, and can often be key to tape An obsolete system of data
programmed by the user to send com- entry in which the data entered by a key-
monly used sequences; the cursor keys are board operator was recorded on a mag-
used to move the screen CURSOR to a new netic tape. The data encoded on the tape
was then verified, often by comparing it
position; and the numeric KEYPAD dupli-
with data entered by a second operator
cates the normal typewriter number keys
working from the same document. The
and speeds up the entry of numerical data
tape could then be transferred to a com-
by allowing one-handed operation. See
puter so that the data could be processed.
also alt key.
kHz See kilohertz.
key frame An animation technique in
which the start and end frames of a se-
kill To stop a process in a program or
quence are defined but the intervening operating system.
frames are interpolated by a computer to
give a smooth animation. kill file A feature of some e-mail clients
or newsgroup readers, or an independent
keypad A small KEYBOARD with only a Internet utility, that is used to block out or
few keys, often 12 or 16, that can be oper- filter unwanted messages or newsgroup ar-
ated with one hand. It is usually used for ticles from particular individuals. See also
encoding a particular sort of information. spam.
For example, a numeric keypad is used to
feed numerical data into a computer, and kilo- /kil-ŏ-/ A prefix indicating a multi-
can be operated much faster than an all- ple of 1000 (i.e. 103) or, loosely, a multiple
purpose keyboard; there are 10 keys for the of 210 (i.e. 1024). In science and technology
digits 0–9 plus additional keys for, say, the decimal notation is usually used and pow-
decimal point and minus sign. A keypad ers of 10 are thus encountered – 103, 106,
may be hand-held or may be part of a etc. The symbol k is used for kilo-, as in kV
larger keyboard. for kilovolt. In communications, kilo- (as
in kilobaud) means 1000. Binary notation
keystroke The action of pressing a key- is generally used in computing and so the
board key to enter a character or initiate a power of 2 nearest to 103 (i.e. 210 or 1024)
command in a computer system. A key- has assumed the meaning of kilo-.
boarder’s efficiency can be measured in The prefix is most frequently encoun-
keystrokes per minute, or ease of use of an tered in computing in the context of stor-
application can be assessed by how many age CAPACITY. With magnetic disks and
116
knowledge domain
main store, the capacity is normally reck- bust interface, such as a TOUCH SCREEN,
oned in terms of the number of BYTES that that anyone can use without training or
can be stored, and the word kilobyte (KB) documentation.
is used to mean 210 bytes; kilobyte is usu-
ally abbreviated to K byte or just to K, thus knowledge acquisition The transla-
a 256 K byte file is 262 144 bytes in size. tion of human knowledge into a form suit-
SEMICONDUCTOR MEMORY – RAM or ROM able for use on a computer. The term is
– is sometimes considered in terms of the used in EXPERT SYSTEMS.
number of bits that can be stored in the de-
vice, and the word kilobit (kb) is used to knowledge base A database of accumu-
mean 210 bits. lated human knowledge on a particular
Kilo- is part of a sequence subject used in EXPERT SYSTEMS. For exam-
kilo-, mega-, giga-, tera-, peta-, … ple, a knowledge base of symptoms for a
of increasing powers of 103 (or of 210). particular disease would be used in medical
diagnosis.
kilohertz /kil-ŏ-herts/ (kHz) A unit of
frequency equal to 1000 HERTZ. knowledge-based system See expert
system.
kiosk A booth providing a computer-
related service such as a theater booking knowledge domain The area of knowl-
service or tourist information service. A edge in which a particular EXPERT SYSTEM is
kiosk requires a very simple, physically ro- involved.
117
license agreement
LCD (liquid crystal display) A device dividual LEDs can be made to emit light
used in many FLAT PANEL MONITORS. The (usually red) so as to form the shape of a
screen is composed of many dots, each of particular character. A simple SEVEN-SEG-
which contains a liquid that is usually MENT DISPLAY can display the numbers 0–9
transparent. Individual dots can be dark- and some letters. LED displays require
ened however (by applying an electric more electrical power than LCDS.
field) and so an image can be built up. Note
that LCDs do not emit light, and can only left-justified See justify.
be seen because light shone from behind
the display makes the dark dots stand out. legacy system A computer, software,
Color images are achieved by each ‘dot’ in operating system, network, or other com-
fact being a triplet of red, green, and blue puter equipment that exists before a new
dots. A variant of LCD technology called system is installed, and is still required af-
ACTIVE MATRIX DISPLAY is generally used in terward. BACKWARD COMPATIBILITY is an
computer monitors. LCDs are also used in important part of the design and imple-
digital watches, calculators, cellphones, mentation of new systems, but cannot al-
etc. See also display. ways be maintained. The question of when
to engage in the expensive operation of up-
LDAP /el-dap/ (lightweight directory ac- grading legacy systems is therefore impor-
cess protocol) A significantly simpler tant.
version of the X.500 standard for a set of
protocols for accessing information direc- LEO /lee-oh/ Lyons Electronic Office.
tories. Unlike X.500 it supports TCP/IP See first generation computers.
and is an open protocol, which means that
applications do not need to know about LET /let/ See Basic.
the type of server hosting the directory.
letter quality (LQ) A quality of printing
leaf An item at the very bottom of a hi- that is as good as a top-quality electric
erarchical TREE structure. In hierarchical typewriter. The term is mainly used of dot
file systems, files are leaves whereas direc- matrix printers and is becoming less com-
tories are NODES. mon with the increased use of laser print-
ers, which generally give better quality
leased line A telephone line that is than electric typewriters.
rented for exclusive use by the customer
from one location to another. Typically lexical analysis The initial phase in the
leased lines are used by businesses to con- COMPILATION of a program during which
nect geographically distant sites for high- the source program is split up into mean-
speed data transfer. ingful units. These units could, for exam-
ple, be NAMES, CONSTANTS, RESERVED
least significant digit (lsd) The digit in WORDS, or OPERATORS. The part of the com-
the least significant, i.e. rightmost, position piler program that does this analysis is
in the representation of a number, and thus called a lexical analyzer. The units recog-
making the smallest contribution to the nized by the analyzer are known as tokens.
value of the number. With a binary num- They are output in some conveniently
ber, this digit is referred to as the least sig- coded form for further processing by the
nificant bit (lsb). compiler. See also syntax analysis.
LED display A device used in some cal- library See program library; subroutine
culators, digital clocks, etc., to display library.
numbers and letters. The characters are
formed from groups of segments. The seg- library program See program library.
ments are small electronic components
called LEDs, i.e. light-emitting diodes. In- license agreement An agreement defin-
119
LIFO
y
10
8 x + y = 10
7
x=y−2
5
x=8
1 x − 4y = k
O x
1 2 3 4 5 6 7 8 9 10 11 12
−1
−2
−3
Linear programming: possible values lie in the shaded area with the minimum value at (8,0)
ing terms for the use of hardware or soft- line 1. A hardware circuit connecting
ware by a user. two devices.
2. A single line in a computer program.
LIFO /lÿ-foh/ (last in first out) See stack. 3. A single data entry in a cache. For ex-
ample in an L2 cache each line is 32 bits
light-emitting diode See LED display. wide and can contain data from one or
more addresses.
light pen An obsolete hand-held penlike
instrument used together with a DISPLAY linear programming The process of
device for the input of data to a computer finding maximum or minimum values of a
graphics system. (The display had to be a linear function under limiting conditions
refreshed cathode-ray tube.) The main or constraints. For example, the function x
function of the light pen was to point to – 4y might be minimized subject to the con-
small areas of the screen, such as a single straints that x + y ≤ 10, x ≤ y – 2, x ≤ 8, x
character or a small graphical object, and it ≥ 0, and y ≥ 0. The constraints can be
could for example indicate a selection from shown as the area on a Cartesian coordi-
a displayed list (a MENU). It could also be nate graph bounded by the lines x + y = 10,
used to draw shapes on the screen. x = y – 2, x = 8, x = 0, and y = 0 (see dia-
120
line printer
gram). The minimum value for x – 4y is anism. Line printers are usually found on
chosen from points within this area. A se- large computers rather than personal com-
ries of parallel lines x – 4y = k are drawn puters, but their use is declining as they are
for different values of k. The line k = –2 replaced by high-volume LASER PRINTERS.
just reaches the constraint area at the point Line printers may be impact or nonim-
(8,0). Higher values are outside it, and so pact printers (see printer). Modern line
x = 8, y = 0 gives the maximum value of printers use the same technology as DOT
x – 4y within the constraints. Linear pro- MATRIX PRINTERS except that they have
gramming is used to find the best possible enough needles to print a line at a time.
combination of two or more variable quan- However, there are two older types still in
tities that determine the value of another use, barrel printers (also known as drum
quantity. In most applications, for exam- printers) and band printers (see illustration
ple, finding the best combination of quan- overleaf).
tities of each product from a factory to give A barrel printer involves a horizontal
the maximum profit, there are many vari- cylinder with rows of characters embossed
ables and constraints. Linear functions with along its length. The cylinder rotates con-
large numbers of variables and constraints tinuously at high speed, and one line is
are maximized or minimized by computer printed during each rotation. The paper is
techniques that are similar in principle to positioned between the cylinder and a set
this graphical technique for two variables. of print hammers; there is one hammer for
each printing position along a line, i.e. for
line editor See text editor. every character in a row on the cylinder.
The complete character set that can be
line feed 1. A control mechanism on a printed is provided at each printing posi-
PRINTER, or a programmed instruction, that tion. During one rotation of the cylinder all
causes the paper to move up a specified dis- the characters of one sort are printed at the
tance until the printing mechanism is op- same time; for example, all the As in a line
posite the next line to be printed. See also are printed, then all the Bs, then all the Cs,
paper throw. and so on.
2. A means by which the CURSOR on a A band printer involves a steel band
screen is moved down one line. If that with characters embossed along its length.
would result in the cursor dropping off the The band forms a loop and is rotated con-
bottom, the whole display moves up one tinuously at high speed. The band moves
line and the cursor is placed on a new horizontally, spanning the line to be
blank bottom line. printed. The paper is positioned between
3. (LF) The ASCII control character, with the band and a set of print hammers; there
the decimal value 10, that is often used to is one hammer for every printing position
invoke either of these operations. Sending along a line. As the appropriate character
a line-feed character to a printer or screen on the band passes beneath a particular
may or may not also generate a CARRIAGE hammer, the hammer strikes and prints it
RETURN. in the required position on the line. Bands
can be replaced when necessary. Mistiming
line printer A PRINTER that produces a of the hammer actions in a band printer
complete line of characters at a time, or produces slight variations in character
more precisely produces a complete line spacing; these are less obvious than the ver-
during one cycle of its operation. Printing tical displacements resulting from mistim-
rates range from 150 to 3000 lines per ing in a barrel printer.
minute (lpm). There are normally between Chain printers and train printers are
80 and 160 character positions on a line. earlier versions of the band printer. The
Each line is first assembled in the printer’s former involves small metal plates linked
memory and then printed. Once the line into a continuous chain rather than a metal
has been printed the paper is moved so that band; the latter involves metal slugs guided
the next line is opposite the printing mech- around a track.
121
link
ribbon
rows of embossed
characters
AAA
paper B B B BA A A
BB
CCCCC
DDDD C
EEEED
E
FF
GG
HH
rotating cylinder
hammers (not all shown)
barrel printer
paper
ribbon
PQR S embossed
AB CD E characters
band printer
Line printers
122
local
linker See link editor. the list components without moving the
data items; only the links are modified. For
link loader See link editor. example, a new item can be inserted in its
correct place by having the preceding item
linkrot /link-rot/ Informal The condi- link to the new item and the new item link
tion in which HYPERTEXT leads nowhere to the succeeding item.
because Web sites have removed or reorga-
nized their Web pages and the links are list box A type of CONTROL on a GUI
outdated. FORM. The user selects its content from a
list of available options. Compare text
Linux /lin-ŭks/ A freely distributable op- box; see also combo box.
erating system that runs on several hard-
ware platforms including Intel and listing An output of a computer system,
Motorola microprocessors, and thus on often diagnostic, that can be printed. The
PCs and Macintoshes. Linux is a form of word is usually qualified in some way to
UNIX written by Linus Torvalds. Its use has give a more precise idea of the contents, as
increased rapidly in recent years, especially in program listing or error listing. A listing
for SERVERS. may be created in a file rather than printed,
which allows it to be examined on line and
liquid crystal display See LCD. perhaps analyzed by program.
123
local area network
other words, it is either part of the user’s local variable See scope.
computer or directly attached to it. See re-
mote. location (storage location) In general,
2. In programming, describing a variable any place in which information can be
that is only defined for one part (subrou- stored in a computer – in MAIN STORE or
tine, procedure, or function) of the pro- BACKING STORE. A storage location is an
gram and does not have GLOBAL scope. area within a computer memory capable of
storing a single unit of information in bi-
local area network (LAN) A communi- nary form. Each location can be identified
cation system linking a number of comput- by an ADDRESS, allowing an item of infor-
ers within a defined and small locality. This mation to be stored there or retrieved from
locality may be, say, an office building, an there. In most cases it is possible to change
industrial site, or a university. Linking the the stored value during execution of the
computers together allows resources, such program. The words location and address
as HARD DISKS, PRINTERS, or an Internet con- are often used as synonyms, as in ordinary
nection, to be shared between them, allows speech.
FILES to be shared, gives all users access Main store is divided into either WORDS
to specialized application SERVERS (e.g. or BYTES: in some computers each location
database servers that manage shared data- holds a word; in other machines each loca-
bases), allows an INTRANET to be estab- tion holds a byte. The memory is said to be
lished, and permits messages to be sent word-addressable or byte-addressable. A
between the computers by E-MAIL. word is now generally 32 bits and is usu-
The computers in the network are di- ally large enough to contain a MACHINE IN-
rectly connected to the transmission STRUCTION. A byte is always 8 bits and thus
medium by electrical INTERFACES; normally is the storage space required to hold a small
the medium is in the form of electric cables, number or a common CHARACTER.
optical fibers, or wireless. There are vari- Locations on disk and tape hold a num-
ous ways in which the interconnections be- ber of bytes and are addressed by TRACK
tween machines can be organized (see and SECTOR or by BLOCK, respectively.
network). The system is usually controlled
overall by a server that validates all users lock 1. A mechanical device on some re-
when they LOG ON and ensures they do not movable storage media that prevents the
access resources or perform operations for- user writing to the media.
bidden to them. Often, and especially on 2. A software security feature that ensures
small networks, this function is combined it cannot be used unless unlocked with a
with that of FILE and PRINT SERVERS on a key or DONGLE.
single machine, known as the server.
Local area networks generally provide locked up The condition of a system or
high-speed data communications and have an application that seems unable to re-
very low ERROR RATES. It is possible to in- spond to the user. See also hang.
terconnect two or more local networks,
and to connect a local network to a larger log /log/ A record containing details of
network, i.e. to a WIDE AREA NETWORK. past activities or actions on a computer sys-
tem, e.g. details of problems found (or not)
local bus A PC architecture design that and actions taken when scanning a disk.
gives high performance by allowing expan-
sion boards to communicate directly with logic /loj-ik/ See binary logic; digital
the microprocessor. logic.
124
logic circuit
A A
OR
NOT B +A
B
(B + A) . (A + B)
AND
NOT A
A +B
OR
B B
inputs _ _ _ _ output_
A B A B A+B A+B (B + A).(A + B)
0 0 1 1 1 1 1
0 1 1 0 0 1 0
1 0 0 1 1 0 0
1 1 0 0 1 1 1
125
logic diagram
tion of the circuit. In the circuit and truth TRUTH TABLES are used to describe the
table shown, +, ., and ¯ are the OR, AND, output of a logic gate for all possible com-
and NOT OPERATORS. The expression de- binations of inputs. A simple gate with 2
rived for the circuit output indicates that it inputs has 22, i.e. 4, possible combinations
is an EQUIVALENCE GATE. of inputs and hence 4 possible outputs;
with 3 or with 4 inputs there are 23 or 24,
logic diagram A diagram in which i.e. 8 or 16, possible outputs. Truth tables
LOGIC SYMBOLS are used to show how the are given at individual entries for gates in
inputs and outputs of components are con- this dictionary. See also logic symbols;
nected together in a LOGIC CIRCUIT. It thus logic diagram.
displays the design of the circuit but gives
no indication of the electronics involved. logic instruction A MACHINE INSTRUC-
TION specifying a LOGIC OPERATION and the
logic element A LOGIC GATE or combi- OPERAND or operands on which the logic
nation of gates. operation is to be performed. An example,
expressed in ASSEMBLY LANGUAGE, might be
logic error An error in the logic of a OR A,B
program that causes incorrect results but This is an instruction to perform the OR
does not stop the program running. operation between the contents of registers
A and B, placing the result in A. See also
logic gate A device, almost always elec- arithmetic instruction.
tronic, that is used to control the flow of
signals in a computer. It does this by per- logic network See logic circuit.
forming LOGIC OPERATIONS on its inputs.
Normally there are between two and eight
logic operation An operation per-
formed on quantities (OPERANDS) that can
inputs to a gate and one output. An input
be assigned a truth value; this value is ei-
signal is either at a high level or at a low
ther true or false. The result of the opera-
level, the level changing with time. (It is ac-
tion is in accordance with the rules of
tually the voltage level of the signal that
BOOLEAN ALGEBRA, and again has a value
changes in an electronic gate.) The signal
true or false. The operands may be state-
thus has a binary nature; the two values
ments or formulae, such as
can be represented as 1 (high) and 0 (low).
It is July
As the input signals switch between their the wire is less than 10 cm long
high and low levels, the output level x + 2 = 16
changes, the value depending on the logic The operands of a logic operation can be
operation performed by the gate. For ex- represented by letters. For example, in
ample, one type of gate has a high output P AND Q, P OR Q, NOT (P OR Q)
only when all its inputs are high at the P and Q are the operands and AND, OR,
same time. In any logic gate the output and NOT are LOGIC OPERATORS that repre-
changes extremely rapidly in response to a sent three commonly used logic operations,
change in input. namely the AND, OR, and NOT OPERATIONS.
There are several simple gates that each Logic operations can be described by
perform a basic logic operation. For exam- means of TRUTH TABLES. LOGIC GATES are
ple, the AND GATE and OR GATE perform the constructed so as to perform logic opera-
AND and OR OPERATIONS, respectively, on tions on their input signals.
their inputs. The NOT GATE performs the
NOT OPERATION on its single input. The logic operator A symbol representing a
other simple logic gates are the NAND GATE, LOGIC OPERATION such as the AND or OR
NOR GATE, NONEQUIVALENCE GATE, and operation. A logic operator has logical
EQUIVALENCE GATE. Selected gates can be quantities as its OPERANDS and delivers a
connected together to form a LOGIC CIR- logical result. For example, the expression
CUIT. ‘I have no money’ and ‘it is Tuesday’
126
long filename
is true if both the operands are true, other- incorporates the concepts of PROCEDURES
wise it is false. In programming languages and LISTS, and helps children to think algo-
logic operators are either spelled out in full rithmically. Logo includes TURTLE GRAPH-
or various symbols are used; some of these ICS: the turtle is a simple pen-plotter that
are shown in the table. can be steered around a large piece of
An example of how logic operators are paper on the floor, under program control,
used, in Basic, is as follows: allowing the children to create complex
IF (A=3) AND (X<10) THEN A=0 patterns. In some microcomputer imple-
See also order of precedence. mentations the turtle pictures may be
drawn on the screen. This is cheaper but
perhaps not so effective.
LOGIC OPERATORS
Operation Operators log off (log out; sign off) To inform a
computer system or NETWORK from a TER-
AND AND .AND. & .
MINAL or WORKSTATION that one’s work is
OR OR .OR. ! | +
– finished for the moment. This log off or log
NOT NOT .NOT. ¬ ! ′
out process is required in any MULTIACCESS
SYSTEM for security and accounting pur-
poses. After a user has logged off, the com-
puter system or network will not be
available on that terminal or workstation
logic symbols The symbols used in a until someone has logged on there. See also
LOGIC DIAGRAM to represent the various log on.
LOGIC GATES and the other possible compo-
nents in a LOGIC CIRCUIT. One set that is log on (log in; sign on) To identify one-
used to represent logic gates is shown in the self to a computer system or NETWORK
diagram: the logic operation performed by when wishing to enter the system from a
the gate is printed inside a circle, and ar- TERMINAL or WORKSTATION. This log on or
rowed lines labeled by letters represent the log in process is required in any MULTIAC-
inputs and output. In another set (widely CESS SYSTEM. The log on sequence involves
used in technical literature) the shape of the first saying who one is by means of one’s
symbol is used to indicate its function. name or account number and then entering
one’s PASSWORD. In some cases where, say,
log in See log on. security is very strict, there may be more to
do. See also log off.
Logo /loh-goh/ A PROGRAMMING LAN-
GUAGE developed in the 1960s by Seymour log out See log off.
Papert for use in teaching young children.
Logo is a simple but powerful language. It long filename In Windows, a FILENAME
P (P AND Q) P (P OR Q)
AND R OR R
Q Q
127
look-up table
that can be up to 255 characters long and tems, a test signal sent to a network desti-
include spaces. The term is used to contrast nation that is then returned as received by
such names with the short filename (or 8.3 the sender. Loopback is used in fault diag-
filename) system used by MS-DOS, where nosis. On the Internet a comparable utility
filenames were limited to 8 characters plus is called PING.
a 3-character EXTENSION and spaces were
not allowed. Windows still maintains short lossless compression /loss-liss/ Any data
filenames for BACKWARD COMPATIBILITY. compression accomplished by a technique
that reduces file size by eliminating redun-
look-up table A TABLE used in table dancies in the data. Data compressed by
look-up. these methods can be restored exactly to its
original value. Lossless compression is
loop A statement or group of statements used for data and programs. Compare
whose execution is repeated either a speci- lossy compression. See also PKZIP.
fied number of times or while or until some
condition is met. Both types of loop are it- lossy compression /loss-ee/ Any data
erative CONTROL STRUCTURES, and appear compression accomplished by a technique
in various forms in almost every program- that reduces file size by discarding some
ming language. data forever, such as small variations in
The for loop appears in several lan- color, whose loss won’t be noticed. Lossy
guages and permits a specified number of files cannot be decompressed to their orig-
repetitions to occur. An example, in Pascal, inal state. Only certain types of data, such
is
as graphics, audio, and video, can tolerate
for i := 1 to 10 do statement
lossy compression. JPEG and MPEG, for ex-
This causes the statement to be executed 10
ample, use lossy compression techniques.
times. In Basic this would be written
Compare lossless compression.
FOR I = 1 TO 10
statement
lost cluster A CLUSTER on a disk storage
NEXT I
device that the FILE SYSTEM thinks is in use
There are two main forms of the second
but is not part of any file. Lost clusters are
type of loop, i.e. the conditional-type loop.
In one form the condition is given at the be- usually caused when a system or program
ginning of the loop and execution contin- is closed down without going through the
ues while this condition is satisfied. This is proper procedures, for example in the case
commonly referred to as a while loop or a of a power cut. Some file systems (e.g. FAT)
while do loop. For example, Pascal has are more vulnerable to lost clusters than
while condition do statement others (e.g. NTFS).
The statement or statements will not be ex-
ecuted at all if the condition does not hold Lotus 1-2-3 /loh-tŭs/ Trademark An
the first time round. The other form is the electronic spreadsheet application intro-
repeat until loop, which is executed at least duced in 1982. It is widely credited with es-
once since the condition is given at the end tablishing the popularity of the IBM PC.
of the loop, causing it to terminate when
necessary. For example, Pascal has Lotus Notes A GROUPWARE application
repeat statement until condition developed by Lotus, now part of IBM.
It is possible for one or more loops to be Notes was one of the first applications to
included within a loop; the incorporated support a distributed database of docu-
loop or loops are referred to as nested ments that could be accessed by users
loop(s). The nesting must be such that each across a LAN or WAN. Its sophisticated
inner loop is entirely contained within the REPLICATION features enable users to work
outer loop or loops. with local copies of documents and have
their modifications propagated throughout
loopback /loop-bak/ In telephone sys- the entire Notes network. With the grow-
128
LZW compression
129
L
L1 cache (primary cache) A small cache electric charge. This charge pattern is
memory in microprocessors. Compare L2 coated with varying amounts of toner (dry
cache. particles of pigment) according to the
strength of the charge. The toner is rolled
L2 cache (secondary cache) Cache onto the paper and fused in place by heat,
memory that was originally external to the producing a permanent image. Laser print-
microprocessor. Modern micr-oprocessors ers work quietly and at high speed, gener-
now include L2 caches in their architec- ating high-quality print. They readily
tures. Compare L1 cache. produce graphs and diagrams and a wide
variety of typestyles. Color laser printers
label An IDENTIFIER associated with a are also available.
STATEMENT in a program and used in other
parts of the program to refer to that state- latency 1. In general, the period of time
ment. that one component in a system is waiting
for another component. For example, in
LAN /lan/ See local area network. accessing data on a disk, latency is the time
it takes to position the sector under the
landscape The arrangement of printed read/write head. See access time.
matter on a page so that the top of the page 2. In networking, the amount of time it
is on the wider side. Compare portrait. takes a packet to travel from source to des-
tination. Together latency and BANDWIDTH
language See programming language. determine the speed and capacity of the
network.
laptop /lap-top/ A small portable per-
sonal computer. Laptops can be run from LaTeX /lay-tekh/ A typesetting system
battery packs or from the mains. Flat panel based on the TeX programming language
display screens are used and a TRACKERBALL for typesetting. It provides for higher level
is often embedded in the keyboard. In macros than TeX, making it easier to for-
other respects the memory capacity, hard mat documents, but in doing so it loses
drives, and CPUs are the same as those some flexibility.
used in a personal computer.
launch To start a program.
large-scale integration See integrated
circuit. layer A level in a program or system that
performs a particular function or functions
laser printer A type of PRINTER in which and has a defined interface for passing re-
a page-sized image is produced in the form sults to/from the layers next to it. OSI is an
of a pattern of very fine dots by the action example of a layered system.
of a laser beam. The principle is similar to
that used in many office copiers. The laser layout In word processing and desktop
beam writes the image on the surface of a publishing, the arrangement of the text and
drum or band in the form of a pattern of graphics on a page.
118
M
M The symbol used for MEGA-, or some- item of data consists of a particular se-
times for megabyte. quence of bits.
It is very slow and tedious for a person
Mac The popular abbreviation for the to write a program in machine code, and
Apple MACINTOSH computer. very difficult to read and understand such
a program. Instead a programmer nor-
MacBinary /mak-bÿ-nă-ree, -nair-ee/ A mally uses a HIGH-LEVEL LANGUAGE (such as
file transfer protocol that preserves Mac- Basic or C++), which is translated into ma-
intosh file formats when stored on non- chine code by a program – either a COM-
Macintosh computers. PILER or an INTERPRETER. A large number of
machine code instructions are required for
machine An actual computer, or some- each high-level language statement. An-
times a model of a computer used in theo- other less easy alternative is for the pro-
retical studies. grammer to use ASSEMBLY LANGUAGE,
which gives a ‘readable’ and hence more
machine address A unique number that convenient representation of machine
specifies a unique storage LOCATION (or in code; a program written in assembly lan-
some computers a particular REGISTER or guage is converted into machine code by a
INPUT/OUTPUT DEVICE). A machine address
program known as an ASSEMBLER.
contains a fixed number of bits, n, and
hence can identify 2n different locations.
machine instruction (computer instruc-
tion) An instruction that can be recog-
The number of distinct locations to which
nized by the processing unit of a computer.
a machine address can refer is known as
It is written in the MACHINE CODE designed
the address space. A machine address is de-
for that particular computer, and can be in-
termined by the computer hardware and
terpreted and executed directly by the CON-
when specified in a MACHINE INSTRUCTION TROL UNIT and the ARITHMETIC AND LOGIC
can be used directly by the hardware. A UNIT of that computer. A machine instruc-
machine address is thus an ABSOLUTE AD- tion consists of a statement of the OPERA-
DRESS. See also address.
TION to be performed, and some method of
specifying the object upon which the oper-
machine code The code used to repre- ation is to be performed (i.e. the OPERAND)
sent the instructions in the INSTRUCTION SET together with an indication of where the re-
of a particular computer. It is therefore sult is to go. There may be more than one
specific to a particular computer. It is the operand involved.
form in which programs must be recorded The portion of the machine instruction
on, say, magnetic disk and entered into in which the operation is specified is
main store, and is the form subsequently known as the operation part. The operand
used for processing. Machine code consists or operands are generally specified in the
of a sequence of MACHINE INSTRUCTIONS. In instruction by stating their actual AD-
order for the computer hardware to handle DRESSES in main store, or by some other
it, machine code is expressed in binary scheme such as INDIRECT ADDRESSING, IN-
form: every operation, every address, every DEXED ADDRESSING, or RELATIVE ADDRESS-
130
magnetic disk
ING; in IMMEDIATE ADDRESSING the operand to record a series of keystrokes and com-
itself is stated explicitly in the instruction. mands, which can be given a name or as-
This information about the operand or signed to a combination of keys. The
operands is given in the address part or macro can then be used over and over
parts of the instruction. again to repeat the recorded actions.
In addition to the operation and ad- In programming languages or applica-
dress parts, special bit positions are re- tions that provide a macro facility, which
quired in the machine instruction to may include a macro language, a set of in-
indicate whether a particular addressing structions can be given a name. When the
mode is to be used. There are thus bit posi- program is compiled or assembled, when-
tions to specify indirect addressing, the use ever the macro name is used the set of in-
of index registers for indexed addressing, structions is substituted by a process called
or the use of base registers in relative ad- macro expansion.
dressing. Additional bits may be used for
other purposes. macro virus A virus written in the
The layout of a machine instruction, macro language associated with an appli-
showing its constituent parts, is known as cation. The document file that uses the
the instruction format. It is laid down by macro carries the virus which is usually
the computer manufacturer, and specifies triggered when the document is opened or
the number of bits allocated to each part – sometimes when other actions are per-
operation part, address part or parts, etc. – formed. A macro virus is often spread as an
and the relative positions of each of these e-mail virus. A well-known example was
parts. the Melissa virus of 1999.
131
magnetic drum
When the head is to write data on the disk purposes; they are found for example on
it receives a coded electrical signal. This credit cards and swipe cards.
signal causes magnetization of tiny regions
in the magnetic film immediately below the magnetic tape A cheap robust storage
head. The regions are magnetized in either medium consisting of a flexible plastic tape
of two directions; magnetization in one di- with a magnetic coating on one side. For-
rection represents a binary 1 and in the merly held on large reels, magnetic tape is
other direction a binary 0. A letter, digit, or now found mainly in small CASSETTES. It is
other character can therefore be encoded essentially the same as that found in do-
as a particular sequence of magnetized mestic video cassettes. SERIAL ACCESS must
spots (see binary representation). This pat- be used for storage and retrieval from tape.
tern of spots is written on the disk in dif- When magnetic tape was first used in
ferent SECTORS of the tracks. The sectors computer systems in the 1950s it was the
may be in a continuous sequence or scat- only form of high-speed BACKING STORE. It
tered in different tracks. When the head is is still a cheap way of storing very large
to read data from a specified track and sec- amounts of information, especially when
tor, it senses the magnetized patterns there the information is accessed rarely, as in FILE
ARCHIVES or BACKUP copies of information
and converts them into a corresponding
electrical signal. Items of data can be held on MAGNETIC DISK.
deleted from a disk and new data added – Items of data are recorded on and re-
trieved from magnetic tape by read and
a magnetic disk is a reusable resource.
write heads operating in a TAPE DRIVE. The
write head receives a coded electrical sig-
magnetic drum An obsolete type of
nal. This signal causes magnetization of
BACKING STORE that has now been displaced
tiny regions in the magnetic film immedi-
by the MAGNETIC DISK. It was a cylindrical
ately below the head. Each spot represents
storage device whose curved outer surface
one BIT: a spot magnetized in one direction
was coated with a magnetic material in
represents a binary 0 while a spot magne-
which data could be recorded.
tized in the opposite direction represents a
binary 1. A letter, digit, or other character
magnetic ink character recognition is encoded as a pattern of magnetized
See MICR. spots.
The data is written on the tape in
magnetic media Material on which BLOCKS; a sequence of blocks makes up a
data can be recorded in the form of a mag- FILE. When data is to be read from a block,
netic pattern on the surface of a magnetiz- the read head senses the magnetized pat-
able film. The data is represented in binary terns there and converts them into a corre-
form by the direction in which points on sponding electrical signal. Blocks of data
the surface are magnetized. Magnetic can be erased from tape and new data
media form the recording basis of MAG- added – magnetic tape is a reusable re-
NETIC DISKS, MAGNETIC TAPE, and MAGNETIC source.
STRIPES. See also digital audio tape.
magnetic storage Any noninternal data magnetic tape unit (MTU) See tape
storage medium that uses magnetic proper- unit.
ties to effect the representation of data, for
example disk. magneto-optic recording A technol-
ogy used with optical disks in which small
magnetic stripe (magnetic strip) A strip areas of the surface are heated by a laser so
of magnetic film on a plastic card on which that they can be easily magnetized and the
machine-readable data is encoded. The regions of magnetization can be confined
most common application of magnetic and precisely located. A lower power laser
stripes is for identification and security is used to read the data from the disk. The
132
MAN
disks are erasable and have a capacity sim- and data awaiting processing, being
ilar to a CD-ROM. processed, or resulting from processing.
Programs can only be executed when
mailbox A network user’s area for stor- they are in main store. Programs and asso-
ing incoming e-mail messages. Mail sys- ciated data are not, however, retained per-
tems usually allow mail to be read, manently in main store. They are kept in
archived, copied to a file, deleted, printed, backing store until required by the proces-
or forwarded to another user. sor. The backing store in a particular com-
puter (large or small) has a larger storage
mailing list 1. A system, usually auto- CAPACITY and is also less expensive (in
mated, that allows users to send e-mail to terms of cost per bit stored) than the main
one address, whereon their message is store in that computer.
copied and sent to all of the other sub- Main store consists of SEMICONDUCTOR
scribers to the list. MEMORY. There is RANDOM ACCESS to data
2. A named list of e-mail addresses. Enter- and the average ACCESS TIME is less than a
ing the name of the list in the address field millionth of a second, typical values being
of an e-mail causes the e-mail to be sent to a few nanoseconds. Semiconductor mem-
all addresses on the list. ory is very reliable and is highly miniatur-
ized.
mail merge A technique used in word- A single item of data is stored in a par-
processing applications for the merging of ticular LOCATION in main store. In most
names and addresses and sometimes other computers each location holds a fixed
facts from a list (perhaps derived from a number of BITS; this number is the same for
database) into a form letter or other basic all locations in the main store of a particu-
document for mass mailing purposes. lar computer, and is usually 8, 16, or 32
bits. Each location can be identified by its
main body The program code that is ex- ADDRESS.
ecuted when the program is started and
that then calls subroutines and functions as maintenance 1. Any activity, including
necessary. tests, measurements, adjustments, replace-
ments, and repairs, that aims to prevent
mainframe /mayn-fraym/ Any large the occurrence of faults in hardware or
general-purpose computer system, as op- corrects an existing hardware fault. See
posed to the smaller MINICOMPUTERS and preventative maintenance; corrective main-
MICROCOMPUTERS. Mainframes are typi- tenance.
cally expensive computers used by large or- 2. (program maintenance) See program.
ganizations and are capable of supporting
a large number of users. malfunction The occurrence of a FAULT,
usually a hardware fault.
main memory See main store.
malware (malicious software) An over-
main store (main storage; main memory; all term denoting any software intended to
RAM; immediate access store, IAS; pri- damage a computer system or subvert its
mary store) The storage closely associ- operation. Malware includes ADWARE, SPY-
ated with the PROCESSOR (or processors) of WARE, TROJANS, VIRUSES, and WORMS.
a computer and from which program in-
structions and data can be retrieved ex- MAN (metropolitan area network) Usu-
tremely rapidly and copied directly into the ally characterized by very high speed con-
processor registers, and in which the re- nections using fiber optic cables or other
sulting data is stored prior to being trans- digital media, a network that can consist of
ferred to BACKING STORE or OUTPUT DEVICES. several LANS as well as satellite and mi-
Main store thus stores instructions waiting crowave relay stations. It is smaller, but
to be obeyed or currently being obeyed, generally speaking faster, than a WAN.
133
Manchester Mark I
134
mean
mark sensing An obsolete process of capacity disk drives used primarily as FILE-
data entry to a computer in which a ma- STORE. See also capacity.
chine sensed pencil marks made by a per-
son in his or her choice of indicated master file A DATA FILE that is subject to
positions on a document. The pencil marks frequent requests for data and frequent up-
were electrically conductive and could thus dating of the values stored. It can therefore
be sensed by electrical methods. Mark be considered an authoritative source of in-
sensing has been displaced by optical mark formation. A master file can be updated by
reading, i.e. OMR, which employs more re- being completely rewritten by the system.
liable (photoelectric) sensing methods. The version prior to a particular update is
called the father file, while the one before
markup language A set of formatting that is the grandfather file. The father file,
codes and indexing and linking informa- and often the grandfather file, are retained
tion included in a text file, indicating how as BACKUP copies of the master file. See also
the file should look when it is printed or file updating.
displayed or describing the document’s
logical structure. The markup indicators math coprocessor A coprocessor de-
are often called tags. Some markup lan- signed to enable quick processing of float-
guages are specific to an application, but ing-point numbers. This speeds up the
others are designed to enable documents to processing of math and graphs consider-
be platform independent, offering great ably if the application is designed for its
portability between applications and ma- use. Formerly important in microcomput-
chines. See also HTML; SGML; XML. ers, the functionality of math coprocessors
is now built into most microprocessors.
mask 1. A pattern of bits that is used to
modify or identify part of a byte, word, matrix A two-dimensional ARRAY.
etc., with the same number of bits. An op-
eration, such as an AND OPERATION or an matrix printer See printer; dot matrix
OR OPERATION, is performed on corre- printer; ink-jet printer.
sponding pairs of bits in the two patterns;
this operation is called masking. For exam- maximize To make a window take up
ple, with an AND operation a 1 in the the full screen.
mask will retain the value of the corre-
sponding bit in the input byte, word, etc., Mb A symbol for megabit.
for subsequent processing, while a 0 in the
mask will set or keep the corresponding bit MB A symbol for megabyte.
as a 0. Thus an AND on a byte containing
an 8-bit number with a mask of 00000011 MBR See memory buffer register.
will clear all bits except the two least sig-
nificant, i.e. it will return a number be- MDR See memory data register.
tween 0 and 3 that is the modulo 4 of the
input. Masking is often used to store more mean In general, a number representing
than one piece of information in a single the average or typical value of a group of
byte or word. For example, the 8 bits of a numerical values, such as measurements of
byte can be made to store 5 TRUTH VALUES a quantity. If the values are represented by
in one bit each and also a 3-bit number, i.e. x1, x2, x3, … xn
between 0 and 7. so that there are_ n values in all, then their
2. A chemical shield used to determine the mean (symbol:
_ x ) is given by the formula
pattern of interconnections in an INTE- x = (x1 + x2 + x3 + … + xn)/n
GRATED CIRCUIT. The mean of 8 numbers is thus the sum of
those numbers divided by 8. For example,
mass storage device Any backing stor- with the numbers
age device that contains one or more high- 2, 5, 3, 7, 5, 1, 3, 2
135
media
the mean is 3.5. The amount of variation for subsequent use by a computer. The
within the group is indicated by its STAN- word is synonymous with store and stor-
DARD DEVIATION. age but is most frequently used in the terms
random-access memory (RAM) and read-
media (singular: medium) Materials only memory (ROM), i.e. semiconductor
used in the entry or storage of data in a memory used for MAIN STORE. The word
computer, or in the recording of results ‘memory’ by itself is generally used to
from a computer. The latter group are mean main store. See also storage device;
known as output media, the former as media.
input media and storage media. Input and
storage media include MAGNETIC MEDIA, memory address A number represent-
CDs, and DVDs. Data is represented in ing a location in MAIN STORE.
coded form on these media. Output media
include the sheets of paper or film or the memory address register (MAR) A
CONTINUOUS STATIONERY used in PRINTERS, REGISTER (i.e. a temporary location) in the
the paper or film used in PLOTTERS, and the central processor, used for holding the ad-
microfiche used in COM. dress of the next location in main store to
be accessed. Whenever the main store is to
mega- /meg-ă-/ A prefix indicating a be used, the required address must be sent
multiple of a million (i.e. 1 000 000 or to the MAR. The corresponding storage lo-
106) or, loosely, a multiple of 220 (i.e. cation can then be selected automatically,
1 048 576). In science and technology dec- and data read from there or stored there.
imal notation is usually used and powers of See also memory data register.
10 are thus encountered – 103, 106, etc.
The symbol M is used for mega-, as in MV memory buffer register (MBR) See
for megavolt. In communications mega- memory data register.
(as in megabaud) means a million. Binary
notation is generally used in computing memory card (memory board) See ex-
and so the power of 2 nearest to 106 has as- pansion card.
sumed the meaning of mega-.
The prefix is most frequently encoun- memory data register (MDR) A REGIS-
tered in computing in the context of stor- TER (i.e. a temporary location) used for
age CAPACITY. With magnetic disks, holding all instructions and data items as
magnetic tape, and main store, the capacity they are transferred between main store
is normally reckoned in terms of the num- and central processor. It thus acts as a
ber of BYTES that can be stored, and the BUFFER, and is therefore also called memory
word megabyte (MB) is used to mean 220 buffer register (MBR). See also memory ad-
bytes; megabyte is usually abbreviated to dress register.
M byte or just to M, thus a 256 M byte
flash drive can hold 268 435 456 bytes. memory mapping 1. A technique al-
Mega- is part of a sequence lowing a processor to access more memory
kilo-, mega-, giga-, tera-, peta-, … than it is ordinarily capable of addressing.
of increasing powers of 103 (or 210). Different banks of memory are electroni-
cally switched so that they appear to the
membrane keyboard A keyboard in processor to have the same address.
which the keys are covered by a transpar- 2. In microcomputers, a way of making
ent plastic shell. It is difficult to type accu- input/output devices appear to be memory
rately and quickly on such a keyboard but locations so that instructions that would
it has the advantage that the components normally store or retrieve information
can be kept very clean. from main store perform output or input to
an external device.
memory A device or medium in which
data or program instructions can be held memory-resident Describing programs
136
MICR
that are permanently in memory. The op- metafile /met-ă-fÿl/ 1. A file that con-
erating system is not permitted to swap tains information about another file, e.g. a
such programs out to a storage device. format for graphics files. The commonest
is the Windows metafile (which has the file
memory size The total number of bytes extension wmf).
of MAIN STORE available on a computer.
metalanguage /met-ă-lang-gwij/ A lan-
menu A list of options, usually displayed guage used to describe other languages in
on a screen, from which a choice can be the theory of computer languages. SGML
made by the user. The list may be displayed and XML are metalanguages.
with a different letter or number in each
option underlined or shown separately op- meta tag A tag used to describe the con-
posite each option, and the selection made tents of a Web page. Meta tags do not af-
by typing in the appropriate letter or num- fect how the page is displayed but provide
ber using a keyboard. The selection can information such as authorship, what the
also be made by positioning the cursor on page is about, and keywords that represent
the selected option using, for instance, a the page’s content. Many search engines
mouse or cursor keys on a keyboard, or by use this information when building their
touching a selected area of a touch-sensi- indexes.
tive screen. The selected item will often be
highlighted in some way; REVERSE VIDEO or metatext /met-ă-tekst/ Text contained in
blinking are commonly used. a document that does not appear on the
screen but is included to trick Web search
Any program that obtains input from a
indexing software into giving the site a
user by means of a menu is said to be
higher relevance rating in a search. For ex-
menu-driven.
ample, by repeating the word aeroplane
100 times, the document is more likely to
menu bar A horizontal menu that ap-
turn up in a search that includes aeroplane
pears at the top of a window. Usually each
as a key word. Search engines are now
option in the menu is associated with a
aware of this trick and compensate for it.
PULL-DOWN MENU.
137
micro
an appropriate shape and size and have a Microcomputers were originally rather
high print quality; one standardized and limited in what they could do. With the
widely used set of type is rectangular in ap- emergence of more powerful, more sophis-
pearance with thickened vertical sections. ticated, but cheaper systems they are now
The most common application of MICR is regarded as serious problem-solving tools
in the handling of cheques by banks; in many fields, including business, science,
cheque number, branch number, and ac- and engineering, and are a standard feature
count number are printed in magnetic ink in many homes. The capability of the sys-
characters along the bottom of the cheque, tem depends not only on the characteristics
the amount of the cheque being added of the microprocessor employed but also
manually. See also OCR; OMR. on the amount of storage provided, the
types of peripheral devices that can be
micro /mÿ-kroh/ See microcomputer. used, the possibility of expanding the sys-
tem with add-on peripherals and addi-
micro- 1. a prefix to a unit, indicating a tional storage, and other possible options.
submultiple of one millionth (i.e. 10–6) of
that unit, as in micrometer, microsecond, microelectronics /mÿ-kroh-i-lek-tron-iks/
or microvolt. The Greek letter µ (mu) is Electronics involving the use of very small
used as the symbol for micro-; micrometer, electronic parts to build circuits and sys-
microsecond, and microvolt are thus often tems.
written µm, µs, µV. Micro- is part of a se-
quence milli-, micro-, nano-, pico-, … microfiche, microfilm /mÿ-kroh-feesh,
of decreasing powers of 1000. mÿ-kroh-film/ See COM.
2. A prefix indicating smallness or compar-
ative smallness, or something concerned microinstruction /mÿ-kroh-in-struk-shŏn/
with small objects, quantities, etc. See microprogram.
138
milli-
operation, etc.). They are used not just in addition to ROM. Use of RAM allows
microcomputers but are embedded in a modifications to be made and specialized
wide variety of devices. microinstructions to be introduced.
The first microprocessor – the Intel
4004, produced by the US company Intel – Microsoft /mÿ-kroh-soft/ Trademark
appeared in 1971. It had a 4-bit word Microsoft Corporation is the largest pro-
length, 4.5 kilobytes of memory, and 45 in- ducer of PC operating systems and applica-
structions. Its 8-bit counterpart, the Intel tions, founded in 1975 by Paul Allen and
8008, was introduced in 1974 and its im- Bill Gates. See also Windows.
proved derivative, the Zilog Z80, in 1976.
The number of microprocessors, and their MIDI /mid-ee/ (musical instrument device
sophistication, has increased rapidly – interface) A standard INTERFACE that al-
alongside developments in the fabrication lows electronic instruments such as musi-
technology of integrated circuits. The Intel cal keyboards and synthesizers to be
PENTIUM 4 series of microprocessors, intro- attached to and used on a computer. The
duced in 2000 and now the usual chip in MIDI standard is supported by most syn-
computers designed for SOHO use, has a 32- thesizers and sounds created by one syn-
bit word length and can address at least 4 thesizer can be played and manipulated by
gigabytes of memory; other microproces- another. Applications are available for
sors, such as the ITANIUM, are more power- composing and editing music on comput-
ful still. ers to MIDI standards.
139
MIMD
volt are thus often written mm, ms, mV. bar button with the application’s name on
Milli- is part of a sequence it.
milli-, micro-, nano-, pico-, …
of decreasing powers of 1000. minitower /min-ee-tow-er/ A microcom-
puter in which the power supply, mother-
MIMD (multiple instruction multiple data board, and mass storage devices are
stream processing) A type of computer stacked on top of one another in a small
architecture used in parallel processing in vertical cabinet.
which CPUs are used to process data and
fetch instructions independently. See CPU; MIP mapping A technique used in 3D
parallel processing. graphics for correcting pixelation and
speeding up rendering times. Different pre-
MIME /mÿm/ (multipurpose Internet mail built texture maps are applied when the
extensions) A method for attaching files object on the screen is close up or far away.
to an e-mail that allows 8-bit binary data MIP is an abbreviation for the Latin mul-
to be included with the e-mail rather than tum in parvo (many in few).
encoding the binary data as 7-bit ASCII text.
An e-mail program is said to be MIME MIPS Millions of instructions per sec-
compliant if it can both send and receive ond; a measure of processor speed.
files using the MIME standard.
mirror An arrangement where data is
mini /min-ee/ See minicomputer. written to more than one hard disk simul-
taneously, so that if one disk fails, the sys-
minicomputer /min-ee-kŏm-pyoo-ter/ tem can instantly switch to the other disk
(mini) Loosely, a medium-size computer, without any loss of data or service. Mir-
i.e. one that is usually considered to be rored disks are a type of RAID.
smaller and less capable in terms of perfor-
mance than a contemporary MAINFRAME mirror site A Web or FTP site that
computer, and hence cheaper. A modern maintains exact copies of material origi-
minicomputer might well operate faster nated at another location, usually in order
than many of the mainframes used in the to provide more widespread access to the
l970s. Again, it might be outperformed by information or to improve the perfor-
a sophisticated MICROCOMPUTER. Mini- mance of a heavily used site.
computers came on the scene before micro-
computers and the boundary between the mod /mod/ Modulo. See integer arith-
two is no longer clear. metic.
Minicomputers are often to be found in
laboratories built into complex experi- model A mathematical or graphical rep-
ments, or controlling industrial processes. resentation of something. Mathematical
They are also used for small multiuser sys- models can usually be changed so that
tems, although networks of PERSONAL COM- ‘What if’ analyses can be carried out. See
PUTERS attached to a server are now also simulation.
commonly used for this purpose. Main-
frame computers often have a number of modem /moh-dem/ A device that con-
minicomputers to perform such tasks as verts a DIGITAL SIGNAL into an ANALOG SIG-
input/output or communications, thus NAL suitable for transmission over an
freeing the power of the mainframe CPU analog communication channel, and that
for more demanding tasks. converts an incoming analog signal back
into a digital signal. It does this by MODU-
minimize To shrink a window down so LATION of the digital signal and demodula-
that it is displayed only on the TASK BAR. A tion of the incoming analog signal. (The
minimized application remains active and word ‘modem’ is a contraction of ‘modula-
can be brought back by clicking on the task tor/demodulator’.) A pair of modems can
140
monadic operator
thus be used to connect two units – two ing ANALOG SIGNAL. The other signal may
computers, say, or a computer and a termi- for instance be a DIGITAL SIGNAL, e.g. a
nal – across a telephone line; the streams of stream of bits. The pattern conveyed by
bits from each unit are digital signals that this signal is superimposed on the carrier
are converted by the modems into continu- by varying one of the properties of the car-
ously varying signals suitable for transmis- rier, e.g. its frequency, amplitude, or phase.
sion. This modified analog signal – the modu-
lated signal – can then be transmitted.
moderator /mod-ĕ-ray-ter/ A person or Demodulation is the reverse process to
group who has the authority to intervene in modulation: a signal is obtained from a
a newsgroup or other on-line discussion
modulated signal (by means of a demodu-
forum and to block messages where appro-
lator) and has the same shape as the origi-
priate. A moderated forum has fewer
nal modulating signal.
FLAMES and less SPAM than unmoderated fo-
See also modem.
rums.
141
monitor
monitor A video display unit in a com- puter systems, the motherboard carries all
puter system. the major functional elements, e.g. the
processor and some of the memory; addi-
monochrome /mon-ŏ-krohm/ Using or tional boards plugged into the mother-
involving only one color. For example, a board perform specific tasks such as
monochrome graphics image uses one memory extension or disk control.
color, shades of that color, and no color
(i.e. white). The color is usually black, in Motorola /moh-tŏ-roh-lă/ Formerly one
which case the image is a GRAY SCALE of the leading manufacturers of micro-
image. processors. Until the early 1990s, Mo-
torola chips were used in all Macintosh
monospaced /mon-ŏ-spaysst/ Describ- computers and in many workstations. To-
ing a font in which all the characters oc- gether with Apple Computer and IBM,
cupy the same width. Typewriter-like fonts they launched the PowerPC architecture in
such as Courier are monospaced. They are 1994. In 2004 Motorola’s semiconductor
often used in software programs where division became a separate company
vertical alignment needs to be maintained. (Freescale Semiconductor), allowing Mo-
See also fixed format. torola to concentrate on cellphones and
other communications equipment.
morphing /mor-fing/ The gradual trans-
formation of one image into another. This mouse A pointing device that is moved
is a common motion-picture special effect by hand around a flat surface: the move-
and is available in computer animation ments on the surface are communicated to
packages. a computer and cause corresponding
movements of the CURSOR on the display.
MOS /moss/ (metal oxide semiconductor) The mouse has one or more buttons to in-
See integrated circuit. dicate to the computer that the cursor has
reached a desired position (by a ‘click’).
Mosaic /mŏ-zay-ik/ Client browser soft- Usually, a click on the main mouse button
ware for access to WWW, Gopher, and is equivalent to ‘enter’ on the keyboard;
FTP servers. It was the first WWW browser however, on Windows systems the default
that was available for Macintosh, Win- is that a click selects an item while a dou-
dows, and Unix, all with the same inter- ble-click is equivalent to ‘enter’ and opens
face. Development ceased in 1997. See also it. Using suitable software, other mouse
NCSA. buttons can be assigned different func-
tions. A mouse is normally connected by
MOSFET /moss-fet/ (metal oxide silicon cable to the computer; a ‘tailless’ mouse
field-effect transistor) A field-effect tran- communicates by means of infrared rather
sistor that uses the MOS capacitor, which than electrical signals (as occurs in a TV
is a two-terminal capacitor structure, as its remote control system).
basis.
mousepad /mowss-pad/ (mouse mat) A
most significant digit (msd) The digit small mat that provides better traction for
in the most significant, i.e. leftmost, posi- a mouse ball than the smooth surface of a
tion in the representation of a number, and desk or table. It usually has a plastic sur-
thus making the largest contribution to the face and a thin non-slip rubber or plastic
value of the number. With a binary number backing.
this digit is referred to as the most signifi-
cant bit (msb). MP (multilink point-to-point protocol)
An extension of the PPP protocol that al-
motherboard /muth-er-bord, -bohrd/ A lows multiple physical connections be-
PRINTED CIRCUIT BOARD into which other tween two points to be combined into a
boards can be plugged. In most microcom- single logical connection. The combined
142
MS-DOS
MP3 files
MP3 files uploaded to the
downloaded Internet to be
from the Internet used by another
user
audio is
digitally
compressed
and encoded
to make MP3
files
MP3
connections are called a bundle and pro- removable flash memory cards, and
vide greater BANDWIDTH than a single con- plugged into the player to swap collections
nection. MP is an open standard specified of music. The illegal sharing of music in
in RFC 1990. MP3 format, in breach of copyright, has
been a cause of controversy in recent years.
MP3 A form of audio compression See also iPod.
adapted from the MPEG format for com-
pressed video. Similar to Dolby Digital, MPEG /em-peg/ (moving picture experts
MP3 compresses the sound by filtering out group) The set of digital video compres-
the differences in the sound signal that the sion standards and file formats developed
human ear cannot detect. Additionally, a by a working group of the ISO. MPEG
musical CD has a sampling rate of 44 100 achieves a high compression rate by only
times a second, whereas MP3 can produce storing the changes from one frame to the
almost the same quality by sampling at next, instead of each entire frame. The
16 000 times a second for music and video information is then encoded using
10 000 times a second for voice. This discrete cosine transform. MPEG uses a
means a sound track can be compressed to type of lossy compression as some data is
about one-tenth its original size. removed. However this is generally imper-
ceptible to the eye. There are several
MP3 player A portable device that can MPEG standards, the latest being MPEG-
store sound tracks in MP3 format, usually 7. Files in this format have the file exten-
music, and play them through earphones. sion mpeg or mpg.
MP3 players can be connected to comput-
ers and tracks transferred from the com- msb (most significant bit) See most sig-
puter to the inbuilt memory (RAM) in the nificant digit.
MP3 player. 32-MB flash memory can
store 32 minutes of CD-quality sound msd See most significant digit.
using 128 kilobits of storage for each sec-
ond of sound. MP3 files can be stored on MS-DOS /em-ess-doss/ Trademark An
143
MSN
144
MUX
multiprogramming system /mul-ti-proh- ing the operating system takes the decision.
gram-ing/ A computer system in which Preemptive multitasking is generally pre-
several individual programs may be exe- ferred, because a single misbehaving pro-
cuted at the same time through MULTITASK- cess that does not give up control properly
ING. can bring other processes to a halt in a co-
operative multitasking system.
multitasking /mul-ti-task-ing/ The pro- Some operating systems allow processes
cess of loading and running at least two themselves to be multitasked, by subdivid-
PROCESSES (tasks) at the same time. A MUL- ing them into parallel units of execution
TIPROCESSING SYSTEM, containing several called threads. In such systems the alloca-
processors, is a multitasking system. It is tion of processor time is performed at the
also feasible for a multitasking system to be thread level rather than the process level. In
operated with only one processor. On any general all a process’s threads share its re-
processor only one processing task can sources, such as memory, files, etc. (by con-
take place at a particular time. In a multi-
trast, each process has its own independent
tasking system there is rapid switching be-
set of resources). The use of more than one
tween the tasks to be performed for each of
thread in a process is called multithread-
the processes running. For example, while
ing.
one process is waiting for an input or out-
put operation to be completed, another
process can be allocated use of the proces-
multithreading /mul-ti-thred-ing/ 1. See
sor. TIME SLICING is also used. This switch- multitasking.
ing procedure is under the control of the 2. In data manipulation, the use of extra
OPERATING SYSTEM, which also controls the
pointers in a tree data structure to facilitate
allocation and release of other resources in moving around the structure.
the system.
Multitasking enables optimum use to multiuser system /mul-ti-yoo-zer/ A
be made of the computer resources as a computer system that is apparently serving
whole: a processor operates very much more than one user simultaneously. A
faster than its peripheral devices, and valu- multiuser system is generally a MULTIAC-
able processor time is wasted if the proces- CESS system.
sor remains idle while a process uses a
peripheral. mutual exclusion A programming
Most current operating systems sup- technique used to ensure that a resource,
port multitasking, which can be coopera- such as a file, can be used by only one pro-
tive or preemptive. In cooperative gram at a time.
multitasking, processes give up control vol-
untarily, whereas in preemptive multitask- MUX See multiplexer.
145
N
nagware /nag-wair/ Informal Shareware space and their identifier. A common syn-
that reminds the user at regular intervals tax, used in C++, is ‘nsa::x’ and ‘nsb::x’ to
that payment or registration is needed. distinguish the two procedures called ‘x’.
Namespaces are not only applicable to
NAK (negative acknowledgment) The programs: XML uses them to prevent dupli-
ASCII character 21 transmitted as a con- cate ELEMENT and ENTITY names. Name-
trol code by a receiving station to a sending spaces are generally hierarchical: i.e. it is
station to indicate that the transmitted in- possible to qualify the contents of a name-
formation has been received incorrectly space by dividing it into two or more sub-
and should be resent. See also acknowledg- namespaces. There are various conventions
ment. to ensure that publicly-used namespaces
are unique; for example JAVA namespaces
name A means of referring to or identi- should begin with the Internet DOMAIN
fying some element in a computer pro- NAME of the author.
gram, in a computer NETWORK, or in some
other system. For example, VARIABLES, AR- NAND gate /nand/ A LOGIC GATE
RAYS, and PROCEDURES in a program would whose output is low only when all (two or
be named, as would be the NODES in a net- more) inputs are high, otherwise the out-
work. In many programming languages a put is high. It can thus be regarded as an
name must be a simple IDENTIFIER consist- AND GATE followed by a NOT GATE: it effec-
ing of a string of one or more characters. In tively first performs the AND operation on
most of these languages names can be cho- its inputs then negates the output, i.e. per-
sen quite freely by the programmer. The forms the NOT operation on it. The TRUTH
names used in a program are stated in a se- TABLE for a NAND gate with two inputs
ries of DECLARATIONS. When the program is (A, B) is shown, where 0 represents a low
compiled or assembled, prior to execution, signal level and 1 a high level. The LOGIC
names are converted to MACHINE ADDRESSES OPERATION implemented by the gate is
by means of a SYMBOL TABLE. Compare lit- known as the NAND operation and can be
eral. written as A NAND B for inputs A and B.
146
negation
NAS See network-attached storage. near letter quality (NLQ) A print qual-
ity that is not quite as good as LETTER QUAL-
native compiler The usual form of ITY, the quality produced by a top-quality
compiler, which runs on a computer pro- electric typewriter, but is better than draft
ducing an object file for that computer, as quality. The term is mainly used of dot ma-
opposed to a CROSS COMPILER. trix printers and is becoming less common
with the increased use of laser printers,
natural language Language as used and which generally give better quality than
spoken by humans as opposed to a pro- electric typewriters.
gramming or machine language. One facet
of ARTIFICIAL-INTELLIGENCE research is the negation 1. In computing, a NOT OPERA-
approximating of natural language in a TION, the performance of a NOT opera-
computer environment. tion, or the result of a NOT operation. The
NOT operation can be performed, for ex-
natural-language processing The study ample, on a binary digit or number or a
of computer systems that can react to and logic statement. The outcome is the COM-
understand both spoken and written nat- PLEMENT of that entity. (It should be noted
ural language. that if a number is stored in two’s comple-
ment, then to negate it is to take the two’s
natural-language query A query to a complement.)
147
negative acknowledgment
2. The changing of the sign of a nonzero connected system. The information is sent
number or quantity. The negation of +42 is as an encoded signal, and is transmitted
–42; the negation of –42 is +42; the nega- along communication lines such as tele-
tion of 0 is 0. phone lines, satellite channels, electric ca-
bles, and optical fibers.
negative acknowledgment See acknowl- It is not sensible (in terms of cost and ef-
edgment; NAK. ficiency) for each computer to be directly
connected to every other computer on the
NEQ gate See nonequivalence gate. network: communication lines are expen-
sive to provide and maintain and would be
nested structure A programming struc- unused for much of the time. Direct con-
ture that is incorporated into another nections are made only at certain points in
structure of the same kind. More than one the network, called NODES. (Strictly, a net-
structure, of the same kind, can be incor- work is a set of interconnected nodes.)
porated. For instance, one or more LOOPS Computers are attached to some or all of
can be nested in another loop. the nodes. Nodes may be at a junction of
two or more communication lines or may
Net See Internet. be at an endpoint of a line. A particular
piece of information has to be routed along
NetBIOS (network basic input output a set of lines to reach its specified destina-
system) An application programming in- tion, being switched at the nodes from one
terface (API) that adds special functions to line to another, often by a computer. A
the MS-DOS BIOS for local area networks. particular destination is specified by its AD-
DRESS, which is indicated by the sender and
netiquette /net-ă-ket/ Internet etiquette; forms part of the transmitted signal.
guidelines for posting messages to on-line The computers on a network may be
services and particularly newsgroups. organized in several ways (see diagram). In
a star structure all communication lines
Netscape Navigator /net-skayp/ A connect directly to a single central or con-
Web BROWSER produced by Netscape Com- trolling node. The endpoints on each line
munications Corporation, which was cannot therefore communicate directly. In
founded in 1994 by James H. Clark and a ring structure there is a single loop of
Marc Andreessen. It was based on MOSAIC communication lines (usually electric ca-
and was one of the first commercial Web bles) between the various nodes on the net-
browsers. It runs on all major platforms, work. Each node is thus connected to two
Windows, Macintoshes, and UNIX. For- adjacent ones. Information can pass in
merly the leading Web browser, it has been only one direction around the loop. A bus
overtaken by INTERNET EXPLORER. configuration is linear rather than circular.
All nodes are connected to a single electric
Netware /net-wair/ A local area net- cable acting as a bus, and information can
work (LAN) operating system developed travel in both directions.
by Novell Corporation. Formerly the lead- See also local area network; wide area
ing network operating system, it has been network; packet switching; message
overtaken by WINDOWS-based systems. switching.
148
network-attached storage
star network
ring network
bus network
nodes
flow of information
controlling node
149
network computer
tion of data storage and its separation from network topology See network archi-
other network functions simplifies both tecture.
network management and the addition of
new storage capacity. NAS is not appropri- neural network /new-răl/ A network of
ate to all networks: it is unnecessarily so- interconnected components (called neu-
phisticated for small networks, while the rons), designed so that an input signal to
faster STORAGE AREA NETWORKS are often each component is multiplied by a weight-
preferred in large networks. ing factor and passed to two or more other
components, and so on. The output signal
network computer See thin client. depends on the weighting factors used, and
these can be adjusted so as to optimize the
network database 1. A type of data- final output. Neural networks are designed
base that is similar to but less rigorously by analogy with the transmission of im-
structured than a HIERARCHICAL DATABASE, pulses in the human brain. They do not
in that records, or nodes, can have more have to be physical networks; usually they
than one parent. Network databases are are simulated by a computer. They can be
less powerful than RELATIONAL DATABASES. ‘trained’ to recognize patterns in large
2. A database that runs in a network. amounts of data and are used for making
3. A database containing the addresses of predictions (e.g. in financial markets).
users on a network. They are also used for research in ARTIFI-
CIAL INTELLIGENCE.
150
NOR gate
designed to interface with a particular net- high and the rest are low; otherwise the
work, protocol, and media. output is low. It effectively performs the
operation of EQUIVALENCE on its inputs and
NLQ See near letter quality. then negates the output, i.e. performs the
NOT OPERATION on it. The TRUTH TABLE for
NNTP (network news transport protocol) a gate with two inputs (A, B) is shown,
The protocol used by client and server soft- where 0 represents a low signal level and 1
ware to carry USENET postings back and a high level.
forth over a TCP/IP network. The NEQ gate is also called the exclu-
sive-OR (or XOR) gate: it excludes the
node 1. A point in a computer NETWORK possibility (allowed in the OR GATE) that
where communication lines, such as tele- the output is high when more than one
phone lines, electric cables, or optical input is high. See also equivalence gate.
fibers, are interconnected, or the endpoint
of such a line. Some nodes are used to con- noninterlaced /non-in-ter-laysst/ Describ-
nect client computers, SERVERS, printers, ing monitors and video standards that do
etc., to the network. Others, especially on not use INTERLACING techniques to improve
larger networks, are devices dedicated to resolution.
the operation of the network itself, for ex-
ample HUBS, NETWORK SWITCHES, RE- nonlocal variable /non-loh-kăl/ See scope.
PEATERS, BRIDGES, and ROUTERS.
2. See tree. nonprocedural language A PROGRAM-
MING LANGUAGE where the programmer
noise Any unwanted electrical signals states the problem but does not specify the
occurring in a system and producing dis- exact steps (procedure) to be used in solv-
turbances in the output of the system. ing it. FOURTH and FIFTH GENERATION LAN-
Noise can disturb a TV picture or make a GUAGES are nonprocedural languages.
telephone message inaudible. It can also in- Compare procedural language.
troduce errors in data communications.
nonvolatile memory /non-vol-ă-tăl/ A
nonbreaking space A character repre- type of memory whose contents are not
senting a space between words that must lost when the power supply to the memory
not be split between lines in text format- is switched off. ROM is nonvolatile mem-
ting. ory. Compare volatile memory.
A B output A B output
0 0 0 0 0 0
0 1 1 0 1 0
1 0 1 1 0 0
1 1 0 1 1 0
Truth table for nonequivalence (NEQ) gate Truth table for NOR gate
151
normalization
inputs are low, otherwise the output is low. notation The set of symbols and for-
It can thus be regarded as an OR GATE fol- mats used in a programming language.
lowed by a NOT GATE: it effectively first per-
forms the OR operation on its inputs then notebook A small LAPTOP computer
negates the output, i.e. performs the NOT usually weighing less than 6 lbs that will fit
operation on it. The TRUTH TABLE for a inside a briefcase.
NOR gate with two inputs (A, B) is shown,
where 0 represents a low signal level and 1 NOT gate (inverter) The only LOGIC
a high level. The LOGIC OPERATION imple- GATE with a single input. When the input is
mented by the gate is known as the NOR high then the output is low; when the input
operation and can be written as A NOR B is low the output is high. The NOT gate
for inputs A and B. thus performs the NOT OPERATION on its
The NOR gate is important because input and has the same TRUTH TABLE. The
any logic gate can in theory be constructed truth table is shown in the diagram, where
from a suitable combination of NOR 0 represents a low signal level and 1 a high
gates. For example, a two-input NOR gate level.
with one input kept permanently low (i.e.
earthed) acts as a NOT gate. An OR gate is NOT operation A LOGIC OPERATION
produced by feeding the output of a NOR with only one OPERAND. When applied to
gate through such a NOT gate, i.e. an operand A (a statement or formula, for
NOT (A NOR B) = A OR B instance), the outcome is false if A is true
and is true if A is false. It thus negates the
This can be checked by means of the NOR
truth value of A. The TRUTH TABLE is shown
and OR truth tables. Again, an AND GATE is
in the diagram, where the two truth values
produced if each of the two inputs to a
are represented as 0 (false) and 1 (true).
NOR gate is first fed through a NOT gate,
The operation can be written in several
i.e.
ways, including ∼
(NOT A) NOR (NOT B) = A AND B
NOT A ¬A A ~A A′
See also NAND gate.
The symbol next to the operand A (i.e.
NOT, ¬, etc.) is a LOGIC OPERATOR for the
normalization 1. In relational database NOT operation.
design, the organization of data to mini- In a computer the NOT operation is
mize duplication. The ideal is that data is used in HIGH-LEVEL LANGUAGES to negate a
organized into tables and relationships be- logical expression. It is used in LOW-LEVEL
tween tables in such a way that field mod- LANGUAGES on a BYTE or WORD, producing
ifications, additions, and deletions need a result by performing the NOT operation
take place in only one table. Although nor- on each bit in turn. If 8-bit words are used,
malization makes databases more efficient then for example with the operand
to maintain, they can become very com- 01010001
plex. the outcome of the NOT operation is
2. In programming, changing the represen- 10101110
tation of a floating-point number so that See also NOT gate.
the most significant digit in the mantissa is
not zero. NSFnet /en-ess-eff-net/ A wide area net-
work that was developed by the National
NOR operation See NOR gate. Science Foundation and replaced
ARPANET as the main USA government
NOS /noss/ See network operating sys- network linking research establishments
tem. and universities. It was replaced in 1995
with a commercial Internet backbone. A
NOT /not/ A Boolean operator that re- new backbone for research and develop-
turns the value true when the operand is ment, the very high speed backbone net-
false and vice versa. work service (vBNS), was also set up.
152
num lock key
153
O
object An item of data, such as a picture, and must be LINKED to other, separately
a spreadsheet, or a CAD document, to be compiled, object modules to form the com-
used in a program. See also object-oriented plete program. An object module contains
programming. object code, which is either MACHINE CODE
or, for such languages as JAVA, a machine-
object code See object program. independent code that is interpreted at run-
time. See also object program.
object file A file containing object code,
which has normally been generated by a object-oriented graphics The repre-
compiler or assembler. sentation of images by the use of a combi-
nation of basic graphical shapes (polygons,
object linking and embedding (OLE) ellipses, curves, lines, etc.). Because of the
The use of a data OBJECT created in one ap- way they are defined such objects can be
plication (the server or object application) layered, rotated, or resized with relative
by another (the client application). A file ease. See also computer graphics.
created by the server application is inserted
at the desired point in the client document. object-oriented programming (OOP)
A compound document is created and the Programming using a language in which all
object can be edited from the client appli- the files, inputs, and outputs are consid-
cation using the server application. For ex- ered as discrete objects, and the programs
ample, a spreadsheet inserted into a dealing with these are part of the same ob-
word-processing document will appear as jects. This is a form of modular program-
a seamless part of the document when it is ming in which the definition of each data
printed, but is updated from within the structure is normally contained in one
word-processing application by automati- module. When a data structure’s internal
cally calling the spreadsheet application. design is altered and the implementation
An object can be inserted in one of two details of actions carried out on it need to
ways. If it is linked, the compound docu- be changed, the program modifications are
ment contains a link to the object’s original localized and easier to control.
file; any changes made to this file are re-
flected in the compound document, and object program The program that is in
vice versa. If the object is embedded, a copy object code and is ready to be loaded into
of the original file is taken and stored a computer for execution. An object pro-
within the compound document; no con- gram consists of one or more OBJECT MOD-
nection between the original file and the ULES.
compound document is maintained and
changes will not be replicated. OCR (optical character recognition) A
process in which a machine scans a docu-
object module A unit of output from a ment and is able to convert the characters
COMPILER or ASSEMBLER, usually the trans- that it recognizes into a code that can be
lated form of one file of source code. An fed straight into a computer. The machine
object module may be a self-contained pro- is called an optical character reader, also
gram, or it may be part of a larger program sometimes abbreviated to OCR. Scanners
154
off-line browser
binary 000 001 010 011 100 101 110 111
octal 0 1 2 3 4 5 6 7
with appropriate software can also func- base-access technologies, ODBC is still
tion as optical character readers. The char- widely used.
acters can be read not only by the machine
but also by a person: they can be letters, odd parity See parity check.
digits, punctuation, and other symbols.
Early optical character readers could only OEM (original equipment manufacturer)
recognize a limited number of rather styl- Usually, an organization that purchases
ized characters. Current devices can now equipment, such as electronic circuitry,
recognize characters in a variety of type- that is to be built into one of its products.
faces, sizes, layouts, etc., and may even rec- In some cases the term is used to refer to
ognize handwritten characters. OCR is the supplier of the equipment.
widely used in billing and banking, opti-
cal characters being used for example on Office (Microsoft Office) A suite of re-
gas and electricity bills. See also MICR; lated applications, developed by Microsoft
OMR. and designed to meet the WORD-PROCESS-
ING, SPREADSHEET, E-MAIL, etc., needs of
octal notation /ok-tăl/ (octal) A number computer users. There are several editions
system that uses 8 digits, 0–7, and thus has of Office, aimed at different types of user.
BASE 8. It is a positional notation (see num- All mainstream editions include WORD,
ber system), positional values increasing EXCEL, OUTLOOK, and POWERPOINT; ACCESS
from right to left by powers of 8. Octal is a and other programs are added in those
convenient shorthand by which people aimed at business users. Although some of
(rather than machines) can handle binary these applications existed prior to Office
numbers. Each octal digit corresponds to a and all can be used on their own, they can
group of 3 binary digits, or bits. Conver- easily share data and successive versions
sion of binary to octal is done by marking have increased common features that assist
off groups of 3 bits in the binary number integration and management in a corpo-
(starting from the right) and replacing each rate environment. First released for the
group by its octal equivalent. Conversion Macintosh in 1989 and Windows in 1990,
of octal to binary is done by replacing each the current versions are Office 2003 (Win-
octal digit by its equivalent binary group. dows) and Office 2004 (Macintosh).
The use of octal notation has declined in
favour of HEXADECIMAL NOTATION. See also off-line 1. Of PERIPHERAL DEVICES or FILES:
binary notation. not under the control of the processing unit
of a computer, i.e. not connected to a com-
ODBC (open database connectivity) The puter system or not usable. A peripheral
Windows interface, part of the WOSA device may be connected to a computer but
(Windows open system architecture), that is off-line if the system has been instructed
provides a common language for applica- not to use it.
tions needing network database access. 2. Involving peripheral devices or files that
The goal of ODBC is to make it possible to are off-line (as in off-line processing or off-
access any data from any application re- line storage).
gardless of which database management Compare on-line.
system is handling the data. In order to
achieve this, both the data and the applica- off-line browser A browser designed to
tion must be ODBC-compliant. Despite the read HTML files from a disk rather than
subsequent release of new Microsoft data- on the Internet.
155
off-line navigator
156
operation
ifications are public. Open architecture al- LOGIC OPERATION is performed. For exam-
lows other manufacturers or developers to ple, in
produce add-ons. (3 + 5)/2
2. A design allowing for expansion slots on 3, 5, and 2 are the operands. The symbols
a motherboard. + and / are called OPERATORS.
2. The parts of a MACHINE INSTRUCTION
open database connectivity See ODBC. that specify the objects upon which the op-
eration is to be performed. For instance, in
Open Group An international group of the instruction
computer hardware and software manu- ADD A,B
facturers, and users, formed in 1996 by the A and B are the operands and could be REG-
consolidation of the Open Software Foun- ISTERS in the central processor, or actual
dation and X/Open Company Limited, to values, or ADDRESSES of values, or even ad-
promote the furthering of multivendor in- dresses of addresses of values. See also ad-
formation systems. The Open Group owns dressing mode.
the trademark of the UNIX operating system
and so certifies whether a particular imple- operating system (OS) A program or
mentation is or is not permitted to describe collection of programs used for managing
itself as ‘UNIX’. the hardware and software resources of a
computer system on behalf of the user(s).
open source A programming methodol- Every general-purpose computer has an
ogy founded on the premise that better operating system. In general a portion of
software will result by making the SOURCE
the operating system is always resident in
CODE available for any user or interested
MAIN STORE; the rest is kept on BACKING
party to study and modify. All contribu-
STORE and is read into main store when re-
tions are then fed back into the master ver-
quired.
sion, which is updated frequently. Open
The operating system provides the USER
source has had some successes, notably the
INTERFACE that interprets commands typed
LINUX operating system. It contrasts with
on a user’s KEYBOARD and movements of
closed source, or proprietary, software,
the MOUSE, and provides information about
where the source code is a closely-guarded
secret of the developer and all development the current state of the system on the VDU.
is strictly controlled. The merits of the two In GUI systems it manages the WINDOWS that
methodologies are hotly debated. appear on the screen. It also manages
queues for slower devices such as PRINTERS,
open standards Standards for the spec- keeps track of where FILES are stored, what
ification of hardware or software that is they are named, and who they belong to,
generally known and can be used by any- and transfers files between disks or be-
one. Open standards encourage the devel- tween disks and main store when in-
opment and use of new technologies. structed. Input from and output to any
connected INPUT or OUTPUT DEVICES is con-
open system Any system that conforms trolled by the operating system. In MULTI-
to standards that are not specific to the TASKING systems the operating system is
particular supplier, but that are generally responsible for deciding in what order and
well defined and well known. for how long each processing task is to
have the central processor, and what sec-
open systems interconnection See tions of memory they are to use. All com-
OSI. munications are managed by the operating
system – not only links between the user(s)
OpenType Trademark See font. and the computer but also links to other
computers or to a NETWORK.
operand 1. In general, a quantity or
function upon which an arithmetic or operation 1. In mathematics, the act of
157
operation code
OPERATING SYSTEMS
BASIC FUNCTIONS
Function Description
CPU management To ensure that every process and application that is running has
enough processor time to funtion properly and to optimize the
use of processor cycles by scheduling processor priorities. Dealing
with hardware and software interrupts.
Memory management The different types of memory in the system must be used
effectively and each process must have enough secure memory
in which to run.
I/O management Largely the management of data in the form of a stream of bits
coming in from or going out to a device such as a keyboard,
monitor, printer, or disk.
Application interface The provision and management of an API that lets applications
use operating system functions and request input/ouput functions
without having to keep track of the CPUs basic operation.
158
OR gate
159
OR operation
put is low only when all (two or more) in- after the market success of Windows
puts are low; otherwise the output is high. prompted Microsoft to withdraw from the
It thus performs the OR OPERATION on its partnership. Following a period of rivalry
inputs and has the same TRUTH TABLE. The in the early 1990s OS/2 is no longer a
truth table for a gate with two inputs (A, B) major force but is still supported by IBM
is shown, where 0 represents a low signal and is used in specialized areas.
level and 1 a high level.
OSI /oh-ess-ÿ/ (open systems interconnec-
tion) A standard description for how
A B output/outcome messages should be transmitted between
two points in a telecommunications net-
0 0 0 work. Its purpose is to guide developers to
0 1 1 produce products that will work with
1 0 1 other products in a consistent fashion. OSI
1 1 1 divides telecommunications into seven lay-
ers of functions.
Truth table for OR gate and OR operation
outage /owt-ij/ The time when a com-
puter system is out of use because of ser-
OR operation A LOGIC OPERATION com- vicing or maintenance.
bining two statements or formulae, A and
B, in such a way that the outcome is false outline font A font in which the shape
only when A and B are both false; other- of the character is defined by a set of com-
wise the outcome is true. The TRUTH TABLE mands. Page description languages such as
is shown in the diagram, where the two PostScript treat everything on a page as a
truth values are represented by 0 (false) graphic and use outline fonts, which are
and 1 (true). The operation can be written more versatile at producing different sizes
in several ways, including of type with different attributes or special
A OR B A ! B A | B A + B effects. TrueType and OpenType are other
A and B are known as the OPERANDS and kinds of outline fonts.
the symbol between them (OR, !, |, +) is a
LOGIC OPERATOR for the OR operation. Outlook (Microsoft Outlook) Trade-
In a computer the OR operation is used mark An E-MAIL and personal-information
in HIGH-LEVEL LANGUAGES to combine two management application developed by Mi-
logical expressions according to the rules crosoft. First released with OFFICE 97, it has
stated above. It is used in LOW-LEVEL LAN- been upgraded several times and the cur-
GUAGES to combine two BYTES or WORDS, rent version is Outlook 2003.
producing a result by performing the OR
operation on each corresponding pair of output 1. The results obtained from a
bits in turn. If 8-bit words are used, then computer system following some process-
for example with operands ing activity. Computer output may be in a
01010001 form that people can understand and use,
11110000 such as words or pictures that are dis-
the outcome of the OR operation is played on a screen or printed or drawn on
11110001 paper. Alternatively, it may be written on,
See also OR gate. say, a magnetic disk or a CD so that it can
be fed back into the computer system – or
OS /oh-ess/ See operating system. into another system. See also output de-
vice.
OS/2 /oh-ess-too/ A multitasking GUI 2. The signal – voltage, current, or power –
operating system created by Microsoft and obtained from an electric circuit.
IBM in the late 1980s. Originally intended 3. To produce a signal or some other form
as the successor to MS-DOS, OS/2 suffered of information.
160
overlay
5
Protocols here are
application set
5
stream for the physical network
Protocols here are Maintains flow control i.e., it integrates data
communications from each application into a single stream
network 3
transport set
output device Any device that converts clocked system to be sold without inform-
the electrical signals representing informa- ing the buyer.
tion within a computer system into a form
that can be understood and used by people. overflow The condition arising when
PRINTERS, VDUS, loudspeakers, and PLOT- the result of an arithmetic operation in a
TERS are the most common types of output computer exceeds the size of the location
device. An output device may also convert allocated to it. For example, a number rep-
internal signals into a coded form on, say, resented by 33 bits would not fit in a 32-bit
magnetic disk that can be fed back into a word. If there were no means to detect
computer system at a later date. Compare overflow, results would be unreliable and
input device. might be incorrect. Facilities are therefore
provided to detect overflow: a particular
output medium See media. bit, known as an overflow flag or bit, is
set to 1 when overflow occurs. The over-
overclocking The process of reconfig- flow can then be corrected. See also under-
uring a computer to operate the CPU at a flow.
higher clock speed than specified for.
Overclocking carries risks to the CPU in- overlay 1. Part of a program, i.e. a sec-
cluding the possibility of permanent dam- tion of code, that is loaded into MAIN STORE
age. Successful overclocking is more likely during the execution of that program,
to be achieved if the system has a well- overwriting what was previously there.
designed motherboard with a fast enough The section of code is thus not permanently
bus and a suitable cooling fan and an Intel resident in main store but is held in BACK-
microprocessor. It is illegal for an over- ING STORE until required by the processor.
161
overrun
The parts of the program to be used as 2. A piece of card or plastic placed over or
overlays, and the other parts that these are around a KEYBOARD to indicate what spe-
allowed to overwrite, are determined by in- cial functions have been assigned to partic-
structions to the LINK EDITOR at link time. ular keys.
This has to be done with care: if an overlay
tries to use a procedure in a part of the overrun The loss of data incurred when
code it has just overwritten, then the pro- data arrives at a port faster than it can be
gram will fail. processed.
In general several overlays are loaded
overtype (overwrite) In text entry, the
into the same area of main store at differ-
mode in which characters being typed re-
ent stages of the program, each erasing the
place those under or to the right of the in-
code already there but no longer needed. sertion point.
This process of repeatedly using the same
area of store during program execution overwrite 1. To destroy the contents of
overcomes space limitations in main store. a storage location by writing a new item of
The technique, which is called overlaying, data to that location.
is used mainly in systems that do not have 2. To write a new version of a file on top of
VIRTUAL STORAGE – principally some older the existing version.
mainframes. 3. See overtype.
162
P
to be transmitted over
e-mail message of a network with a fixed
4300 bits packet length of 1024
bits
6
protocol
packet number
Message packet
163
packing
number of packets. Each packet is cording to rules set by the user. However a
switched as a composite whole from one ‘hard’ or ‘manual’ break can be placed in a
communication line to another. At each document to force the following text to be
switching point along its route the packet is printed or displayed on a new page.
checked for errors, stored temporarily, and
forwarded to the next point when the nec- page description language (PDL) A
essary resources become available. The type of programming language in which
packets of a partitioned message, docu- commands are used to describe the layout
ment, etc., are reassembled when they of a complete page of text and graphics, in-
reach their final destination (see packet as- cluding font styles and sizes and picture
sembler/disassembler). Compare message shapes and positions. POSTSCRIPT is a
switching. widely used PDL.
packing The storage of data in a com- page orientation The position of a page
pact form in order to reduce the amount of according to whether the top of the page is
storage space required. The term is used as along the short edge of the paper (portrait)
a synonym for the simpler methods of or along the long edge of the paper (land-
COMPRESSION – i.e. those that remove obvi- scape).
ous redundancies in the data without using
a sophisticated encoding algorithm. For page printer A PRINTER that produces a
example, several BYTES may be stored in complete page at a time, an example being
one WORD, or multiple occurrences of a the LASER PRINTER.
character or word may be replaced by a
single instance of the character or word paging The process of managing VIR-
and an indication of the number of times it TUAL MEMORY by dividing the space into
occurs. The original data can be recovered fixed-size pages, which can then be
from its packed form by unpacking it. mapped onto an available physical ad-
dress. Address translation from physical to
packing density 1. The number of com- virtual is done by memory management
ponents, or LOGIC GATES, per unit area of hardware.
an INTEGRATED CIRCUIT.
2. The amount of information in a given di- paint To infill a part of a drawing on the
mension of a storage medium, for example screen with color or a pattern.
the number of tracks per inch of a magnetic
disk. paint program A simple drawing pro-
gram that allows freehand drawing and
PAD /pad, pee-ay-dee/ See packet assem- also provides tools to facilitate the drawing
bler/disassembler. of lines and curves. Paint programs pro-
duce raster graphics files.
padding The adding of blanks to extend
a string or record to a set length. palette 1. The tools provided in a graph-
ics program, which enable different pat-
page The unit of interchange between terns, colors, brush shapes, and line widths
main store and a backing-store device used to be selected for use.
for SWAPPING. A page contains instructions 2. The set of different colors available in a
or data, or both, and generally the number program.
of words or bytes is fixed for a given com-
puter system. See also virtual storage. palmtop /pahm-top/ A hand-sized com-
puter. Often palmtops have no keyboard
page break The position in a document but the screen serves as both an input and
at which text will move to the top of a new output device. A stylus is used to enter in-
page. Word-processing systems will usu- formation; text can be entered by manually
ally paginate documents automatically ac- writing on the screen with the stylus. A
164
parity check
palmtop is generally used for simple appli- a character) are transmitted at the same
cations such as personal organization and time along different paths. For example, if
note taking. See also PDA. two devices wish to communicate a 16-bit
unit of data, the sending device would
PANTONE /pan-tohn/ Trademark A transmit all the 16 bits at the same time,
system of coding colors used in commercial one bit per wire. More than 16 wires are
color printing. required, however, since control signals are
sent simultaneously, again one bit per wire,
paperless office The (mistaken) idea indicating that data is available (see hand-
that the introduction of computers would shake). The receiving device would then
mean that paper would no longer be neces- signal that the data had been received. See
sary in offices. also serial transmission.
parallel port An interface for PARALLEL parity check A simple CHECK used to de-
TRANSMISSION. tect ERRORS in items of information in bi-
nary form. The item may represent a piece
parallel processing Strictly, the execu- of data or part of a program, and may be a
tion of two or more PROCESSES at the same single CHARACTER, BYTE, or WORD. An extra
time in a single computer system. This im- BIT, called the parity bit, is added to the
plies that two or more PROCESSORS are in leftmost end of the item. There are two
operation at the same time. The term is kinds of parity: even parity and odd parity.
often loosely applied to MULTITASKING, a In the former, the parity bit is used to make
situation in which a number of processes the total number of 1s an even number; in
are potentially active, but only one is actu- the latter, the parity bit is used to make the
ally being run at any particular instant. See total number of 1s an odd number. For ex-
also multiprocessing system. ample, the 7-bit ASCII code for W is
1010111. The number of 1s is 5; the parity
parallel transmission A method of bit for even parity is thus a 1, making the
sending data between devices, typically a number of 1s up to 6 (an even number) and
computer and its peripherals, in which all the 8-bit ASCII code with parity
the BITS associated with a unit of data (e.g. 11010111.
165
park
The parity bit is recomputed after the password A means of verifying the
augmented item – original item plus parity identity of someone wishing to gain access
bit – has been transferred or otherwise ma- to a computer system or to certain parts of
nipulated. The new value is compared with a system. A password is a particular se-
the received value, and most simple bit er- quence of characters (letters, numbers,
rors (certainly all single bit errors) will be etc.) that is allocated to or chosen by an au-
detected. thorized user and is unique to that person.
The same principle can be used to check A potential user must feed his or her pass-
larger blocks of data. In most RAID disk ar- word into the computer to establish iden-
rays, for example, the equivalent bits on all tity. The password entered is checked
but one of the disks generate a parity bit on against the stored version of the password.
the last disk. Only if the two sequences agree will the
person be granted access.
park To lock the read/write heads in a Passwords form part of the LOG ON se-
secure position (over an unused track or quence in most modern computer systems.
beyond the surface of the disk) before shut- They are used, for example, in systems that
ting down the drive, so that data cannot be hold sensitive data (i.e. for security rea-
corrupted when the computer is physically sons) or when charges are made for use of
moved. On all modern drives parking is system resources. Passwords can also be
done automatically but it can also be granted to programs in order to control ac-
achieved by using a disk utility. The disk cess to data.
Methods commonly used to make a
will unpark itself automatically when the
password system more effective are:
power is turned back on.
1. the passwords are stored on the com-
puter in code so that unauthorized people
parsing, parser /par-zing, par-zer/ See
looking at the password file cannot read
syntax analysis.
the passwords;
2. the password does not appear on the
partition A logical division of a hard
screen when it is typed in, thus making it
disk. It enables the user to have different
harder to discover users’ passwords by
operating systems on the same drive (in dif-
watching them log on;
ferent partitions) and can also create the 3. users are urged, and in some cases forced
appearance of having separate drives, by the system, to change their passwords
which can make it easier to organize files frequently.
and provide access for multiple users. 4. users are urged, and in some cases
forced, to use passwords that are not sim-
Pascal /pass-kăl/ A PROGRAMMING LAN-
ple words or predictable numbers (e.g.
GUAGE named after the French philosopher birthdays) but a mixture of letters, num-
and mathematician Blaise Pascal and de- bers, and other characters, sometimes in-
veloped by Niklaus Wirth in the late termingled in a random fashion.
1960s. Pascal was intended to be used for 5. the password must be of a minimum
teaching students the principles and prac- length, often 6 or 7 characters.
tice of programming, and is a relatively
small concise HIGH-LEVEL LANGUAGE with paste To insert information that has
what were at the time a number of novel been cut or copied from a document into a
features. Pascal has certain drawbacks different place in that document or into an-
when it comes to large-scale programming other document.
but is nevertheless popular and has been
implemented on all sizes of computer. patch A small section of code, often pro-
vided by a software supplier, that allows a
passive matrix display An inexpensive user to correct or modify a program. It is
type of low-resolution LCD. See also active used for purposes of convenience and
matrix display. speed of change. The patch may be written
166
Pentium
in MACHINE CODE and introduced into the PCI (peripheral component interconnect)
compiled (or assembled) version of a pro- A local bus developed by Intel Corporation
gram, i.e. the OBJECT PROGRAM. and some other large companies to allevi-
ate problems caused by the slow speed of
path 1. A link between two nodes in a ISA. Introduced in the mid-1990s, it has
network. continued to develop and is still the stan-
2. The route followed by an operating sys- dard expansion bus for PCs.
tem through a disk’s directories and subdi-
rectories to a file. PCL (printer control language) Trade-
mark A control language developed by
pattern recognition The ability of a Hewlett Packard for laser printers.
computer to recognize visual images or
sound patterns that are represented as an PDA (personal digital assistant) Any
array of numbers. small portable hand-held device, possibly a
PALMTOP computer, that provides a com-
pause instruction An instruction that puting capability covering information
specifies that the execution of a program is input, retrieval, and storage for business,
to be suspended. After a pause, a program college, or personal use. A PDA is often
may be continued from where it left off or used as a personal organizer with calen-
was abandoned. For example, a program dars, diaries, and address books. Increas-
might pause to allow the screen to be read, ingly PDAs are combining with cellphones
or so that the paper may be changed on a and paging systems. Data can be trans-
printer. ferred between a PDA and a personal com-
puter.
pause key A keyboard key positioned to
the right of the function keys. When PDF (portable document file) See Acro-
pressed it pauses the execution of a pro- bat.
gram.
PDL See page description language.
payload The part of a VIRUS or WORM
that is intended to damage the infected sys- peer-to-peer Describing the linking of
tem, as opposed to the part concerned with computers together in a NETWORK without
spreading the program to other machines. a dedicated SERVER. Each individual com-
puter acts as a server to other computers
PC See personal computer. (its peers) on the network and is also a
client to all its peers acting as servers. Re-
PCB (pcb) See printed circuit board. sources can be shared and messages passed
between users. All versions of Windows
PC-compatible See IBM-compatible. since Windows 95 have offered limited
peer-to-peer networking.
PC-DOS /pee-see-doss/ Trademark An
OPERATING SYSTEM produced by Microsoft PEM See privacy-enhanced mail.
Corp. for IBM and intended for use on the
IBM Personal Computer. (Its official name pen computer Any computer whose
in IBM documentation was ‘DOS’.) It was primary input device is a stylus (‘pen’). See
essentially the same as MS-DOS, but made palmtop.
assumptions about the underlying hard-
ware that were only fully justified on IBM pen plotter See plotter.
PCs. IBM took over development in the
1990s and PC-DOS diverged from MS- Pentium /pen-tee-ŭm/ Trademark A fam-
DOS. However, its use declined as WIN- ily of processors first introduced in 1993
DOWS became the dominant PC operating by Intel. These had speeds up to 133 MHz.
system. The original Pentium has been succeeded
167
peripheral component interconnect
by the Pentium Pro, the Pentium II, the computer system and containing a ma-
Pentium 3, and the current version, the chine-readable sequence of characters that
Pentium 4. Speeds of 3 GHz are now stan- identifies that person. It may be a card or
dard. Each model has, in its turn, been the badge. The PID must be inserted into a ter-
standard microprocessor for PCs, although minal of the computer system before access
other companies have brought out compet- to the system is granted. In many cases a
itive clones. Since the Pentium II each person must use a PID in conjunction with
model has been marketed in three versions: a PIN (personal identification number) to
the Pentium, aimed at desktop and note- gain access.
book PCs; the Xeon, a high-end version
aimed at servers and high-specification personal identification number (PIN)
workstations; and the Celeron, a reduced A number allocated to an authorized user
specification aimed at the bottom end of of a computer system that is unique to that
the home and small business market. person, and which he or she must feed into
the computer system in order to establish
peripheral component interconnect identity. A PIN is a form of PASSWORD. It is
/pĕ-riff-er-ăl/ See PCI. often used in conjunction with a PID (PER-
SONAL IDENTIFICATION DEVICE). PINs and
peripheral device (peripheral) Any de- PIDs are used, for example, at cash dis-
vice that can be connected to and con- pensers.
trolled by a computer. It is external to the
CENTRAL PROCESSOR and MAIN STORE of the personal information manager See PIM.
computer. Peripherals may be INPUT DE-
VICES, OUTPUT DEVICES, or BACKING STORE. peta- (pet-a) A prefix indicating a multi-
ple of a thousand million million (i.e. 1015)
Perl /perl/ (practical extraction and report or, loosely, a multiple of 250 (i.e. 1 125 899
language) A script programming lan- 906 842 624). In science and technology
guage with a syntax similar to C and in- decimal notation is usually used, and pow-
corporating some popular UNIX features. ers of 10 are thus encountered - 103, 106,
It has good text-handling features and can 109, etc. The symbol P is used for peta-, as
also deal with binary files, making it suit- in PV for petavolt. Binary notation is gen-
able for developing common gateway in- erally used in computing, and so the power
terface software. of 2 nearest to 1015 has assumed the mean-
ing of peta-.
personal computer (PC) A general- The prefix is most frequently encoun-
purpose computer designed for operation tered in computing in the context of stor-
and use by one person at a time. Such a de- age CAPACITY. With magnetic disks, mag-
vice became a reality when MICROCOM- netic tape, and main store, the capacity is
PUTERS became available. All personal normally reckoned in terms of the number
computers are microcomputers, but micro- of BYTES that can be stored, and the
computers are not necessarily personal petabyte is used to mean 250 bytes;
computers. Personal computers range from petabyte is usually abbreviated to P byte,
cheap domestic machines to expensive and Pb, or just to P.
highly sophisticated systems. Often the ab- Peta- is part of a sequence
breviation ‘PC’ is used in a more restricted kilo-, mega-, giga-, tera-, peta-, ...
sense to mean a computer that will run of increasing powers of 103 (or of 210).
Windows, as opposed to an Apple Macin-
tosh computer. phishing An attempt to obtain personal
information from Internet users through
personal digital assistant See PDA. fake Web sites and fraudulent e-mails. A
common tactic is to send an e-mail al-
personal identification device (PID) A legedly from a bank that links to a Web site
device issued to an authorized user of a where users are asked to ‘confirm’ such de-
168
pin header
tails as their online-banking username and PICS /piks, pee-ÿ-see-ess/ (platform for In-
password, credit-card number and expiry ternet content selection) A specification
date, etc. These are then used to steal money of the format conventions used by Internet
from the user’s account or run up bills on rating systems. This was developed to en-
the credit card. Alternatively the user can able parents, teachers, and administrators
be asked for personal information, which to limit the type of material accessed on the
is then used to set up bank accounts, obtain Internet, especially that available to chil-
credit, etc. Such frauds can vary from the dren. It now also includes the protection of
crude to the highly sophisticated: the fake privacy.
Web site, for example, is often identical in
look and feel to the bank’s genuine one. PICT /pikt, pee-ÿ-see-tee/ A Macintosh
The term ‘phishing’ derives from the anal- file format for encoding graphical images,
ogy with fishing: while many people will which can hold both object-oriented im-
ignore the bait (the e-mail), some will bite ages and raster graphics images. Although
and be reeled in. To phish is to engage in now superseded by more sophisticated for-
phishing, and the people who do so are mats, it is still widely used.
sometimes known as phishers.
picture processing See image process-
phono connector /foh-noh/ A connec- ing.
tor used to attach a device such as a micro-
phone or pair of headphones to a computer PID See personal identification device.
peripheral with audio capability.
Pilot (programmed inquiry, learning or
PhotoCD /foh-toh-see-dee/ A Kodak teaching) An early programming lan-
system that can store 35-mm film pictures, guage that was used largely in CAL.
negatives, slides, and scanned images as
digitized images on CD. The images can PIM (personal information manager)
usually be viewed on a computer with a Software that organizes personal informa-
CD drive and the appropriate software or tion, such as addresses, diary notes, and
by using a special player. appointments, in a useful manner.
photo illustration Computer art that PIN /pin/ See personal identification
starts with a digitized photograph that is number.
subsequently manipulated by image-
enhancement software to apply special ef- ping /ping/ (packet Internet groper) 1.
fects. A basic Internet program that verifies that
a particular Internet address exists and can
PHP (Hypertext Preprocessor) An OPEN accept requests. It sends a special signal
SOURCE SCRIPTING LANGUAGE widely used message to the address that the host echoes
for SERVER-SIDE SCRIPTING. First released in back. Ping informs the user not only
1995, its syntax is closely based on C, JAVA, whether the host is available but also how
and PERL. long the signal took to return.
2. The process of sending a message to all
physical record See block; record. the members of a mailing list requesting an
acknowledgment (ACK). It is used in e-
pico- /pÿ-koh-/ A prefix to a unit, indi- mail systems to test which users are cur-
cating a submultiple of one million mil- rent.
lionth (i.e. 10–12) of that unit, as in
picosecond. The symbol p is used for pico- pin header A device similar in form to a
; picosecond can thus be written ps. Pico- is DIP but containing no circuitry. Instead
part of a sequence each leg is extended vertically through the
milli-, micro-, nano-, pico-, … package allowing pins to be connected to-
of decreasing powers of 1000. gether in any configuration by soldering
169
pipe
small pieces of wire across the relevant clude ‘softlifting’ (users sharing software),
pins. A pin header is more flexible but hard-disk loading (the installation of unau-
clumsier than a DIL SWITCH. thorized copies of software on computers
for sale), counterfeiting (making counter-
pipe See pipeline processing. feit copies), and on-line piracy.
170
plug and play
plasma display (plasma panel display) up and down the length of the sheet of
See display. paper or film. The pen can therefore be
moved to any specified point on the surface
platen /plat-ĕn/ The cylinder around of the paper or film to make a mark. It can
which the paper is wrapped in most impact either touch the surface as it moves so that
printers. it makes a line, or it can be lifted off the
surface as it moves.
platform 1. The underlying hardware or In the drum plotter the bar is fixed in
software system on which application pro- position and it is the paper that is moved.
grams can run. Thus one speaks, for exam- The bar is parallel to the axis of a drum.
ple, of a Windows platform as opposed to The paper is wrapped around part of the
a Macintosh platform. surface of the drum; the holes down each
2. Any base technology on which other side of the paper engage with pegs on the
technologies are built. In a computer, drum. Rotating movements of the drum
which is a layered device, the layer at the cause the paper to move, and different po-
bottom is often referred to as a platform. sitions down the length of the paper can be
Once a platform has been defined develop- brought precisely under the pen. As in the
ers can produce appropriate software. flatbed plotter the pen can be moved along
the bar to the required position across the
platform for Internet content selec- width of the paper, and it can be lifted off
tion See PICS. the surface of the paper if necessary. The
drum plotter is used when large numbers of
platform-independent Describing drawings are required; these can be plotted
software that is independent of operating in sequence down the paper and then split
system and hardware. up. Several pens can be mounted on the bar
of a flatbed or drum plotter to provide a se-
PL/I A high-level PROGRAMMING LAN- lection of colors, which can be changed by
GUAGE whose name is derived from Pro- computer control.
gramming Language I (one). It was To control the movement of the pen of
developed by IBM in the late 1960s to a plotter, the plotter is sent a stream of
combine what were considered the best commands from the computer. These tell it
features of FORTRAN, ALGOL, and COBOL, where to move the pen to and whether the
and was intended to replace all three. PL/I pen is to be up or down during the move-
is a very large comprehensive language, ment. Most plotters also have built-in
and although it was used fairly widely in hardware characters, i.e. the plotter will
IBM installations it was never taken up to draw a character at the current pen posi-
any large extent by other manufacturers. tion without being given the detailed
movement required to construct it. Some
plotter An OUTPUT DEVICE that converts plotters can also draw arcs, circles, and el-
coded information from a computer into lipses given only the start and end posi-
graphs or pictures drawn on paper or tions, radii, etc.
transparent film by means of one or more In recent years plotters have adopted
pens. There are many types of plotter, ink-jet technology, and indeed large ink-jet
varying in the accuracy, quality, and speed printers have superseded plotters in many
of production of the graphs or pictures, the situations.
size produced, and the use (if any) of color.
Two basic designs of plotter are the plug See connector.
flatbed plotter and the drum plotter (see il-
lustration overleaf). In both cases the pen is plug and play The ability of an operat-
mounted on a bar that spans the width of ing system to recognize new hardware that
the paper. The pen can be moved to precise has been attached to computer and to con-
positions along the bar. In the flatbed plot- figure the system automatically to use the
ter the bar can be moved on tracks running hardware. If a suitable device driver cannot
171
plug compatible
paper or film bar
fixed in position
track
pen
Flatbed plotter
pen rotatable
drum
paper
Drum plotter
be found the user will be prompted to in- format for compressed raster graphics files
stall one. that may eventually replace the widely
used GIF format. The PNG format was de-
plug compatible See compatibility. veloped to be patent-free by an Internet
committee and does not have the legal re-
plug-in A small program that can be in- strictions then associated with GIF. PNG
stalled as part of other software to increase uses lossless compression and a PNG file
the functionality. It is often used to refer to can be 10–30% more compressed than the
such programs that can be used as part of equivalent GIF file.
a Web browser, allowing animation, video,
and audio files embedded in HTML docu- podcasting The packaging of audio ma-
ments to be recognized. terial for download over the Internet and
playback on a computer or MP3 PLAYER.
PNG (portable network graphics) A file This technique was first widely used with
172
polling
music tracks but has been extended to WAN to which a user can connect with, for
other types of audio content. An item made example, a local telephone call. For exam-
available in this way is called a podcast, the ple, an ISP has a point of presence on the
act of doing so is to podcast, and a person Internet.
or organization that does so is a podcaster.
Many radio broadcasters are making their point of sale (POS) The place where
programs available as podcasts and are ex- goods are paid for in a store, etc., often
pecting podcasting to become an impor- with a POINT-OF-SALE TERMINAL linked to a
tant distribution method. However, pod- computerized transaction system.
casting is not confined to established media
companies: anybody with appropriate point-of-sale terminal (POS terminal)
hardware and software – both of which are A specialized cash register, credit-card
cheap and widely available – can create an recording system, or ticket dispenser that
MP3 file, post it on the Web, and advertise records the details relating to the sale of
its content through a WEB FEED. The term goods, generally using a scanner to read
‘podcasting’ was first used in 2005 and is a bar codes or tags, and feeds the informa-
blend of ‘iPod’ and ‘broadcasting’ – a trib- tion into a central computer. This system
ute to the IPOD’s iconic status as the arche- improves stock, cash, and credit control.
typal MP3 player, although neither an iPod
nor any other MP3 player is essential to lis- point-to-point protocol (PPP) A pro-
ten to podcasts. It is expected that podcasts tocol used by machines that support serial
will soon encompass video as well as audio interfaces, such as modems and ADSL
content. transceivers. PPP is commonly used to pro-
vide and access dial-up services to Internet
point 1. In typography, a unit equal to service providers. PPP can share a line with
just under 1/72 of an inch that is used to other users, and unlike SLIP it can handle
measure the height of a character and the asynchronous as well as synchronous com-
distance between lines of text (leading). In munication and has error detection in-
computer DTP software, fonts, and page cluded.
description languages, a point is usually re-
garded as equalling exactly 1/72 of an inch. policy In policy-based networking, a set
2. A single pixel on a screen, which is iden- of definitions showing how the network’s
tified by its row and column numbers. resources are to be allocated among its
users. Parameters for allocation, which can
point-and-click To select data or an be static or dynamic, include items such as
object on a screen by using a mouse (or time of day, client priority, and resource
similar device) to move the cursor to the availability. Policies are created by net-
object (point) and then pressing a button work managers and used by network-man-
(click) on the mouse to select the object. agement software to make decisions.
The term applies to any pointing device. Policies can also be used to control how
much freedom networked users are permit-
pointer (link) A character or group of ted in other areas, for example to what ex-
characters that indicates the LOCATION of tent they may customize their DESKTOP.
an item of data in memory.
polling A technique whereby a com-
pointing device Any device that identi- puter checks each of a number of input de-
fies a point on a VDU and transmits its lo- vices in rotation to see if any data is
cation to a computer. Examples are waiting to be read. Polling depends for its
CURSOR KEYS, a MOUSE, JOYSTICK, TRACKER- success on each device being checked at
BALL, TOUCH-SENSITIVE DEVICE, or DIGITIZ- least as frequently as the data is arriving; in
ING PAD. addition the time spent processing any data
that has arrived must not be too long so
point of presence (POP) A point in a that data on other devices is missed. Polling
173
pop
can be contrasted with the use of INTER- program. Portable programs are different
RUPTS, where it is the devices that inform from PLATFORM-INDEPENDENT programs in
the processor when they have data avail- that the compiled versions of the latter will
able. In some systems an interrupt will alert run on any supported platform unchanged,
the processor to the fact that one of a num- while a portable program will at least be
ber of devices has data available, and the recompiled for the new platform. See also
processor then polls the devices to find out configure.
which one it is.
portable computer A small computer
pop See stack. that is designed for ease of transportation.
LAPTOPS, NOTEBOOKS, and PALMTOPS are all
POP /pop/ See point of presence. types of portable computer.
POP3 /pop-three/ (Post Office Protocol 3) portable document file (PDF) See Ac-
The version of the Post Office Protocol cur- robat.
rently in use on TCP/IP networks. Most e-
mail applications use this protocol, portable network graphics See PNG.
although some use IMAP (Internet mes-
sage access protocol). portal A gateway to the Internet pro-
vided by a Web site. There are general and
pop-up menu In a graphical user inter- specialized portals. General portals include
face, a menu that appears on screen when Yahoo!, MSN, Netscape, Excite, and
a user performs a certain action and usu- AOL.
ally disappears when the user selects one of
the menu items. portrait The arrangement of printed
matter on a page so that the top of the page
port 1. A point at which a connection is on the shorter side. Compare landscape.
can be made between an input/output de-
vice and the central processor of a com- POS /poss, pee-oh-ess/ See point of sale.
puter system, allowing data to be passed.
On personal computers ports usually have positional notation See number sys-
particular types of socket for connecting tem.
the mouse, keyboard, printer, etc.
2. A point through which data can enter or positive acknowledgment See ac-
leave a computer NETWORK. It forms part knowledgment.
of a NODE.
3. To move a program, usually written in a POSIX /poz-iks/ (portable operating sys-
PORTABLE form, to another computer sys- tem interface for UNIX) A defined set of
tem. operating system interfaces, agreed by the
IEEE, based on the UNIX operating sys-
portable Denoting programs that can be tem.
readily transferred from one computer to
other computers. For maximum portability post To publish a message in an on-line
a program should be written in a HIGH- forum or newsgroup.
LEVEL LANGUAGE possessing an internation-
ally agreed standard, and any extensions or POST /pohst/ (power-on self-test) The
deviations permitted by a particular COM- first task performed by the BIOS when a
PILER should be avoided. Where it is ab- PC is switched on. It checks that all the
solutely necessary to make some reference basic hardware components, such as the
to particular hardware or to include sys- keyboard and display, are running and that
tem-dependent features, then these should the CPU and memory are functioning
be concentrated in a few well-documented properly. If an error is found a message will
routines and not scattered throughout the be displayed or if the monitor is faulty a se-
174
preemptive multitasking
ries of beeps will be heard. After a success- and run visual presentations. Developed by
ful POST, control is passed to the system’s Forethought, Inc., it was first released for
bootstrap. the Macintosh in 1987, then purchased by
Microsoft and released for Windows in
Post Office Protocol 3 See POP3. 1988. It has been upgraded several times
and the current version is PowerPoint
PostScript A page description language 2003. PowerPoint is very widely used for
introduced by Adobe in 1985. PostScript business and other presentations and forms
has a series of instructions for drawing part of Microsoft OFFICE.
graphics and placing text on a page. The
FONTS used (known as type 1 fonts) are de- PPP See point-to-point.
fined as outline fonts and can be embedded
in the PostScript file. PostScript is an inter- PRAM /pee-ram/ (parameter RAM) A
preted language; the interpreter, known as battery-operated form of RAM used in
a RIP (raster image processor), is present in some Macintosh computers in which infor-
all PostScript printers or imagesetters. mation such as date, time, and computer
Most applications, especially word-pro- configuration are stored. Occasionally
cessing or desktop-publishing applications, PRAM gets corrupted and needs to be re-
are able to generate PostScript output. See stored.
also Acrobat.
precedence See order of precedence.
POTS (plain old telephone service) The
basic public switched telephone network precision The degree of exactness to
with no additions. The main differences which a numerical value is stated, usually
between POTS and non-POTS services,
based on the number of digits to which the
such as ISDN and ADSL, are bandwidth
number is represented. The value of some
and speed.
quantity given to 6 digits thus has 6-digit
precision. For example, π to 6-digit preci-
power-fail recovery See recovery.
sion and to 8-digit precision is respectively
3.141 59 and 3.141 5927
Power Macintosh A Macintosh com-
In a computer each item of information
puter based on the PowerPC chip. The
6100/60, 7100/66, and 8100/70 were must fit into a WORD, which contains a
available in 1994. The latest generation, fixed number of bits. This sets a limit on
the G5, has been released in various ver- the precision to which a number can be
sions since 2003. See also iMac. represented.
Many programming languages offer
power on To switch on a piece of com- extended precision (for FLOATING-POINT
puting equipment. NUMBERS). More powerful computers will
perform extended-precision floating-point
power-on self-test See POST. arithmetic in hardware. In the simplest
case two words can be operated as a single
PowerPC A RISC-based microproces- double-length word; this doubles the num-
sor open architecture developed jointly by ber of bits available to represent a number,
IBM, Apple, and Motorola. The name is and leads to what is known as double pre-
derived from Performance Optimization cision. (The use of only one word is then
with Enhanced RISC. PowerPC offers an called single precision.) The increased
alternative to Intel architectures and is number of bits is normally added to the
used primarily in some IBM machines and mantissa section of the word, leaving the
in Apple Macintosh computers. exponent section unchanged in length.
175
presentation graphics
presentation graphics The use of such with the circuit design; the metal strips are
techniques as bar or line charts and pie conducting tracks. The components, usu-
graphs to present information. Presenta- ally INTEGRATED CIRCUITS, are then soldered
tion-graphics applications usually produce into position between the appropriate con-
high-quality output used principally for ducting tracks to complete the circuit.
presentations on projectors for lectures, Double-sided PCBs are commonly pro-
meetings, etc. duced; these have components and some of
the conducting tracks on one side of the
presentation layer The sixth of the board with additional tracks on the other,
seven layers in the OSI communication ref- interconnections being made through the
erence model, which ensures that the com- board.
munications passing through are in the A PCB connects via an appropriate
appropriate form for the recipient. socket to the internal wiring of, say, a com-
puter system. Smaller modular PCBs may
preventative maintenance MAINTE- be connected to a PCB to augment its func-
NANCE that is performed on a regular basis tion. See also expansion card; edge connec-
and is intended to prevent failures or to de- tor; motherboard.
tect failures that are just about to happen.
When the performance of, say, an electric printer An OUTPUT DEVICE that converts
circuit is suspect, the circuit can be checked the coded information from a computer
by deliberately trying to induce a failure. into a readable form printed on paper.
Compare corrective maintenance. There are many types of printer, varying in
the method and speed of printing and the
preview To display a document in its quality of the print.
printed form on the screen before sending Some printers print a single character at
it to the printer. a time and are thus often referred to as
CHARACTER PRINTERS; the DOT MATRIX
primary color One of a set of three or PRINTER is an example. There are also LINE
four colors that can be mixed to produced PRINTERS, which produce a complete line of
many other colors. There are two sets of characters at a time. Page printers, such as
primary colors: the RGB SYSTEM - red, green, the LASER PRINTER, produce a complete
and blue - used in such video equipment as page at a time.
televisions and computer monitors; and the Printers can produce their printed char-
CMYK SYSTEM - cyan, magenta, yellow, and acters either by mechanical impact or by
black - used for color printing. nonimpact; in some devices different colors
can be printed. With impact printers a
primary store See main store. character is produced by a hammer action
causing an ink ribbon to press against the
print To output information from a paper. In some cases the printed characters
computer to a printer. Application pro- have a solid form, being created by an em-
grams can also output printed information bossed character struck against the ribbon;
to a file for printing at a later date, rather the band printer and barrel printer are ex-
than directly to the printer. This is known amples of this type of impact printer. Al-
as printing to file. ternatively, characters may be built up as a
pattern of closely spaced (or possibly over-
PRINT See Basic. lapping) dots, each dot produced by the
impact of a thin rod. In nonimpact printers
printed circuit board (PCB; pcb) A characters are produced without the use of
thin rigid board of insulating material, usu- mechanical impact; examples include INK-
ally fiberglass, with electronic components JET, THERMAL, and laser printers. The char-
and interconnections on one or both sides. acters are usually composed of fine dots.
A specific pattern of metal strips is chemi- Printers – either impact or nonimpact –
cally formed on the board in accordance that produce characters as a pattern of dots
176
print spooler
DIPs (containing
integrated circuits)
conducting tracks
are known as matrix printers. Matrix the printed matter itself or the paper sheets
printers can produce a large selection of or stack of CONTINUOUS STATIONERY on
shapes and styles of letters and digits, and which the matter is printed.
may also print Arabic characters or the
ideograms of oriental languages. In addi- print queue Part of a computer’s mem-
tion to characters it is possible to produce ory set aside as a buffer, to hold documents
diagrams, graphs, etc. waiting to be printed. When the printer is
ready the next document will be taken
printer control language See PCL. from the queue.
printer format See format. print screen key A key on PC and com-
patible keyboards that, when depressed,
printer port A port to which a printer will place a raster graphics image of the
can be connected. On PCs this is usually a screen on the Windows clipboard. See also
parallel port and is given the logical name screen dump.
LPT by the operating system. Serial ports
can also be used to connect printers (logi- print server On a network, a SERVER
cal device name COM), but they often re- that makes one or more printers connected
quire configuration. Modern printers to it available to network users. Print
increasingly connect to a computer via a servers are becoming less common as print-
USB port. On Macintosh computers ers increasingly act as their own servers
printer ports were formerly serial and are and connect directly to the network.
now USB.
print spooler Software that takes docu-
printout The output of a PRINTER, either ments to be printed and stores them tem-
177
privacy
178
program
distinct processes that run simultaneously cessing unit. A processor may form part of
and independently with their own re- a unit, as with an intelligent TERMINAL. The
sources (memory, files, etc.), communicat- words processor and computer are often
ing only to coordinate their activities. See used as synonyms. See also front-end
also multitasking. processor; microprocessor.
179
program counter
2. To design, write, test, and debug pro- tinction is often drawn between systems
grams. analysts and programmers. See systems
analysis.
program counter See instruction ad-
dress register. programming error See error.
180
PSTN
are added after rather than during manu- tion. These would typically be set at the be-
facture and cannot be altered from that ginning of the program, when the user
time. These contents are fixed electroni- chooses the desired type of automobile,
cally by a device known as a PROM pro- and remain constant thereafter. Software
grammer. It is only outside a computer that components usually have properties whose
a PROM can be programmed. Within a values are set when the program is de-
computer the contents cannot be changed; signed. A clock CONTROL, for example,
they can only be read. See also EPROM. might have properties determining whether
it displays the time in 24-hour or 12-hour
PROM programmer See PROM; format, in analog or digital format, etc.
EPROM.
protection See data protection; file pro-
prompt A short message displayed on a tection; storage protection.
VDU by a COMMAND LINE INTERFACE to in-
dicate that additional data is required from protocol /proh-tŏ-kôl/ An agreement
the user to allow processing to continue. that governs the procedures used to ex-
The prompt frequently has some mne- change information between entities in a
monic content to describe the type of data computer NETWORK. In general, a protocol
expected. Ideally a prompt should be ex- will govern the way in which information
plicit and informative for those unused to is encoded (see code), the generation of
the system, but capable of being shortened checking information (see check), and the
to increase the speed of use for an experi- FLOW CONTROL of information, as well as
enced user. For example, a novice prompt actions to be taken in the event of ERRORS.
may take the form
Enter the name of a file, proxy server An intermediate server be-
1 to 8 characters starting tween a LAN and another network, espe-
with a letter: cially the Internet, that provides security
The equivalent expert prompt could then (e.g. a FIREWALL), administrative control,
be and caching facilities. Performance can be
Filename: improved with a proxy server as its cache
This state of affairs is seldom realized, can serve all users and supply frequently re-
and many operating systems in particular quested data, such as popular Web pages.
prompt with a single character such as >
or $. pseudocode /sew-doh-kohd/ A pro-
gramlike but informal notation consisting
property A data item that forms part of of text in natural language (e.g. English)
a CLASS or software component. Assigning and used to describe the functioning of a
different property values to different in- procedure or a program. The flow of con-
stances of the class or component can alter trol is usually expressed in programming
its behavior to suit the precise circum- terms, e.g.
stances. Some properties are transient, if … then … else …
holding temporary values generated during repeat … until …
the execution of a program. A games pro- while the actions are elaborated in prose.
gram, for example, might contain a class Pseudocode is used mainly as a design aid
called ‘Auto’ that models the characteris- and is an alternative to FLOWCHARTS.
tics of an automobile, with such properties
as ‘current speed’, ‘fuel remaining’, etc. pseudorandom numbers /sew-doh-ran-
These would change in response to user dŏm/ See random numbers.
input and elapsed game time. Other prop-
erties might define more fundamental char- PSN (packet switching network) See
acteristics of the particular type of packet switching.
automobile being modeled, such as maxi-
mum speed and maximum rate of accelera- PSTN (public switched telephone net-
181
public-domain
work) The worldwide public telephone down from the menu bar and remains open
system. The PSTN provides much of the In- until either another menu is pulled down or
ternet’s long-distance infrastructure. ISPs it is closed by the user.
pay long-distance providers for access and
share circuits among many users by packet punched card An obsolete data MEDIUM,
switching. Thus Internet users only pay for consisting of a rectangular paper card on
usage to their ISPs and for local telephone which data could be encoded in the form of
calls. holes punched in columns. Since a hole
could either be present or absent, it could
public-domain Describing any work be used to represent a binary 1 or a binary
that is not copyrighted. Such work can be 0. A pattern of holes and spaces could
copied freely by anyone and used for any therefore represent a sequence of 1s and 0s
purpose. Much of the information on the that in turn represented a character (see bi-
Internet is of this form but note that copy- nary representation). Punched cards were
righted information on the Internet is not employed extensively for the input, output,
in the public domain. and storage of data in early computers.
They were in fact used for data processing
public-domain software Uncopyrighted before computers were available.
software available for anyone to use. See
Each letter, number, or other character
also freeware; shareware.
was represented on a card by holes pro-
duced mechanically – by a card punch – in
public key One of two keys used in PUB-
one or more rows within a column. When
LIC-KEY ENCRYPTION. A value is provided by
some designated authority (the user) and used for the input of data, the card punch
released to the public as a key that can be was operated by means of a keyboard; it
used to encrypt messages to be sent to the was then known as a keypunch. The pres-
user or for decrypting the user’s DIGITAL ence of holes in a column, and their posi-
SIGNATURE. Public keys are usually very tions, was sensed (photoelectrically) by
large integers. another machine – a card reader. The
reader converted the punched data into a
public-key encryption The use of dif- binary coding of 1s and 0s that could be
ferent PUBLIC and PRIVATE KEYS for encryp- fed into a computer. Binary data from a
tion, known as asymmetric cryptography. computer could be punched on to cards –
A public key encrypts data and a corre- again by a card punch – as output.
sponding secret private key decrypts data.
The process is reversed for digital signa- purge To clean up, usually by removing
tures. old or redundant data or files.
182
Q
QBE (query by example) A simple data- be used to store requests for access to pe-
base querying technique in which the crite- ripherals, such as disks or printers.
ria are set in a blank database record. This
method is easier than using a query lan- QuickDraw Trademark A graphics dis-
guage. play system for Apple Macintosh comput-
ers that enables programs to create and
quadbit /kwod-bit/ See nibble. display graphical objects. It has been su-
perseded but is still widely used.
quantum computing /kwon-tŭm/ Re-
search into the use of quantum mechanical QuickTime Trademark A standard de-
effects as the basis of computer processors. veloped by Apple Macintosh for displaying
multimedia on computers. QuickTime is
query by example See QBE. built into the Macintosh operating system.
PCs can also run files in QuickTime format
query language A specialized language if they have the appropriate software.
that is used for obtaining information from QuickTime files usually have the file exten-
a database. sion mov.
query processing The retrieval of one quit To exit from a program or system in
or more values from a FILE or DATABASE, a clean and orderly way.
leaving the contents unchanged.
QWERTY /kwer-tee/ Describing the
queue (pushup list) A LIST that is con- standard form of a Latin-alphabet-based
structed and maintained in computer mem- keyboard, laid out in a typewriter style, in
ory so that all insertions are made at one which the first six keys on the upper left of
end of the list and all removals are made at the alphabetic keys are QWERTY. It orig-
the other end. The next item to be removed inated in the 1880s to reduce the jamming
is thus the earliest to have been inserted, of type bars in typewriters by fast typists.
i.e. the queue works on a first in first out Various alternatives have been pro-
(FIFO) basis. In contrast, a STACK works on pounded but QWERTY is so entrenched
a last in first out (LIFO) basis. Queues may that these have had little impact.
183
R
radio button One of a set of small icons no redundancy at all, but is very fast; it is
used to select an option in a graphical user sometimes considered not to be a proper
interface. Typically radio buttons are small form of RAID storage. RAID disk systems
circles. Only one button in any set can be are becoming standard on all but the small-
selected at any one time. est FILE SERVERS.
184
random access
RAID CONFIGURATIONS
ical drive. When the computer is switched other stored data to be read first. The stor-
off all the data is lost. This system has been age locations can be accessed (read or writ-
superseded by the use of cache memory. ten to) in any order. There is random
access with MAGNETIC DISK storage. There is
random access A method of retrieval or also random access with main storage, i.e.
storage of data that does not require any with RAM and ROM, but with a much
185
random-access file
shorter ACCESS TIME than with disk storage. RDF Site Summary See RSS.
See also file. Compare serial access.
read /reed/ 1. To sense and retrieve data
random-access file See file. from a storage medium such as MAGNETIC
DISK or SEMICONDUCTOR MEMORY. This
random-access memory See RAM. data can then be transferred to another
storage medium or to an output medium.
random numbers Numbers that are See also readout.
drawn from a set of permissible numbers 2. To sense and interpret data on an input
and that have no detectable pattern or bias. medium such as a document or a BAR CODE,
Any number in the set should therefore and to convert the data into an electrical
have an equal probability of being selected. signal that can be fed into a computer.
Random numbers are required, for exam-
ple, when selecting the winners of a lottery. READ /reed/ See Basic.
In a computer true random numbers are
difficult to obtain. Instead programs are reader An INPUT DEVICE, such as a DOCU-
designed to generate what are known as MENT READER or bar-code reader, that is
pseudorandom numbers. In principle the able to READ data, i.e. sense and interpret
numbers produced depend on their prede- data, on an input medium.
cessors and so are not truly random; in
practice the pseudorandom numbers gen- read head /reed hed/ See tape unit.
erated by a particular program are suffi-
ciently random for the purpose intended. read instruction A MACHINE INSTRUC-
TION that causes an item of data to be READ
raster graphics /rass-ter/ A method of from a specified location in a memory de-
producing pictorial images in which the de- vice and copied into a BUFFER store or into
sired shape is built up line by line. Each line a REGISTER.
is composed of closely spaced elements
that can be any one of a number of colors read-only memory See ROM.
or shades. This is the technique used in
most VDUs, in TV monitors, and also in readout 1. Information retrieved from
some plotters. The term is also used of a MAIN STORE after processing and displayed
type of COMPUTER GRAPHICS image that uses on a screen or copied into backing store.
these principles. Compare vector graphics. 2. The process of reading data from a
memory device. In most types of memory
raster image processor See RIP. this does not alter the actual representation
of the data; the readout is then described as
raw data Data in the form in which it nondestructive.
reaches a computer system from the out-
side world. It is data that has not been vet- read/write head See head. See also disk
ted for correctness, nor sorted into any drive.
particular order, nor processed in any
other way. See also data cleaning. read/write memory A type of memory
that in normal operation allows the user to
ray tracing A technique used mainly in read from or write to individual storage lo-
three-dimensional graphics for showing cations within the memory, i.e. to retrieve
objects with realistic shadows cast by an and store information. It may involve SER-
imaginary light source. In ray-tracing ap- IAL ACCESS or RANDOM ACCESS. RAM is
plications primitive objects (e.g. cubes, read/write memory.
cylinders, spheres, etc.) can be placed in
position, sized, distorted, rendered (see real address The absolute machine ad-
rendering), and illuminated and viewed dress that specifies a physical memory lo-
from different directions. cation.
186
record
187
record length
maintained that records their starting posi- wise there would be an endless series of
tions or lengths. calls. Recursion is not allowed in every
This collection of related data items is programming language. Improper use can
considered in terms of its contents or func- cause problems during program execution,
tion rather than on how it is stored and being difficult to detect during compila-
moved in a computer. It is often referred to tion.
as a logical record. In contrast, a physical
record is the unit in which data is trans- reduced instruction set computing See
ferred to and from peripheral devices; it is RISC.
also called a BLOCK. One logical record
may be split up and held in different phys- redundancy The provision of addi-
ical records. Several logical records or tional components in a system, over and
parts of logical records may be located in above the minimum set of components re-
one physical record. quired to perform the function of the sys-
The record is one of the fundamental tem. This is done to increase the reliability
ways of collecting and organizing data. of the system and, in the event of errors oc-
DATA FILES held in backing store are often curring, to increase its ability to recover
treated as sequences of records. Many pro- from such situations. For example, two or
gramming languages permit operations on three identical pieces of hardware may be
an entire record as well as on its individual used in a computer so that if a fault occurs
components. See also database. in one piece that item can be replaced.
Again, additional bits are attached to data
record length The storage space needed when it is transferred from one location to
to contain a record, usually stated in bytes. another, enabling a CHECK to be performed
– a redundancy check – and any errors oc-
recoverable error An error that is not curring in the data during transfer to be de-
fatal, i.e. one from which the program can tected.
recover and continue running.
reentrant program /ree-en-trănt/ A pro-
recovery The procedure whereby nor- gram that uses code that can be shared by
mal operation is restored after the occur- other programs at the same time. Operat-
rence of a FAULT. Power-fail recovery, for ing systems often use reentrant code so that
example, deals with a loss of the incoming only one copy is needed in memory to ser-
power supply to a computer system. The vice many live applications.
system is equipped to detect any long-term
deviation of the power-supply voltage reference file A FILE that contains refer-
from acceptable limits. When such a devia- ence material and thus changes infre-
tion is detected, and before it can cause any quently.
problems, a signal (INTERRUPT) is sent to
the processor: the sequence of instructions reformat To FORMAT a disk that con-
being carried out is suspended, relevant in- tains data. Reformatting will totally re-
formation is dumped in NONVOLATILE move any data on the disk.
MEMORY, and processor activity is halted.
When the voltage is restored the program refresh 1. To recharge a dynamic mem-
or programs that had been running can be ory bank so that information is not lost.
RESTARTED. This is an automatic function of the mem-
ory board. See also dynamic RAM; refresh
recursive routine A PROCEDURE, FUNC- cycle.
TION, or SUBROUTINE that CALLS itself, re- 2. To retrace a video screen after a certain
sulting in a continued repetition of the interval. See also refreshed display; refresh
operation that the procedure, function, or rate.
subroutine is designed to perform. The re-
cursive call must be CONDITIONAL, other- refresh cycle The repeated cycle of elec-
188
regular expression
trical pulses used to renew stored electrical particular REGISTERS or between a specified
charges in DYNAMIC RAM, without which storage LOCATION in main store and a par-
data would be lost. ticular register. For example, loading the
contents of address 10 in main store into
refreshed display A display in which the ACCUMULATOR would involve
the image lasts only a very short time and (1) loading address 10 into the MEMORY
therefore has to be repeatedly redrawn ADDRESS REGISTER,
electronically. An ordinary TV screen is a (2) copying contents at address 10 to
refreshed display. If the same image is re- accumulator.
drawn on a refreshed display then the This can be written in symbolic form:
display appears continuous, without flick- (1) 10 ⇒ MAR
ering. Alternatively different information (2) MEM ⇒ ACC
can be fed to the screen to give the where the arrow symbol means ‘is trans-
appearance of motion in the image. A re- ferred to’.
freshed display can be used for instance to
produce an animated cartoon. registry A database used by Windows
systems to hold hardware and software
refresh rate The number of times a dis- configuration information. When a new
play’s image is redisplayed per second. piece of a program is installed it usually
This is expressed in hertz. The refresh rate writes entries in the registry. When they
for each screen depends on the video card are no longer required they should be un-
used and can be changed in the display set- installed in a proper manner. In this way
tings. However, if an incompatible rate is
the registry entries will also be removed.
chosen the display will go blank or the
The registry can be edited by using the
image will be distorted. The older and
REGEDIT.EXE program provided with
slower refresh rate of 60 Hz caused image
Windows. However care must be taken
flickering and consequently operator tired-
when doing this as the system can be dis-
ness and headaches. Newer rates of 75 to
abled if the registry is altered incorrectly.
85 Hz are commonly used.
register A storage device in the process- regression testing The testing of a com-
ing unit of a computer in which data has to plete program after modification of some
be placed before it can be operated upon by part rather than only testing the modified
instructions. It is therefore a temporary lo- routines, to ensure that the program still
cation, and has a CAPACITY of only a WORD, works in a proper manner.
a BYTE, or a BIT. The registers within a pro-
cessing unit are quite separate from the regular expression A compact method
storage LOCATIONS in main store and back- of specifying more than one character
ing store. Registers are either general-pur- STRING that is often used in searching. For
pose devices or are intended for a special example the regular expression (or pattern)
purpose. Special-purpose registers include ‘abc[def]g’ represents the three strings
the INSTRUCTION ADDRESS REGISTER and IN- ‘abcdg’, ‘abceg’, ‘abcfg’: enclosing the
STRUCTION REGISTER in the CONTROL UNIT. characters ‘def’ in square brackets means
In larger computer systems there may be that they are alternative values for the
several hundred registers. fourth character in the string. It is often
Information must be stored in and re- more useful to perform one search using a
trieved from registers extremely rapidly. regular expression defining a set of strings,
They are therefore high-speed semiconduc- which produces one set of results, than to
tor devices (usually a group of FLIP-FLOPS, search for each string separately. There is
each flip-flop storing one bit). no standard syntax for regular expressions
but some features, such as the use of square
register transfer An operation whereby brackets described above, are universally
information is transferred between two recognized.
189
relational database
190
resolution
remote access The ability to obtain ac- report generator See generator.
cess to a computer or network from a re-
mote location. request for comments (RFC) A formal
Internet document or standard that has
remote procedure call (RPC) A proto- evolved from drafts and reviews by inter-
col using the client/server model that can ested parties. Some RFCs are for informa-
be used by a program to request a service, tion only.
usually a PROCEDURE or FUNCTION call,
from another program located on another requirements description See program.
computer in a network.
rerun 1. To RUN a program again from
rendering The process of converting a the start, usually following a malfunction
wireframe three-dimensional image into a of the computer. See also restart.
solid image. Three-dimensional drawing 2. A repeat run.
programs have a wide range of options for
‘covering’ wireframe structures with differ- reserved character A keyboard charac-
ent textures and colors. See also ray trac- ter, for example an asterisk (*), forward
ing. slash (/ ), or backward slash ( \), that has a
special meaning to a program and cannot
repaginate /ree-paj-ă-nayt/ To rearrange be used for other purposes (such as naming
the pages of a desktop or word-processing files or keyboard shortcuts).
document after editing, to take account of
inserted and deleted data. This is usually an reserved word A word that has a spe-
automatic procedure in the application, cific role in a programming language and
which the user can override if necessary. cannot therefore be used as an IDENTIFIER.
For example, in many languages the words
repeater A device that receives an elec- if, then, and else are used to separate the
tromagnetically or optically transmitted different parts of conditional statements.
signal, amplifies it, and then retransmits it These are thus reserved words.
along the next leg of the medium. Signals
can be sent over long distances by the use reset (restore) 1. To cause a device to
of repeaters. They are used to interconnect take up an earlier value. For example,
segments in LANs and to extend WAN when a program is about to be RESTARTED
transmission. on a processor, the contents of the working
registers of the processor must be reset to
repeat until loop See loop. the values they last held when the program
was previously running.
replace To substitute one set of data for 2. To set the whole computer back to its
another: often used in word processing in initial condition when it was switched on.
search and replace operations. This process is used in smaller computers
to clear faults that have corrupted the OP-
replication The process of creating and ERATING SYSTEM and made further process-
managing duplicate versions of a database. ing impossible.
Changes made in one copy are reflected in
all other copies. These changes are syn- resident Describing a program that is
chronized so that many users can work on present in memory.
their own local copy of the database in ge-
ographically dispersed situations, but ap- resolution /rez-ŏ-loo-shŏn/ In general,
191
resource
the amount of information or detail that 2. The sharing of RESOURCES among the
can be produced by a device such as a VDU users of a computer system to prevent or
or plotter or can be revealed in an image. reduce the unfair use of the system by some
The device or image can be described qual- users at the expense of others. It can be
itatively as having a high resolution (hi res) done, for example, by allocating units of
or a low resolution, etc. The resolution can processor time to each user, depending on
also be given in numerical terms. The reso- their status, work, etc., and by allocating
lution of a display screen, for example, can fewer units at peak time rather than, say, at
be expressed as the number of PIXELS (pic- night.
ture elements) that are available in the hor-
izontal and vertical directions of the resource information file format See
screen, e.g. 1024 by 768 pixels. The reso- RIFF.
lution of a graphics image can be given in
dots (i.e. pixels) per inch, e.g. 300 dpi. The resource sharing See resource alloca-
resolution of a plotter can be given as the tion.
smallest possible pen movement. The reso-
lution of a printer output can be in dots per response time Usually, the time that
inch, e.g. 2400 dpi. elapses from some action by a computer
user to the receipt of some response or
resource Any of the facilities that are feedback from the computer. With a per-
available in a computer system and that are sonal computer, with one user, the re-
required by a processing activity. The re- sponse time depends entirely on the size of
sources in any computer system include the the task to be performed; there is a fast re-
following devices: one or more PROCES-
sponse for simple tasks like displaying the
SORS, some form of memory – MAIN STORE
contents of a file on the screen and a slow
and BACKING STORE – in which to store the
response for, say, compiling a large pro-
instructions for the processor(s) and the
gram. When there are a large number of
data awaiting manipulation, and INPUT/
terminals linked to a mainframe or mini-
OUTPUT DEVICES through which instructions
computer the response time increases as
and data are fed to the computer and in-
the number of users increases; a lightly
formation is fed back to the outside world.
Program instructions and data are held on loaded mainframe responds extremely
backing store as FILES. Files are also re- rapidly. In a network, the response time de-
sources. pends on how heavily the server responsi-
Any facility required from the OPERAT- ble for servicing the requested function is
ING SYSTEM of a computer is regarded as a
loaded, and in extreme cases the loading of
resource. These include programs for the network itself. See also interactive.
scheduling and supervising the running of
users’ programs on the computer. restart The resumption of the execution
of a program after a temporary halt, using
resource allocation (resource sharing) data dumped at a particular point in the
1. The sharing of the RESOURCES of a com- program (see dump). The program is said
puter system between various processing to be restarted. Much of the data from the
tasks. A resource, such as the processor, previous RUN is therefore preserved, and a
that can only be used by one task at a time, restart is often described as a warm start.
is shared by allowing each task in turn a set In contrast, there is a cold start when a pro-
time period. In the case of storage devices, gram has to be rerun from the start. Some
for example magnetic disks, requests from programming languages allow the pro-
tasks are queued and serviced one at a grammer to specify the points at which a
time. Resource allocation is under the con- program can be restarted.
trol of the OPERATING SYSTEM of the com-
puter. restore See reset.
192
RISC
retrieve To locate and return data to the attribute of most VDUS, used to draw atten-
user or program requesting it. tion to a character. The character is dis-
played on the screen in the opposite
return 1. In programming, the transfer contrast to surrounding information. For
of control from a called routine or program instance, with a screen that normally has
back to the calling routine or program. dark characters on a bright background, a
2. See return key. character in reverse video will be a bright
character within a dark character-sized
RETURN See Basic. rectangle.
return key (enter key) A key on a KEY- RGB system A method of defining col-
BOARD that sends a carriage return charac- ors by the proportions of R (red), G
ter to the computer. It is often used to (green), and B (blue) present. See also
signal that the current line of typing is com- CMYK system.
plete and may be processed. In GUI inter-
faces it often initiates the default action of ribbon cable An electric cable in which
the currently selected object, e.g. a pro- a number of similar wires lie side by side in
gram is executed, a word-processing docu- a flat plastic strip and are electrically insu-
ment is opened in the appropriate word lated from each other. A ribbon cable pro-
processor, etc. vides a continuous path along which
electrical signals can be conveyed from one
reverse Polish (RP) A notation used in a point in a system to another. It can be used
computer to evaluate arithmetic expres- for PARALLEL TRANSMISSION of data.
sions. In RP the ARITHMETIC OPERATOR (e.g.
+, –, *, /) is placed after its operands rather Rich Site Summary See RSS.
than between them as in conventional no-
tation. The addition of two numbers, a and rich text format See RTF.
b, would thus be expressed as ab+.
Expressions in RP are scanned from left RIFF /riff/ (resource information file for-
to right and the values of the operands mat) A tagged file-format specification
stored until an operator is encountered. used to cover different types of multimedia
The operator is then applied to the last two files. Tags are used to identify data ele-
values and replaces them by a single result, ments in a file by length and by type. The
which may be used as a value for the next specification can be extended to cover new
operator, and so on. Hence kinds of data elements when necessary
ab–cd+* while still supporting older formats.
would be evaluated as
(a – b) * (c + d) right-justified See justify.
i.e. b would first be subtracted from a; c
and d would then be added together and ring network See network.
the sum would finally be multiplied by
(a – b). RIP (raster image processor) Hardware
The translation of conventional nota- or a combination of hardware and soft-
tion to RP is achieved by an algorithm. A ware that converts vector graphics images
programmer would not have to write ex- into raster graphics images. See also Post-
pressions in RP but a COMPILER producing Script.
machine code from a high-level language
might well generate RP for expression eval- RISC /risk/ (reduced instruction set com-
uation. puter) A design philosophy for CPUs. In
contrast to CISC, RISC processors have a
reverse video (inverse video) A display smaller and simpler INSTRUCTION SET; they
193
RMON
therefore operate more quickly, because it highly robust if it recovers and continues to
takes less time to decode and execute each operate despite numerous error conditions.
instruction, and can be manufactured more
cheaply. Also, research has shown that rogue value See terminator.
many of the missing CISC instructions
were rarely used in practice, and so were roll-out roll-in See swapping.
needless complications. The MACHINE CODE
of programs for RISC processors tends to ROM /rom/ (read-only memory) A type
be longer and more complex, because op- of SEMICONDUCTOR MEMORY that is fabri-
erations that are intrinsic to CISC proces- cated in a similar way to RAM but whose
sors must be performed as a series of contents are fixed during manufacture and
smaller operations. Elements of the RISC cannot therefore be modified. The contents
philosophy have been incorporated into can only be read. As with RAM, ROM is
modern CISC chips, which often work as composed of an array of cells, each of
RISC devices internally through MICROPRO- which can be identified uniquely by its po-
GRAMS. PowerPC chips are the most widely sition in the array. There is thus direct and
used RISC-based chips in computers. extremely rapid access to any cell in ROM,
i.e. there is RANDOM ACCESS to any of the
RMON /ar-mon/ (remote monitoring) A storage locations. Since the contents can-
network management protocol that allows not be altered, ROM is NONVOLATILE MEM-
a single workstation to gather network in- ORY. See also PROM; EPROM.
formation (e.g. usage statistics). The
routers, hubs and switches, and other de- ROM cartridge (ROM pack) A mod-
vices must support RMON for this to ule containing software that is perma-
work. nently stored as ROM. The module can
easily be plugged into and later removed
robot /roh-bot/ A programmable device from a computer, games console, etc. with-
that has been engineered to perform many out the integrated circuitry being handled.
simple repetitive tasks formerly done by
human labor. It consists of one or more ar- root directory The main directory in a
ticulated mechanical manipulators, typi- HIERARCHICAL FILE SYSTEM. It is the direc-
cally an arm and hand, that are linked to a tory that includes all other directories.
computer and have considerable freedom
of movement. In more advanced types ROT-13 /rot th’er-teen/ The encryption
there is also some form of sensory mecha- of a message by exchanging each of the let-
nism, such as a TV camera acting as an ar- ters in the first half of the alphabet with the
tificial eye. Most robots in current corresponding letters in the second half.
industrial use follow a fixed sequence of in- Thus A becomes N, B becomes O and so
structions but can be reprogrammed to on. Other characters are not changed.
carry out another task. They have little if ROT-13 is sometimes used to conceal sen-
any sensory capability. They are designed sitive parts of on-line documents, for ex-
to be flexible and transportable and can be ample offensive words, the ending of a
used, for example, for loading and unload- movie or book, etc.
ing machines, positioning and transferring
parts, and welding pieces of metal or plas- rotate See shift.
tic.
rounding The process applied to a num-
robustness A measure of the ability of a ber whereby its length is reduced to a spec-
system – hardware or software – to recover ified number of significant digits, the
from conditions caused by ERRORS. The er- nearest approximation to its true value
rors may be external to the system or may being taken. For example, rounding the
occur within the system itself. The system number 3.257 69 to 5, to 4, and to 3 digits
can be described as robust or possibly yields 3.2577, 3.258, and 3.26. The REGIS-
194
running head
TERS in a computer can hold only a fixed or between a computer and a printer or
number of binary digits (bits). When the re- some other peripheral device. Although su-
sult of an arithmetic operation requires perseded by the USB interface, RS232C
more than this fixed number of bits, either ports are still supplied as standard on most
rounding or TRUNCATION will take place. PCs.
Since the rounded value is an approxima-
tion to the true value of the number, a RSS (Really Simple Syndication; RDF site
rounding error is introduced; on average summary; Rich Site Summary) Several
this is less than the truncation error. Re- widely-used formats for WEB FEEDS. They
peated rounding can cause a build-up of use XML to provide a standardized concise
such errors and the programmer should try summary of an item on a Web site and a
to minimize this problem. link to the full item. RSS originated in 1999
and has since split into two incompatible
router /rowt-er, root-er/ A hardware de- series of formats; however, most RSS-
vice that routes data from a LAN to a tele- aware software can understand both of
phone line or another connection that them.
leads to another network. It also handles
errors and gathers network statistics. A RTF (rich text format) A file format for
router is connected to at least two net- saving word-processor files in ASCII form
works and decides which way to send each with additional information that retains
information packet according to its ‘under- the formatting, font information, color,
standing’ of the current state of those net-
etc., of the original document. RTF files are
works. Routers will only allow authorized
often used for transferring files between
machines to send data through a LAN so
different word-processing applications.
that private information is kept secure.
They have the file extension rtf.
routine /roo-teen/ A subroutine (in as-
RTS (request to send) A hardware signal
sembly language) or procedure or function
that is often used in serial communications
(in high-level language); the term is usually
and sent over an RS-232 connection re-
used in combinations, as in input routine,
recursive routine, diagnostic routine. questing permission to transmit to a de-
vice.
RP See reverse Polish.
rubber banding In computer graphics,
RPC See remote procedure call. the action of grabbing a point on an an-
chored line of a graphics object and pulling
RS-422, 423, 449 The standards adopted it to a new position to change the shape of
for serial communications over distances the object.
greater than 50 feet. RS-449 incorporates
RS-422 and RS-423. rule-based system See expert system.
195
run time
pears in the top margin of each page of a too small to be stored – it would be re-
document. See also header. placed by zero.
run time The period of time during run-time system A collection of PROCE-
which a program is actually being exe- DURES that support programs derived from
cuted, or the time taken for execution. It a high-level language at RUN-TIME, provid-
does not include the time during which the ing services such as STORAGE ALLOCATION,
program is loaded, or is compiled or as- INPUT/OUTPUT, DEBUGGING, etc., but are not
sembled. actually a part of the program code as
stored on disk. This avoids storing large
run-time error (execution error) An
numbers of copies of commonly used pro-
ERROR detected at RUN TIME. The error may
be fatal, in which case execution is termi- cedures, and allows new versions of them
nated; a fatal error would result, for in- to be installed without the need to recom-
stance, from an attempt to divide by zero pile all the programs that might use them.
or read past the end of a file. Alternatively In computer systems where a number of
the error may be nonfatal, in which case a high-level languages are supplied by a com-
warning is generated and execution contin- mon manufacturer, it is often possible for a
ues; a nonfatal error would result, say, single run-time system to service programs
from an attempt to calculate a real number written in any of these languages.
196
S
sampling A process by which the value clines unacceptably as users increase is said
of an analog (i.e. continuous) signal is ‘ex- to be not scalable or to scale badly. It is im-
amined’ at distinct intervals of time. The portant that such types of software as OP-
sampled values can, for example, be con- ERATING SYSTEMS, DATABASE MANAGEMENT
verted into a digital form using an A/D CON- SYSTEMS, and WEB SERVERS scale well. The
VERTER for subsequent processing by a term is also applied to other aspects of
computer. computing, for example hardware.
2. Denoting something that is designed to
SAN See storage area network. work equally well at more than one size.
For example, OUTLINE FONTS and vector
SATA (S-ATA) See Serial ATA. graphics are scaleable, whereas BITMAPPED
FONTS and raster graphics are not.
satellite computer A computer that is
attached to another computer by a com- scalable font Any font that can be
munications link and is controlled by the scaled to produce characters of different
main computer. sizes without any loss in quality. See also
outline font.
save To write data to a storage medium.
Scalable Link Interface See SLI.
SAX (Simple API for XML) A de facto
standard for software that processes an scale See scalable.
XML file in discreet units, contrasted with
software that constructs a representation scanner A device for capturing diagrams
of the whole file in the computer’s memory and photographs and converting them into
before beginning processing. SAX-based electrical signals, so that the image can be
programs are generally faster and use less stored in digital form. Scanners are also
memory but cannot perform some com- used as optical character readers. For very
plex types of processing. SAX was first re- high-quality work, as in the reproduction
leased in 1998 and the current version is of photographs for book and magazine
version 2 (SAX2). printing, drum scanners are used. In these
the image to be scanned is wrapped around
scalable 1. Denoting something that a cylindrical drum. FLATBED SCANNERS are
works as well at large sizes as it does at suitable for many applications, and HAND-
small sizes. A piece of software, for exam- HELD SCANNERS are also in use.
ple, will generally run very efficiently if it
only has a small number of simultaneous scheduling The method by which the
users. Software that can handle an unpre- use of a PROCESSOR is controlled, i.e. by
dictably large number of simultaneous which time on the processor is allocated.
users with a minimal decline in perfor- Scheduling is necessary in a system where
mance is said to be scalable, to scale well, the processor has to be shared between a
or simply to scale. Any software that either number of programs; this occurs for exam-
has a fixed, and relatively small, maximum ple in a MULTITASKING system. The choice
number of users or whose performance de- of which processing activity should be
197
schema
granted access to the processor, and can a variable or other element is normally in-
safely be granted access, is determined by dicated by the positioning of or a qualifica-
the scheduling algorithm. This is designed tion within its DECLARATION.
to ensure that the computer is used as effi-
ciently as possible. scrapbook A Macintosh file used to
hold text and graphics images for later use.
schema 1. In XML, a file that defines the See also clipboard.
structure of a document or a set of similar
documents. It details such matters as which scratchpad A temporary storage area in
ELEMENTS are allowed in the document, in a computer.
what order they must appear, how many of
each type can be used together, what type screen The front surface of a VDU or
of data they can contain (STRING, INTEGER, monitor on which computer text or graph-
FLOATING-POINT NUMBER, etc.). Documents ics can be displayed. See display.
that are correctly structured are said to
conform to their schema, and schemas can screen dump A way of transferring the
be used to create software that will cor- entire graphical or textual contents of a
rectly process any conforming document. VDU screen to a printer or to a raster
Schemas thus perform the same role as graphics file. See also dump; print screen
DTDS but in a more sophisticated and flexi- key.
ble way, and they use normal XML syntax
rather than the special syntax (inherited screen editor See text editor.
from SGML) used by DTDs. However, there
are currently several varieties of schema screen saver A utility program that dis-
whereas the syntax for DTDs is standard- plays a moving image on the screen after a
ized. set time interval if no keyboard or mouse
2. Any abstract model that describes the actions have been performed. On older
structure of data, operation of hardware or monitors this stopped the possibility of a
software, etc. permanent static image being burned onto
the screen. Modern monitors do not suffer
Schottky TTL /shot-kee/ See TTL. from this problem.
198
searching
199
secondary store
200
server
concerned with specifying the meaning of sequential flow See control structure.
various constructs – STATEMENTS, CONTROL
STRUCTURES, etc. These constructs must serial access (sequential access) A
conform to the SYNTAX, or grammar, of the method of retrieval or storage of data that
language; if they do not then they are must be used for example with magnetic
meaningless. Most programming lan- tape. Blocks of data are read from the stor-
guages have their semantics defined by de- age medium in the actual order in which
scriptions in English or some other natural they occur until the required item or stor-
language. Because natural language is im- age location is found. The ACCESS TIME thus
precise and hence ambiguous, different in- depends on the location of the item. Com-
terpretations of the semantics are possible pare random access.
in the early stages of development of a lan-
guage. Serial ATA (SATA; S-ATA) A develop-
ment of and the successor to the ATA disk
semiconductor memory /sem-ee-kŏn- INTERFACE standard. Developed in the early
duk-ter/ (solid-state memory) Any of var- 2000s and a formal specification since
ious types of compact memory composed 2003, it offers greater data-transfer speeds
of one or more INTEGRATED CIRCUITS fabri- by using a serial rather than a parallel BUS
cated in semiconductor material, usually and is now used in most new PCs.
silicon. The integrated circuit is composed
of an array of microscopic electronic de- serial port An interface used for SERIAL
vices, each of which can store one bit of TRANSMISSION. When used of microcom-
data – either a binary 1 or a binary 0. puters, the term denotes an RS232C INTER-
There may be millions of these storage lo- FACE, not a USB interface.
cations within a single integrated circuit
measuring only a few square millimeters in serial printer A printer that connects to
area. Data can be accessed extremely a SERIAL PORT on a computer. Formerly
rapidly from these locations. standard on Macintoshes and sometimes
The categories of semiconductor mem- used on PCs, they are being superseded by
ory include RAM (which is read/write mem- printers that connect to a USB port.
ory), ROM (read-only memory), PROM
(programmable ROM), and EPROM serial transmission A method of send-
(erasable PROM). There is RANDOM ACCESS ing data between devices, typically a com-
to all these types of semiconductor mem- puter and its peripherals, in which the
ory. Semiconductor RAM has been used to individual BITS making up a unit of data
build the MAIN STORE in computers since (e.g. a character) are transmitted one after
the early 1970s. another along the same path. For ex-
ample, if two devices wish to communicate
sentinel See flag. an 8-bit unit of data, the sending device
will transmit the 8 bits in sequence (per-
separator A symbol that separates haps with a start bit preceding them and a
STATEMENTS in a program. The symbol stop bit following to aid synchronization).
used depends on the programming lan- The receiving device can then reassemble
guage. A semicolon is the separator in sev- the stream of bits into the original 8-bit
eral languages, including Pascal: unit. See also parallel transmission; hand-
begin read (x); y := x*x; write (x,y) end shake.
Note that the final statement needs no
semicolon. Compare terminator. server A program that provides services
to other programs, called clients. Usually a
sequential access /si-kwen-shăl/ See ser- server and its clients run on different com-
ial access; file. puters attached to a network. There are
many types of server. For example, a file
sequential file See file. server – the original type of server – allows
201
server-based application
clients to access its computer’s hard disks stock market prices) and/or user selection
and read or store files; a database server of content (e.g. online shopping). See also
runs a DBMS and handles requests from ASP; PHP.
clients to query the database, update its
contents, etc.; a mail server implements an session A recording made on a CD-R or
e-mail system for the network, usually with CD-RW at a particular time. The session
an external connection to the Internet; and has lead-in and lead-out information on
so on. The computer running a server pro- the CD. Multisession recordings can be
gram is often dedicated, i.e. it does nothing made on a single CD at different times, one
else. It is called the host, the server com- session following on from another.
puter, or, simply, the server.
Session Initiation Protocol See SIP.
server-based application An applica-
tion program that is installed on a server set-top box See Web TV.
and is shared over a network.
seven-segment display A simple dis-
server-side scripting A technology that play in which the numbers 0–9 and some
allows the content of WEB PAGES to be var- letters can be formed from a group of seven
ied each time they are accessed. It operates segments (see diagram). The segments can
on the Web server rather than the user’s be individually darkened or illuminated to
machine (compare dynamic HTML). A form the different characters. The figure
normal (or static) Web page is written in eight for instance uses all seven segments;
HTML and must be edited to change its con- the figure one uses only two segments. See
tent or appearance. Server-side scripting also LCD; LED display.
embeds fragments of program written in a
SCRIPTING LANGUAGE in the Web page; SGML (standard generalized markup lan-
when processed by the Web server, this guage) A METALANGUAGE used to define
program produces HTML that varies in re- MARKUP LANGUAGES adopted by the Inter-
sponse to choices made by the user or to national Organization for Standardization
changing external data. Such pages are (ISO) in 1986 that allows documents to be
known as dynamic Web pages and are independent of any particular application.
widely used for any purpose that requires The text in such a document has ‘tags’
either up-to-date information (e.g. current defining the type of information. For ex-
202
shift instruction
ample, in this book the basic text is tagged that provides a COMMAND-LINE INTERFACE.
as follows Some operating systems, in particular
UNIX, have a choice of shells.
<en><hw>SGML</hw>
<vg>(<var>standard generalized shift 1. A concerted movement of some
markup language</var>)</vg> or all of the bits held in a storage location
<def>A form of…</def></en> in a specified direction, i.e. to the left or to
the right. Shifts are performed in SHIFT REG-
Here <en> and </en> enclose the whole ISTERS. A left (or right) shift of, say, 3 bits
entry; <hw> and </hw> enclose the head- means that 3 bits are removed from the left
word; and <def> and </def> enclose the de- (or right) end of the register. The bits are
finition. Also <vg> and </vg> denote a said to be shifted left, right, out, etc.
group of variant‘s of the headword. In this There are three kinds of shift. In a logi-
case there is only one variant, which is cal shift the bits shifted from the end of the
tagged using <var> and </var>. Note that register are lost, and 0s are shifted in at the
the tagging is for information content opposite end. In a circular shift (or rotate)
rather than typeface, although often tags the bits shifted from the end are reinserted
such as <b> and <i> are used for bold and at the opposite end, in the order in which
italic type. Another feature of SGML is the they were shifted out. In an arithmetic shift
use of ASCII characters only. Nonstandard the bit pattern is regarded as representing a
characters are represented by strings, as in binary integer, and only the bits represent-
é for acute e (é), á for acute ing the magnitude of the number are
a (á), ° for degree sign (°), and so on. moved; this preserves the positive or nega-
The rules for SGML are quite complex. A tive sign (see complement). If the bits
particular feature, which makes the data shifted out are all zero, a left shift of n bits
portable, is the use of a document type de- is equivalent to multiplying by 2n and a
finition (DTD), defining the way the tags right shift of n bits is equivalent to integer
are used in a document or class of docu- division by 2n.
ments. For example, HTML is defined in The effects of different kinds of 3-bit
SGML with a DTD detailing the structure shift on the same 8-bit word are shown in
of Web pages. See also XML. the table; the left arithmetic shift is equiva-
lent to multiplication by 23 (i.e. 8), the
SGN See Basic. right arithmetic shift to division by 23.
2. On a keyboard, a movement with a SHIFT
share See shared folder. KEY.
shared folder (share) The term used for shift instruction A MACHINE INSTRUC-
a NETWORK DIRECTORY on Windows- and TION specifying that the contents of a SHIFT
Macintosh-based networks. REGISTER are to be shifted either to the left
or to the right by a specified number of po-
shareware /shair-wair/ Software that is
available for use for a trial period, after
which a decision is taken not to use it or it original word 00010111
must be registered and paid for. Shareware
logical left 10111000
programs are less expensive than commer-
logical right 00000010
cial software but may be more error-prone
and not supported. circular left 10111000
circular right 11100010
sheet-fed scanner A SCANNER with a
single-sheet feeding mechanism that can arithmetic left 10111000
automatically scan multipage documents. arithmetic right 00000010
shell The part of an operating system Effects of different 3-bit shifts on an 8-bit word
203
shift key
sitions. The bits that are introduced as re- signature 1. A set of bytes that can be
placements at the opposite end of the regis- used to identify a virus.
ter depend on the kind of SHIFT. In the case 2. A set of characters used for identifica-
of a circular shift, the bits shifted off one tion.
end of the register are reintroduced at the
other end. With a logical shift and an arith- sign bit A single bit used to represent the
metic shift, zeros are introduced and the sign of a number; normally 0 represents +
bits shifted out are often discarded. (plus) and 1 represents – (minus). See com-
plement; sign-magnitude notation.
shift key A keyboard key that when de-
pressed at the same time as another key ei- significant digits (significant figures)
The digits in a number that make a contri-
ther gives an upper-case letter or the
bution to the value of that number. The
character marked on the upper part of the
number 0.1234 has four significant digits:
second key. The shift key is also used in
the zero does not contribute to its value.
combination with other keys to perform The number of significant digits can be re-
special functions. duced by TRUNCATION or ROUNDING.
shift register A REGISTER in which SHIFTS sign-magnitude notation A notation
are performed. that can be used to represent positive and
negative integers in a computer. Normally
Shockwave /shok-wayv/ A technology the leftmost bit in a WORD is used to denote
developed by Macromedia that enables the sign (0 for +, 1 for –) and the remaining
Web pages to include multimedia objects. bits are used to represent the magnitude of
In order to see a Shockwave object with a the integer; the leftmost bit is called the
Web browser a plug-in is usually needed, sign bit. It is more usual, however, for a
which is freely available on the Internet. computer to use two’s complement nota-
tion to represent signed integers. See com-
shortcut key See accelerator key. plement.
short filename See long filename. sign off See log off.
204
SMTP
205
SNMP
tween servers. An e-mail client can then re- can be run on a computer system, as dis-
trieve the messages using POP or IMAP. tinct from the physical components – the
HARDWARE – of that computer system.
SNMP (simple network management pro- There are two basic forms: SYSTEMS SOFT-
tocol) A TCP/IP protocol. Hardware or WARE and APPLICATIONS programs. Systems
software SNMP agents can monitor net- software, which includes the OPERATING
work device activity and report to a work- SYSTEM, is an essential accompaniment to
station. the hardware in providing an effective
overall computer system; it is thus nor-
snowflake curve See fractal. mally supplied by the computer manufac-
turer. The applications programs relate to
SOAP A protocol that uses XML and the role that the computer system plays
HTTP to send REMOTE PROCEDURE CALLS and within a given organization. The DOCU-
other structured communications between MENTATION associated with all these pro-
programs running on different computers grams is also considered part of the
linked by the Internet. It was initially de- software.
signed in 1998 with support from Mi-
crosoft and is now controlled by the W3C. software engineering The entire range
SOAP initially stood for ‘Simple Object of activities involved in the design and
Access Protocol’, but this meaning has development of software. The aim is to
been dropped. produce programs that are reliable and ef-
ficient by following standards of quality
soft font A FONT where the instructions and adhering to specifications, as is done in
that define the shape of each character are other engineering disciplines.
held as data, separate from the software
and hardware in a VDU, printer, etc., and software house A commercial organi-
are used to interpret the font data and ren- zation whose main business is to produce
der the character. Soft fonts may be held in software or assist in the production of soft-
the ROM of a particular output device, but ware. Software houses may offer a range of
they are more often stored as data files in a services, including consultancy, the hiring
computer’s backing store. In the latter case out of suitably qualified personnel to work
the same data is used to create the font on within a client’s team, and a complete sys-
each output device, ensuring compatible tem design and development service.
results. See also downloadable font.
software interrupt See interrupt.
soft keyboard A KEYBOARD in which the
meaning of each key, i.e. the code to be software package See application pack-
generated by each key, can be allocated age.
and possibly changed by program control.
Generally, not all the keys are ‘soft’; some software piracy See piracy.
are permanently wired so that they have a
fixed meaning. software tool One of a set of programs
that are used in the development, repair, or
soft return A line break inserted in a enhancement of other programs or of the
document by a word processor or text edi- computer’s hardware. All the programs in
tor to prevent a line exceeding the maxi- the tool set have a common mode of use
mum length allowed. The positions of soft and employ files and other facilities in a
returns are adjusted automatically as the well-defined standard way. A typical set of
text is edited. tools might consist of a TEXT EDITOR, COM-
PILER, LINK EDITOR, and some form of
soft-sectored disk See sector. DEBUG TOOL. See program.
software /sôft-wair/ The programs that SOHO /soh-hoh/ (small office home of-
206
speech recognition
fice) The market for computers and soft- source program A program that is
ware composed of home computer users written by a programmer in a PROGRAM-
and small businesses. MING LANGUAGE, but cannot be executed
directly by a computer. It must first be con-
solid-state memory See semiconductor verted into a suitable form by a specialized
memory. program. This program may be a COM-
PILER, INTERPRETER, or ASSEMBLER. See also
sorting The process of rearranging in- object module; high-level language.
formation into a desired order. This is
achieved by means of sort keys that are as- space character A member of a CHAR-
sociated with each RECORD of information. ACTER SET that is usually represented by a
Sorting is achieved by comparing sort keys blank site on a screen, printout, etc. It
and rearranging the records accordingly. causes the print or display position to move
The sort keys are derived from the values forward by one position without produc-
stored in the fields of the records in such a ing any mark. Within the computer a space
way that by ordering the sort keys in as- is treated in exactly the same way as any
cending or descending sequence, the asso- other character.
ciated records will be placed in the desired
order. For instance, if the sort key were to spam /spam/ Junk mail on the Internet.
be derived from date of birth and surname, The volume of spam has increased hugely
then the result of a sort would be to place in recent years. While such countermea-
the records in order of age, with those born sures as KILL FILES used to be adequate, to-
on the same day in alphabetical order.
day’s avalanche of spam requires more
There are many methods of sorting. For
sophisticated solutions and anti-spam fil-
example, in the straight insertion sort each
ters are now a standard part of e-mail sys-
sort key in a collection of records is
tems.
checked in turn, and the record associated
with the key is repositioned (if necessary)
special character A symbol in a CHAR-
with respect to all previous sort keys. In a
ACTER SET that is a GRAPHIC CHARACTER but
bubble sort a pair of records is inter-
is not a letter, nor a digit, nor (usually) a
changed if their sort keys are not in the cor-
SPACE CHARACTER. Punctuation marks,
rect order, different pairs being checked on
different passes through the collection of brackets, and arithmetic symbols are spe-
records. This continues until no further in- cial characters. The special characters on
terchanges are necessary. the keyboard vary from one country to an-
other, and would include, for instance, the
sort key See sorting. national currency symbol. Compare al-
phanumeric character; control character.
SoundBlaster /sownd-blass-ter/ A range
of sound cards manufactured by Creative specification /spess-ă-fă-kay-shŏn/ See
Technology and its subsidiaries; it has be- system specification.
come the de facto standard.
speech generation device A means of
sound card An expansion board giving producing spoken messages in response to
the capability of playing sound from CDs, signals from a data processing or control
DVDs, MIDI files, or other sound files. system. The selection of messages are pro-
duced by assembling speech sounds from a
source code See source language. set of fundamental sounds that may be ar-
tificial in origin or may have been extracted
source language The programming by processing human speech.
language in which a SOURCE PROGRAM is
written. The program is in the form of speech recognition The process whereby
source code. a computer interprets spoken words in
207
speech synthesis
order to determine their data content. See on to the array of cells; if there are more
also voice input device. rows and columns than will fit on the
screen, then the spreadsheet can be scrolled
speech synthesis The production of ar- horizontally or vertically to bring into view
tificial speech by a computer, usually from previously hidden rows or columns. To
textual input. The technique involves stor- change a value it is only necessary to move
ing prerecorded words or sounds, or artifi- the CURSOR into the required cell displayed
cially generated versions of such sounds, on the screen and type in the new value.
and using software that analyzes text and Spreadsheets can be used for storing
joins the sounds together for output. and amending accounts, ‘what if?’ finan-
cial projections, and many other applica-
speech understanding The process of tions involving tables with interdependent
using SPEECH RECOGNITION in order to per- rows and columns. A spreadsheet is often a
form some task making use of speech, such component of an INTEGRATED OFFICE SYS-
as voice input. TEM.
208
storage
209
storage allocation
medium in which data or program instruc- 2. The minimum period of time required
tions can be held for subsequent use by a between successive accesses (read or write)
computer. The two basic types of storage to a storage device.
in a computer are MAIN STORE (or main
storage) and BACKING STORE (or backing storage device A device that can receive
storage). See also storage device; media. data and retain it for subsequent use. Such
2. The retention of data or program in- devices cover a wide range of storage CA-
structions in a storage device or storage PACITIES and speeds of ACCESS. The semi-
medium. conductor devices used as MAIN STORE have
3. The act of entering data in a storage de- very fast speeds of access but the cost of
vice or storage medium. storing each bit is relatively high. In com-
parison the devices – MAGNETIC DISKS, CDs,
storage allocation The allocation of DVDs, etc. – used as BACKING STORE have a
specified areas of storage to the various lower cost per bit and a greater storage ca-
processing tasks that are active in a com- pacity, but take much longer to retrieve
puter. This controlled use of storage is nec- data.
essary in a computer system in which
several individual processes may be run- storage location (store location) See lo-
ning. Each process is allocated sufficient cation.
WORKING SPACE in a way that does not in-
terfere with another process’s allocated storage medium See media.
working space. The allocation, and the
amount of storage allocated, is controlled
storage protection Any of many facili-
ties that limit and hence control access to a
by the OPERATING SYSTEM of the computer;
storage device or to one or more storage lo-
with some operating systems the amount
cations. The intention is to prevent inad-
can be changed while a program is run-
vertent interference by users and/or to
ning. Storage allocation is one form of RE-
provide system SECURITY. It is achieved by
SOURCE ALLOCATION.
prohibiting unauthorized reading or writ-
ing of data, or both. In storage devices that
storage area network (SAN) A high- operate at high speed, protection is imple-
speed device connected to a network’s mented by hardware (to maintain speed).
servers, on which their data is stored rather For slower devices it may be done entirely
than on local hard disks. It typically con- by software.
sists of a network of RAID disk arrays and
supporting hardware. This consolidation store 1. See storage.
of data storage and its separation from 2. To enter or retain data or instructions
other server functions simplifies both net- for subsequent use by a computer.
work management and the addition of new
storage capacity. A SAN is faster but less stored program A set of instructions
flexible than NETWORK-ATTACHED STORAGE that can be submitted as a unit to some
and is usually found on large networks. computer system and stored in its memory,
The two technologies can be combined, in advance of being processed. It is then
with an NAS device using a SAN to store possible for instructions to be extracted au-
its data. tomatically from memory at the appropri-
ate time. This is done, without human
storage capacity See capacity. intervention, by the CONTROL UNIT of the
computer.
storage cycle 1. The sequence of events The data to be manipulated by the com-
occurring when a unit of a storage device puter, in accordance with the program in-
goes from one inactive state through a read structions, is also stored in the computer
and/or write phase, and back to an inactive memory. Instructions and data are both
state. represented in binary form within the com-
210
string
puter – i.e. as sequences of the binary dig- streaming files, meaning the user can
its 0 and 1 – and cannot be distinguished watch a video or listen to a sound file while
(see binary representation). As a result a it is still being downloaded from, say, the
program and the data handled by that pro- Internet.
gram can be stored together, sharing the
same storage space. Stretch See second generation comput-
The idea of program and data sharing ers.
the same memory is fundamental to the
great majority of modern digital comput- string One form in which a collection of
ers. The concept of the stored program data items can be held in computer mem-
was documented in 1946 by John von ory. A string is a list of letters, digits, or
Neumann, forming the basis of his design other CHARACTERS, examples being
of EDVAC (see first generation comput- cat
ers). The concept was also considered 735
by John W. Mauchly and J. Presper Eckert Sue Jones
at about the same time. The first opera- COMP
tional stored-program computers were It is therefore an item of textual data.
the Manchester Mark I and EDSAC. All Strings are in fact used mainly for handling
digital computers since then have been text. The length of a string, i.e. the number
stored-program computers, and stored of characters in it, is not fixed. There may
programs have long been called, simply, be only one character. An EMPTY STRING is
programs. also a possibility. Any sequence of charac-
The storage of programs and data in ters within a string is called a substring; bc
READ/WRITE MEMORY allows the stored data
and abc are possible substrings of abcde.
to be modified by the operations per-
A string is represented in memory as a
formed on it, and if necessary the program
line of adjacent LOCATIONS; either the first
itself to be modified by the program in-
location holds a number that gives the
structions. Modifying a running program
string length and the remaining locations
is normally avoided since it is a major
hold the characters in correct order, or the
source of error. The opportunities for pro-
string is terminated by a location contain-
gram modification in present-day comput-
ing a special noncharacter code. A string is
ers are reduced by dividing the program
into two parts: one part must not be mod- denoted in a program by various means,
ified, and is therefore either stored in a pro- depending on the programming language.
tected area of memory or is permanently In Basic, for example, a dollar sign ($) as
resident in read-only memory (ROM, PROM, the final character of a VARIABLE name de-
or EPROM); the other part may be modified, clares it to be a string. In many languages
and is stored in an area of read/write mem- the characters making up a string are
ory. placed in quotation marks. An example in
See also program. Basic would be
PET$ = ‘HORSE’
storm A sudden excessive increase in Various operations can be performed
traffic on a network. on strings in addition to input/output and
storage. They can be joined together (see
straight insertion sort See sorting. concatenation); they can be compared to
find out if they differ; they can be searched
streamer See tape streamer. to see if they contain a given sequence of
characters; a given substring can be ex-
streaming The movement of data from tracted and replaced by another sequence
one computer to another in such a way that of characters. The operations that can be
it does not have to be completely trans- performed are known collectively as string
ferred before the receiving computer can manipulation, the choice of operations de-
start using it. Often multimedia files are pending on the programming language.
211
string manipulation
string manipulation See string. combined with the library file by the LINK
EDITOR, the library is searched for the re-
striping A method of improving HARD quired routines, and only they are ex-
DISK performance. Several disks are tracted. If searching is not used, all the
grouped together and act as one large disk, library routines would be combined with
with the data being distributed across all the user program whether they were re-
the disks in a way designed to minimize ac- quired or not. See also program library.
cess times. RAID systems use striping.
subscript /sub-skript/ (index) A value,
structured programming An approach usually an integer or integer expression,
to producing an actual program in a high- that is used in selecting a particular element
level language in which only three struc- in an ARRAY. In computing, subscripts can-
tures need be used to govern the flow of not be put in a subscript position (e.g. xi),
control in the program. These three struc- as the name suggests, and some other nota-
tures – CONTROL STRUCTURES – allow for se- tion must be used. This usually involves
quential, conditional, and iterative flow of brackets, as in x(i) or x[i].
control. Arbitrary transfer of control – in
the form of the GOTO STATEMENT – is ex- substring /sub-string/ See string.
pressly forbidden. As a result, for each
statement or group of statements in a struc- suitcase A Macintosh file containing
tured program there is precisely one entry fonts or desk accessories.
point and one exit point. The program is
thus (in theory) easy to follow, to debug, suite /sweet/ A set of PROGRAMS or pro-
and to modify. gram MODULES that is designed so as to
meet some overall requirement, each pro-
structured query language See SQL. gram or module meeting some part of that
requirement. For example, an accounting
StuffIt /stuff-it/ A Macintosh program suite might consist of separate programs
for compressing and decompressing files. It for stock control, inventory, payroll, etc.
is now commercially available for both
Macintoshes and PCs. Sun Microsystems A US company
founded in 1982 that builds computer
subprogram /sub-proh-gram/ See pro- hardware and software. It is best known
gram unit. for developing workstations and operating
environments for UNIX operating systems
subroutine /sub-roo-teen/ 1. A section and also for the development of JAVA.
of a program that carries out a well-defined
operation on data. Subroutines function supercomputer /soo-per-kŏm-pyoo-ter/
like PROCEDURES except that they cannot A computer that is capable of working at
have PARAMETERS. They are found in AS- very great speed, and can thus process a
SEMBLY LANGUAGE and a few HIGH-LEVEL very large amount of data within an ac-
LANGUAGES, particularly Basic. ceptable time. Supercomputers are in fact
2. The term used for a procedure in such now purpose-built to manipulate numbers,
languages as Fortran. and are the most powerful form of NUMBER
CRUNCHER. Advanced supercomputers can
subroutine library A collection of SUB- handle tens of billions of floating-point op-
ROUTINES, PROCEDURES, and FUNCTIONS erations per second. These machines are
kept in the form of object code in a single used, for example, in meteorology, engi-
file on BACKING STORE. A subroutine library neering, nuclear physics, and astronomy.
will often consist entirely of routines con- The main manufacturers of supercomput-
nected with a common subject area, such ers are Cray Inc., IBM, and NEC.
as graphics or numerical analysis. When a
user program requiring library routines is superuser See administrator.
212
symbolic logic
213
symbol table
way. It is sometimes called mathematical tor’s baton), and thus take place at regular
logic. and predictable times.
The simplest system of symbolic logic In synchronous transmission between
is propositional logic (sometimes called two devices, the devices can operate con-
propositional calculus) in which letters, tinuously and each bit of data is transmit-
e.g. P, Q, R, etc., stand for propositions or ted at a fixed and predictable time.
statements, and various special symbols Compare asynchronous.
stand for relationships that can hold be-
tween them. syntax /sin-taks/ Part of the description
of a PROGRAMMING LANGUAGE, akin to the
symbol table A list kept by a COMPILER grammar of English or some other natural
or an ASSEMBLER of the IDENTIFIERS that language. The syntax of a language is the
have been used in a program. An identifier set of rules defining which combinations of
consists of a string of characters chosen by letters, digits, and other characters are per-
the programmer to identify a VARIABLE, an mitted in that language. The rules allow a
ARRAY, a FUNCTION, a PROCEDURE, or some valid sequence of characters to be distin-
other element in the program. The proper- guished from an invalid, i.e. a meaningless
ties of these identifiers, such as their ad- combination. They determine the form of
dress, value, or size, form part of the table. the various constructs – STATEMENTS, CON-
As the translation of the source pro- TROL STRUCTURES, etc. – but say nothing
gram proceeds, the compiler or assembler about the meaning of the constructs. The
recognizes occurrences of identifiers – or syntax of a language can be expressed pre-
symbols – in the text and looks them up in
cisely and unambiguously, allowing SYN-
the table. If a symbol is already in the table,
TAX ERRORS to be detected. It is often
then the associated value is used by the
expressed in a symbolic notation called
compiler or assembler instead of the sym-
Backus-Naur form (BNF) or some deriva-
bol. If the symbol is not in the table, then it
tive of BNF. See also semantics.
is entered and eventually, when a defini-
tion of the symbol is found, the correct in-
formation is also entered. The compilation
syntax analysis (parsing) A phase in the
COMPILATION of a program during which
or assembly will terminate with errors un-
less all symbols are defined. the program is checked for compliance
with the SYNTAX of the programming lan-
symmetric digital subscriber line See guage that has been used. If syntax errors
SDSL. are found then the compilation process is
stopped and the errors can be corrected.
symmetric multiprocessing See SMP. Syntax analysis follows the process of LEX-
ICAL ANALYSIS, and is achieved by means of
214
systems software
215
t
tab key A key positioned toward the top tape cassette See cassette.
left of the keyboard and marked with two
opposing arrows. It is used for inserting a tape deck See tape drive.
tab character into a document. Generally,
the tab key will also move the FOCUS from tape drive (tape transport; tape deck) A
place to place in a GUI system. mechanism for moving MAGNETIC TAPE and
controlling its movement. A tape drive to-
table A collection of data values, each gether with magnetic read and write heads
one of which can be identified by some and associated electronics is known as a
method. The values may be arranged in an TAPE UNIT. The term tape drive is often used
ARRAY and a particular item is then identi- however as another name for the whole
fied by one or more subscript values. Alter- tape unit.
natively the data may be stored in the form
of RECORDS; one or more fields in each tape format See format; magnetic tape.
record form a unique KEY, and the key is
the means of identification. tape streamer (streamer) A type of TAPE
Values can be extracted from tables UNIT specifically designed for the rapid
using the method of table look-up (TLU). BACKUP of magnetic disks using TAPE CAR-
The subscript or key in the table is then the TRIDGES. The tape streamer is an efficient
value to be looked up. The table may, for way of backing up the fixed disk systems.
example, be a list of square roots. The 4th
entry in the table would then be the square tape transport See tape drive.
root of 4, i.e. 2. Again, the table may con-
sist of prices of numbered items on a menu. tape unit (magnetic tape unit, MTU) A
Entering the number of a particular dish peripheral device containing a mechanism
gives its price; the table can thus be used to for moving MAGNETIC TAPE and controlling
calculate the bill for a particular selection its movement, together with recording and
of dishes. sensing HEADS (and associated electronics)
that cause data to be written to and read
table look-up (TLU) See table. from the tape. Magnetic tape is held in
TAPE CARTRIDGES, CASSETTES, or, in older
tag See SGML; XML. systems, on reels. The reel, cassette, or car-
tridge is mounted in the tape unit, and the
tape See magnetic tape. tape is driven past the heads.
Data is recorded on the tape by the
tape cartridge Usually, a casing con- write head: the head receives an electrical
taining MAGNETIC TAPE and from which the signal coded with the data and converts it
tape is not normally removed. The casing into patterns of magnetization on the tape.
therefore protects the tape. The cartridge Data on the tape can be sensed by the read
can be loaded in a suitable TAPE UNIT with- head: the patterns of magnetization induce
out the tape being handled. Cartridges are a coded electrical signal in the read head,
similar to CASSETTES but are faster and have and this signal can be fed to the computer.
larger storage capacities. Data is stored on and retrieved from mag-
216
terminal
netic tape by the method of SERIAL ACCESS, users to participate in a conference or some
i.e. by first reading through all previous other joint activity despite being at differ-
items of data on the tape until the required ent places and/or communicating at differ-
location is reached. ent times. Users typically have access to
Tapes are nowadays no longer used as computers or terminals interconnected by
on-line backing store. However, they are communication lines. A contribution made
still an important backup medium because by any member of the conference will be
they are relatively cheap and their main brought to the attention of each of the
disadvantage – long access times – is unim- other members when they next use their
portant. computers or terminals. Sophisticated tele-
conferencing systems use audio and video
tar A UNIX shell command that makes a equipment.
single file (called an archive) from a set of
files or extracts files from such an archive. television monitor See monitor.
The files are not compressed and the ex-
tension of the archive file is tar. Tar is a teleworking /tel-ĕ-wer-king/ The use of
contraction of ‘tape archive’. computers and communications technolo-
gies to interact with customers and busi-
target language The language into ness colleagues. See also SOHO.
which source code is compiled.
Telnet /tel-net/ A high-level communica-
task See process. tions protocol that lets a user log on to a re-
mote computer.
task bar A bar containing the start but-
ton, which appears at the foot of a Win- tera- (ter-a) A prefix indicating a multi-
dows screen. Icons representing active ple of a million million (i.e. 1012) or,
applications or documents are also shown loosely, a multiple of 240 (i.e. 1 099 511
here and can be selected with the mouse. 627 776). In science and technology deci-
mal notation is usually used, and powers of
T-carrier A long-distance digital com- 10 are thus encountered – 103, 106, 109,
munications line used for voice communi- etc. The symbol T is used for tera-, as in TV
cation and Internet connection using for teravolt. Binary notation is generally
pulse-code modulation and time-division used in computing, and so the power of 2
multiplexing. The T-carrier service has sev- nearest to 1012 has assumed the meaning of
eral different capacity levels, from T1 to tera-.
T4. The prefix is most frequently encoun-
tered in computing in the context of stor-
TCO See Total Cost of Ownership. age CAPACITY. With magnetic disks, mag-
netic tape, and main store, the capacity is
TCP/IP /tee-see-pee-ÿ-pee/ (transmission normally reckoned in terms of the number
control protocol/Internet protocol) A set of BYTES that can be stored, and the ter-
of protocols developed originally by the US abyte is used to mean 240 bytes; terabyte is
Department of Defense, which has become usually abbreviated to T byte, Tb, or just
the de facto standard for communications to T.
between computers and networks. Tera- is part of a sequence
kilo-, mega-, giga-, tera-, peta-, ...
telecommuting /tel-ee-kŏ-myoot-ing/ The of increasing powers of 103 (or of 210).
use of telecommunications and computer
networks to work outside the normal of- teraflop /te-ră-flop/ See flop.
fice or workplace.
terminal An input/output device that is
teleconferencing /tel-ee-kon-fĕ-rĕn-sing/ used for communication with a computer
A computer-based system enabling its from a remote site. (Some terminals are re-
217
termination
stricted to the output of data or to the testing Any activity that checks whether
input of data.) The terminal may be linked a system – software or hardware – behaves
by cable to the computer, which may be in in the desired manner. It is achieved by
the next room or a nearby building, or the means of a test run in which the system is
two may be connected over a telephone supplied with input data, and the system’s
line. In larger computer systems there may responses are recorded for analysis. The
be many terminals linked to one computer input data is referred to as test data. Test-
(see multiaccess system). ing can be performed on individual com-
The most common type of terminal is ponents of the system in isolation; when
the VDU paired with a KEYBOARD: the user the components are brought together, fur-
of the VDU types information to the com- ther testing can be performed to check that
puter and information from the computer the components operate together correctly.
is received on the VDU screen. Terminals Alternatively the behavior of a system can
can be designed for a particular applica- be investigated without concern for indi-
tion; examples include POINT-OF-SALE TER- vidual components and their internal inter-
MINALS and cash dispensers at banks. If a faces. Testing can never absolutely prove
terminal has a built-in capability to store the correctness of a program. It can only
and manipulate data – i.e. it contains a mi- show that the program works under cer-
crocomputer – then it is classed as an intel- tain circumstances. See also debugging.
ligent (or smart) terminal; without this
capability it is described as dumb. test run See testing.
termination The end of execution of a TeX /tek/ A large and complex typeset-
ting system available as freeware for most
program or some other processing task. A
operating systems, including Windows,
processing task that reaches a successful
UNIX, and Macintosh. It was created by
conclusion terminates by returning control
Donald Knuth for the production of high-
to the OPERATING SYSTEM; this is described
quality scientific, mathematical, or other
as a normal termination. If the processing
technical documents. See also LaTeX.
task reaches a point from which it cannot
continue, it ABORTS itself, or is aborted by texel /teks-ĕl/ The smallest graphical
the operating system or the user; this is re- element in a two-dimensional texture map-
ferred to as an abnormal termination. ping used in the rendition of a three-di-
mensional object. Pixels and texels are
terminator 1. A character or group of similar, but not the same, as although there
characters used to mark the end of a pro- may be a one-to-one correspondence in
gram STATEMENT, CONTROL STRUCTURE, or some parts of a rendition, they cannot usu-
some other item of data. For example, in ally be paired off in this way.
Fortran every statement is terminated by
an end of line while in the language C every text box A type of CONTROL on a GUI
statement must have a semicolon at the FORM. The user types in content, which
end. Compare separator. may be validated in some way but is not re-
2. (rogue value) A value that is added at the stricted to a set list of choices. Compare list
end of a list of data values and that is rec- box; see also combo box.
ognized by the computer as a signal to ter-
minate some operation on the data. It must text editor A UTILITY PROGRAM used
not be possible for the terminator to ap- specifically for input and modification of
pear in the list of data values. If the data information in textual form. For example,
values consist of, say, numbers from 0 to a document in English can be keyed into a
50, then 99 could be selected as the termi- computer and a text editor used to create a
nator. FILE. The same applies to a program in a
high-level programming language. A text
test data See testing. editor can also be used to inspect the con-
218
third generation computers
application server
capable of hosting
many consecutive user
sessions and where
most of the processing
work is done
Thin client
219
thrashing
220
trace
time slicing A technique whereby the quired of the program or system. This is
time of a computer can be shared among broken down into a succession of different
several PROCESSES, a brief period being allo- levels that are progressively more detailed.
cated (by the OPERATING SYSTEM) to each At each level more basic elements are in-
process in turn. During such a period – troduced to describe exactly what is re-
known as a time slice – the process is per- quired. These elements are then defined in
mitted to use the resources of the com- the next stage by introducing even more
puter, i.e. the processor, main store, etc. basic elements. The process stops when
Time slicing differs from TIME SHARING in there is sufficient detail for the elements to
that it disregards which jobs or users own be written in the chosen programming lan-
the processes and so makes no effort to guage or to be brought together to form the
share the available time evenly between desired system.
jobs or users. See also multitasking.
Total Cost of Ownership (TCO) A
TLU (table look-up) See table. method of calculating the cost of software,
hardware, etc., that takes into account not
token See lexical analysis. only the purchase price but also all pro-
jected subsequent costs over the item’s life-
token ring network A LAN in which time: upgrades, repair, support, user
the nodes are connected together in such a training, etc. For example a low-price soft-
way that each node can only communicate ware application that is difficult to use and
with its neighbors on either side, leading to requires extensive user training and sup-
a logical ring. In practice star clusters of up port might have a higher TCO that a rival
to eight workstations can be connected up high-price application that can be config-
to a multiaccess unit that acts as a node on ured easily and used intuitively.
the ring. The single token is a series of bits
sent around the ring from node to node, touchpad /tuch-pad/ See touch-sensitive
and it is used to arbitrate who has access to device.
the ring at any one time. When a node
wants to communicate it must grab the touch screen See touch-sensitive device.
token when it is empty, and then it can
send data to any other node; this data must touch-sensitive device A flat rectangu-
be delivered successfully before any other lar device that responds to the touch of,
node can grab the token. Originally devel- say, a finger by transmitting the coordi-
oped by IBM and mainly used on IBM sys- nates of the touched point (i.e. its position)
tems, the use of token ring networks has to a computer. The touch-sensitive area
declined in recent years in favour of ETHER- may be the VDU screen itself, in which case
NET. See also network. it is called a touch screen; alternatively it
may be part of the keyboard or a separate
toolbar A block of icons showing in an unit that can be placed on a desk. In the
application window, often in a row or col- first case the computer user can, for exam-
umn, that, when clicked on with the ple, make a selection from a number of op-
mouse, activate functions within the appli- tions displayed on the screen by touching
cation. For example, a drawing toolbar one of them; in the latter two cases move-
could contain icons to draw a circle, pick ment of the finger across the so-called
up a paintbrush, shade an object, and so touchpad can cause the CURSOR to move
forth. Toolbars can often be customized by around the screen. See also mouse.
the user and moved and resized on the
screen. TP See transaction processing.
221
track
trace program records – or traces – the se- tions that must be treated as a unit; in par-
quence in which the statements in the pro- ticular, they must all be processed success-
gram are executed, and usually the results fully or not processed at all. This means
of executing the statements. It is thus pos- that, in the case of failure, any updates to
sible to follow changes in data values at data, etc., made by previous instructions in
each stage of the program. There may be the transaction must be undone (rolled
additional options offered by a trace pro- back). For example, making an entry in an
gram, for example it may trace changes to accounting system typically involves cred-
the value of a specific variable in the pro- iting one account with an amount and deb-
gram. The trace information is stored in a iting another account with the same
trace table. A trace program can be used in amount; were the credit to be carried out
DEBUGGING a program. but the debit to fail, or vice versa, the data
files would be left in an inconsistent state.
track The portion of a magnetic storage Grouping the credit and debit as a transac-
medium along which data is stored. In tion ensures that this cannot happen.
MAGNETIC DISKS the tracks are concentric Transactions are used especially with data-
circles along which data is recorded as a bases.
stream of bits. There are several thousand 2. A request for computer action that must
tracks on a hard disk and 80 on a floppy be treated as such a transaction. Examples
disk. The tracks are subdivided into SEC- include making a booking or purchase on-
TORS for the purposes of reading and writ- line, getting cash from a cash dispenser,
ing items of data. etc.
train printer See line printer. transfer rate See data transfer rate.
222
tree
223
trojan
F root node
D H
arcs
B E G
J
A C I K
leaf nodes
Binary tree
The leaf nodes at the bottom of the dia- because the REGISTERS in a computer can
gram have empty links. Note that trees are hold only a fixed number of binary digits
drawn upside down with the root at the (bits). When the result of an arithmetic op-
top and the leaves at the bottom. eration requires more than this fixed num-
ber of bits, either truncation or ROUNDING
trojan /troh-jăn/ (trojan horse) A small will take place. The truncated value is an
destructive program disguised as a game, approximation to the true value of the
application, or utility, which when acti- number, and leads to a truncation error.
vated causes damage although appearing Repeated truncation can cause a build-up
to be of use. of such errors and the programmer should
try to minimize this problem.
true color The specification of the color
of a pixel on a display screen using a 24-bit truth table A table of values that de-
value, 8 bits for each of the three primary scribes a particular LOGIC OPERATION or
RGB colors, giving the possibility of LOGIC GATE. Truth tables were first used in
16 777 216 colors. connection with statements in logic that
could be assigned one of two truth values –
TrueType /troo-tÿp/ A scalable font true or false. They can be used in comput-
technology introduced first by Apple and ing to describe the logic operations, such as
then by Microsoft to enable high-grade AND, OR, and NOT, that are performed
fonts to be used in their operating systems. for example in the ARITHMETIC AND LOGIC
To view and print a TrueType font it is
necessary to have the font file itself and
TrueType raster graphics, which is inbuilt P Q P AND Q NOT (P AND Q)
in Windows and Macintosh operating sys-
tems. TrueType fonts can be embedded in 0 0 0 1
Web pages or word documents and viewed 0 1 0 1
by users who do not have the font installed 1 0 0 1
on their machines. Windows and Macin- 1 1 1 0
tosh versions of TrueType fonts are not
compatible. See also font.
P Q P OR Q NOT (P OR Q)
truncation /trunk-ay-shŏn/ The process 0 0 0 1
applied to a number whereby all digits 0 1 1 0
after a specified number of significant dig- 1 0 1 0
its are ignored. For example, truncation of 1 1 1 0
the number 3.257 69 to 4 or to 5 digits
yields 3.257 or 3.2576 respectively. In
computing, truncation is usually required Truth tables for AND, OR, and NOT operations
224
turnkey operation
UNIT of a central processor. The truth table ized by fairly high SWITCHING SPEEDS.
of a particular operation lists all possible Schottky TTL has higher switching speeds
combinations of the two truth values of the and lower power requirements than the
quantities on which the operation is to be standard TTL. Both versions have a lower
performed (the operands), together with PACKING DENSITY than CMOS. See also ECL.
the truth value of the outcome of each of
the possible combinations. If there are 2 tunneling /tun-ĕl-ing/ A method of trans-
operands, P and Q, there are 22, i.e. 4, pos- mitting data over connected networks
sible outcomes. Truth tables for AND, OR, using different protocols. The packet based
and NOT operations are shown in the dia- on one protocol is wrapped in the packet of
gram. The truth values are represented as 1 another protocol so that it can travel over
(for the value true) and 0 (for the value the network using the second protocol.
false). Tunneling is used in virtual private net-
Truth tables are also used to describe or works (VPNs).
determine the function of a logic gate. The
truth table of a particular gate lists the out- Turing machine /tew-ring/ An abstract
put obtained for all the possible combina- model of a computer, put forward by Alan
tions of inputs. With 2 inputs, each of Turing in 1936, that consists of a control
which can be at one of two voltages, there or processing unit and an infinitely long
are 4 possible output voltages. With 3 in- tape divided into single squares along its
puts there are 8 possible outputs. The truth length. At any given time each square of
tables for AND and OR GATES with two in- the tape is either blank or contains a single
puts are identical to those shown in the symbol s1, s2, …, sn. The Turing machine
table for P AND Q and P OR Q. The truth moves along the tape square by square and
table for a NOT GATE (which always has one reads, erases, and prints symbols. At any
input) is identical to that for NOT P. given time the Turing machine is in one of
a fixed finite number of states represented
truth value Either of two values, true or by q1, q2, …, qn. The ‘program’ for the ma-
false, that can be assigned to a logical ex- chine is made up of a finite set of instruc-
pression or conclusion in BOOLEAN ALGE- tions of the form qisjskXqj, where X is
BRA. The values are symbolized by T and F. either R (move to the right), L (move to the
When used in computing these so-called left), or N (stay in the same position). Here,
logical values can be represented by and qi is the state of a machine reading sj, which
stored as a single bit (usually 1 for true and it changes to sk, then moves left, right, or
0 for false). Larger units, such as a byte or stays and completes the operation by going
word, are also used to store logical values into state qj.
since this is simpler and quicker, although The Turing machine is considered the
less efficient in storage space. See also truth theoretical model for digital computers. It
table. is still important in computer science.
TSR (terminate and stay resident) De- turnaround time 1. The time that
scribing MS-DOS programs that stay in elapses between the submission of a job to
memory at all times. Such programs can be a computer facility and the return of the re-
instantly accessed from within other pro- sults. This is the only time that really mat-
grams. TSRs are not needed in Windows ters to someone using a computer in this
but are still supported for BACKWARD COM- way.
PATIBILITY. 2. The time taken to reverse the direction
of transmission on a one-way (SIMPLEX) or
TTL (transistor-transistor logic) A fam- a HALF-DUPLEX communication channel.
ily of LOGIC CIRCUITS that are all fabricated
with a similar structure by the same INTE- turnkey operation /tern-kee/ The deliv-
GRATED-CIRCUIT techniques. The circuits ery and installation of a complete com-
are bipolar in nature and are all character- puter system plus applications programs so
225
turtle graphics
that the system can be placed into immedi- scanner or digital camera. Nearly all such
ate operational use. This operational sys- devices contain a TWAIN driver, which
tem is called a turnkey system. hides the detailed operation of the device
behind the standard interface. A TWAIN-
turtle graphics A method by which in- aware application can thus use any device
formation from a computer can be turned with a TWAIN driver.
into pictures or patterns. The original
drawing device was a simple pen-plotter tweening /tween-ing/ A technique used
known as a turtle, a small motorized car- in computer animation that enables
riage carrying one or more pens and con- smoother animation to be achieved by the
nected to the computer by a flexible cable.
automatic insertion of extra frames be-
The carriage wheels could be precisely con-
tween the originally created frames.
trolled, enabling the turtle to be steered by
commands from the computer in any di-
twisted pair An electric cable consisting
rection across a floor or other flat surface
of two similar wires that are each electri-
covered in paper or similar material; the
pens could be raised or lowered by control cally insulated by a plastic sheath and are
signals to produce the drawing. twisted around one another to improve the
The action of the turtle can now be sim- transmission of signals. Bundles of tens or
ulated by graphics on the VDU screen: the hundreds of twisted pairs are commonly
screen turtle is usually a triangular arrow encased in an outer protective sheath to
that may or may not produce a line as it is make large cables.
made to move across the screen. See also
Logo. two’s complement See complement.
226
U
227
unconditional jump
228
user port
source; it provides enough running time to which new peripherals can be automati-
cover short breaks and then, if necessary, cally detected and configured without
for the system or device to be shut down in shutting the computer down and without
an orderly fashion so that data is not lost. the need for an adapter card. All modern
The device also protects against sudden microcomputers support USB.
power surges.
Usenet /yooz-net/ A worldwide network
uptime /up-tÿm/ The amount of time a of UNIX systems with thousands of sub-
system has been powered up and running. ject-based newsgroups. It has now become
part of the Internet and has set the stan-
upward compatibility (forward compat- dard for other bulletin-board systems.
ibility) The ability of a computer system
or software to work beside or on more ad- user datagram protocol /day-tă-gram/
vanced equipment or versions. See also See UDP.
backward compatibility.
user-friendly Denoting a piece of hard-
URI (Uniform Resource Indicator) An ware or software that a user should find
identifier for a physical or abstract object easy to use, convenient, or helpful, i.e. that
that follows the syntax laid down in RFC a user who is not a computer specialist
3986 (2005). Internet URLS form one subset should not find confusing or intimidating.
of URIs: in ‘http://www.microsoft.com’,
the part before the colon (’http’) is call the user group A group of people with an
scheme and the remainder of the URI fol- interest in a particular application or com-
lows the syntax specified for the http
puter system. Some user groups are large
scheme. Other schemes include the ‘tel’
and provide forums in which ideas can be
scheme for telephone numbers: the URI
exchanged, problems aired, and help pro-
‘tel:+1-800-123-4567’ represents the US
vided.
telephone number 800-1230-4567. The
use of such standardized names makes
user interface The means of communi-
their manipulation be computer easier.
cation between a person and a computer
system, referring in particular to the use of
URL /yoo-ar-el/ (uniform resource loca-
tor) The address of a resource on the In- input/output devices with supporting soft-
ternet, which specifies the protocol to be ware. The operating system provides the
used in accessing the resource, the name basic facilities in the form of a GRAPHICAL
USER INTERFACE and/or a COMMAND LINE IN-
of the server where the resource is to be
TERFACE, which individual applications
located, and if necessary the path to the
resource. An example address is adapt to suit their requirements. A good
http://www.microsoft.com user interface is important to the success of
indicating a WWW page on the commer- an application.
cial server microsoft.com. See also URI.
username /yoo-zer-naym/ The name by
USART /yoo-sart/ (universal synchronous/ which a user is recognized on a computer
asynchronous receiver–transmitter) A or network system. When logging on a user
module that can send and receive both must enter a username and, usually, a valid
asynchronous and synchronous communi- password.
cations. See also UART.
user port An electrical socket in a mi-
USB (universal serial bus) A serial bus crocomputer that with associated circuitry
used to connect up to 127 peripherals to a allows the computer user to connect addi-
microcomputer through a single general tional peripheral devices. The individual
port, by daisy-chaining peripherals to- pins of the port may be set or tested from a
gether. It is a plug-and-play INTERFACE in user’s program. The port can be used to
229
user program
230
V
231
vCard
vCard /vee-kard/ A specification for cre- universal set, E, is shown as a rectangle. In-
ating an electronic business card for the ex- side this, other sets are shown as circles. In-
change of information in e-mails and tersecting or overlapping circles are
teleconferencing. intersecting sets. Separate circles are sets
that have no intersection. A circle inside
VDSL (very high bit-rate digital subscriber another is a subset. A group of elements, or
line) A variant of DSL technology that a subset, defined by any of these relation-
transfers data over an ordinary copper tele- ships may be indicated by a shaded area in
phone wire at very high speeds, but only the diagram.
over short distances. Like ADSL it can share
the wire with a telephone and transfers verification /ve-ră-fă-kay-shŏn / The
data at different speeds in the two direc- checking of the accuracy of data after it has
tions: up to 52 Mbps downstream (i.e. been transcribed. It is done most com-
when downloading) and 16 Mbps up- monly when data has been encoded by an
stream (i.e. when uploading). It is expected operator at a keyboard reading from a doc-
that VDSL will become more widely used ument. The document is subsequently
as copper wires are replaced by fiber-optic rekeyed by another operator, and the two
cables, which overcome the distance prob- sets of input data compared by machine.
lem. See also SDSL. Any differences are indicated and suitable
action is taken.
VDU (visual display unit) A device that
can display computer output temporarily Veronica /vĕ-ron-ă-kă/ (very easy rodent-
on a screen. The information can be com- oriented network index to computerized
posed of letters, numbers, or other charac- archives) A GOPHER searching service de-
ters, or can be in the form of pictures, veloped by the University of Nevada, simi-
graphs, or charts. The information can be lar to ARCHIE, which is used on FTP servers.
changed or erased under the control of a
computer; this computer may be at a re- vertical application A specialized ap-
mote site or may be a microcomputer of plication designed for use in a particular
which the VDU forms a part. The VDU is area of business or by a particular type of
usually paired with a KEYBOARD, by which user; for example, insurance, government,
information can be fed into the computer. or hospitals.
A VDU plus keyboard is now the most
commonly used computer TERMINAL. very high bit-rate digital subscriber
Modern GUI-based operating systems line See VDSL.
treat all output to VDUs as graphics. How-
ever, VDUs also have a facility for HARD- VESA /vee-să/ (Video Electronics Stan-
WARE CHARACTER GENERATION that allows dards Association) An organization de-
simple text-only output; this is normally voted to the improvement of standards for
used only for low-level maintenance or video and multimedia devices.
compatibility.
VESA local bus (VL bus; VLB) A local-
vector /vek-ter/ A one-dimensional ARRAY. bus architecture used in the early 1990s on
computers with an Intel 80486 CPU. It was
vector graphics A method of producing superseded on Pentium-based machines by
pictorial images in which the image is the PCI local bus.
stored as mathematical instructions that
define curves, lines, etc., rather than pixels. VFAT /vee-eff-ay-tee/ (virtual file alloca-
See also computer graphics. Compare tion table) An extension of the FAT FILE
raster graphics. SYSTEM, introduced in Windows 95, that
allowed long file names.
Venn diagram /ven/ A diagram used to
show the relationships between sets. The VGA (video graphics array) A computer
232
video conferencing
A B
B⊂A
A B
A∩B
A B
A∪B
233
video display board
between two or more participants in differ- and during execution parts of the program
ent locations achieved by using computer not currently needed remain on backing
networks. Each participant has a video store.
camera and speakers attached to his or her Data and instructions are divided by the
computer, and audio and video data is operating system into blocks of fixed
transmitted over the network. Participants length; these blocks are called pages. When
can use a virtual conference room and ap- a reference to a virtual address is made, the
pear as if they were sitting in a room next system hardware detects whether or not
to each other. Hardware cost has made the required address is physically present in
video conferencing too expensive for the main store. If it is not, an INTERRUPT is gen-
small user in the past, but now inexpensive erated and the page containing the re-
video conferencing software and digital quired location is transferred, by the
cameras are available. operating system, from backing store to
main store. A page is thus the unit of inter-
video display board See video adapter. change between backing store and main
store. Once in main store, special hardware
video graphics board A video adapter or software maps the relevant virtual ad-
that generates the signals for displaying dresses onto the actual physical addresses.
graphical images on a video screen. The When a page held in main store is no
term has fallen into disuse because nearly longer required by the processor, it may be
all modern video adapters have a graphics transferred back to backing store. The
capability. backing-store device is known as a swap-
ping device.
videophone /vid-ee-oh-fohn/ A tele-
phone that also has a camera and screen at- virtual network The connection of sev-
tached and is capable of transmitting and eral LANs together so that they appear as
receiving video as well as audio signals. one network or as a part of a network. For
With conventional telephone lines, the example, workstations on several sites can
only video that can be transmitted is freeze be grouped in the LAN definition in terms
frame. of department or type of user, rather than
by geographical location, making the con-
video RAM (VRAM) The RAM in- trol and management much easier.
cluded on a video adapter board. All types
of video RAM are special arrangements of virtual reality The use of a computer
dynamic RAM and act as a buffer between system to simulate a three-dimensional en-
the CPU and the display. Most forms of vironment and allow the user to experience
video RAM are dual ported, which enables and interact with a realistic set of condi-
the processor to write a new image to the tions, provided by sound, real-time motion
RAM while the display is reading from the effects, and sometimes with tactile feed-
RAM to refresh its current content. back. It is used in such applications as
flight simulation, the design of buildings,
virtual address See virtual memory. medical teaching, and computer games. See
HMD.
virtual memory (virtual storage) The
use of BACKING STORE, especially a hard virtual reality modeling language
disk, as if it were part of the MAIN STORE of (VRML) A language for creating interac-
a computer. In virtual memory systems, tive three-dimensional image sequences
programs can be written requiring more and possible user interaction with them.
main store than is physically present in the Links to VRML files can be embedded in
computer (or more than the OPERATING SYS- HTML documents, but a VRML-enabled
TEM is prepared to allow them). The pro- browser is needed to view such Web pages.
grams can address areas of main store that
do not in fact exist using virtual addresses, virtual storage See virtual memory.
234
VoIP
virus /vÿ-rŭs/ A program or macro code mailbox and can copy, store, or forward
that attaches itself to other programs or the messages as appropriate.
files and can cause data corruption or
hardware failure when activated, usually voice output The ability of a computer
when the program is started or the file to produce spoken messages. Software can
opened. Viruses can be transmitted as at- achieve this by either producing the sounds
tachments to e-mails, by downloading an that make up words or assembling together
infected program, or by using a file from an prerecorded words. See also speech gener-
infected portable disk. See vaccine; worm. ation device.
235
volatile memory
and affect the reliability of VoIP, and it can then used to produce a true three-dimen-
be difficult to trace the geographical loca- sional image. Voxel imaging is used in
tion of emergency calls. Nevertheless, it is magnetic resonance imaging to produce ac-
expected that VoIP will be major growth curate models of the human body. It is also
area in the next few years. See also ENUM; used in computer games.
SIP.
VPN (virtual private network) The use
volatile memory /vol-ă-tăl/ A type of of encryption techniques and tunneling
memory whose contents are lost as soon as protocols by a group of users on a public
the power supply to the memory is network, especially the Internet, to ensure
switched off. Compare nonvolatile mem- that their messages are secure. The net-
ory. work nodes appear to be connected by pri-
vate lines and so form a virtual private
voxel /voks-el/ A volume pixel, i.e. the network. The effect is to give companies
smallest distinguishable box-shaped part the same capabilities as on a dedicated
of a three-dimensional image. Depth is communications system, but at much
added to an image using voxelization. The lower cost.
image is divided into two-dimensional
slices each made up of pixels, and a volume VRAM /vee-ram/ See video RAM.
is produced by stacking slices together in
memory. Opacity transformation, which VRML /vee-ar-em-el, ver-măl/ See vir-
gives voxels different opacity values, is tual reality modeling language.
236
W
wafer /way-fer/ A thin flat piece of semi- W3C /dub-’l-yoo-three-see/ (World Wide
conductor crystal on which multiple copies Web Consortium) A group of industrial
of a circuit are formed, and then cut out, in organizations seeking to promote stan-
the manufacture of integrated circuits. dards for the evolution of the Web and in-
teroperability of Web products. It is
wait state The state in which a program responsible for various standards crucial to
or system or part of a system is inactive the Web, e.g. HTML, XML, etc.
while waiting for some event or process to
finish. For example, a CPU has to enter a WDM (wavelength division multiplex-
wait state if requested data cannot be ing) A type of multiplexing that modu-
transferred from main store quickly lates each of several data streams onto a
enough to match its processing speed. different part of the light spectrum. It was
developed for use with optical fibers and is
wallpaper A pattern or picture, chosen the optical equivalent of frequency division
by the user, that is displayed as a screen multiplexing.
background.
Web See World Wide Web.
WAN /wan/ See wide area network.
Web address See URL.
wand /wond/ A hand-held penlike de-
vice that is connected to a computer and is Web browser A program developed to
used to read printed BAR CODES or OCR allow users to browse documents on the
characters. It is moved steadily over the WWW. A Web browser can also provide
surface of the printing, and an audible access to documents on a network, an in-
and/or visual signal is produced if the data tranet, or a local hard drive. Web browsers
has been sensed satisfactorily. interpret HTML documents and can use
hyperlinks to allow the user to jump from
WAP /wap/ (wireless application proto- document to document in any order. They
col) A specification for the protocol used are also capable of downloading and trans-
by wireless devices, such as cellular tele- ferring files, providing newsgroup access,
phones and radio transceivers, to commu- playing audio and video clips associated
nicate with the Internet. with a document, displaying graphics files,
and running small programs, such as Ac-
warm boot To restart a computer with- tiveX controls. See also Internet Explorer.
out turning the power off, usually by com-
manding the operating system to restart or webcam A digital camera that provides
by pressing the reset button combination. live images to a WEB SERVER, from where
they can be viewed by anybody browsing
warm start See restart. the Web. Webcams do not provide contin-
uous video but rather a stream of still im-
WAV /wayv/ A Windows file format for ages taken at predefined intervals, usually
sounds stored as waveforms. These sound a minimum of about a second. Many web-
files have the file extension wav. cams remain in place for long periods
237
Webcasting
showing panoramic views, for example of WWW and the ability to send and receive
weather conditions on volcanoes in the e-mails on a television set by the use of a
Cascade mountain range or of Times modem contained in a set-top box. A user
Square, New York. Webcams are also used must subscribe to a Web TV network and
in VIDEO CONFERENCING. also have an Internet Service Provider
(ISP).
Webcasting /web-kass-ting, -kahss-/ The
broadcasting of information on the World Webzine /web-zeen/ An electronic mag-
Wide Web. It is received automatically by azine available at a Web site.
all subscribers to the relevant broadcast
service. while loop (while do loop) See loop.
Web server A program that serves the wiki (wee-kee) 1. A WEB SERVER applica-
files that form Web pages to Web users. tion that allows Web users to alter the con-
Every computer on the Internet that con- tents of the WEB PAGES it manages and to
tains Web pages must run a Web server. create new ones. Wikis facilitate the col-
They are often included in a larger package laborative authoring of large Web sites, for
of Internet and intranet programs. example Wikipedia, an online encyclope-
dia. The term ‘wiki’ derives from the
Web site A group of Web pages that are Hawaiian ‘wiki wiki’ (‘fast’ or ‘quick’).
usually connected by hyperlinks and have a 2. A WEB SITE created and managed by such
home page as their starting point. A Web an application. The first wiki was created
browser and an Internet connection are in 1995 for the Portland Pattern Reposi-
needed in order to access a Web site. tory.
238
Wintel
wildcard character /wÿld-kard/ A sym- system in the early 1990s, but suffered
bol used to represent any character or char- from problems of stability and perfor-
acters. The asterisk (*) is commonly used mance. Windows 95 and its successors,
for this purpose for any number of charac- Windows 98 and Windows ME, were true
ters, and the question mark (?) is used to operating systems, but had to make design
represent a single character. Wildcard compromises for BACKWARD COMPATIBILITY
searches are available in many applica- with their MS-DOS and Windows 3.x her-
tions. For example, in word processing itage. A completely different line of Win-
searching for d*g would find all words (of dows development, begun with Windows
any length) starting with ‘d’ and ending NT (Windows new technology) in 1993,
with ‘g’. Searching for d???g would find all was independent of MS-DOS and provided
such words with five letters. better stability, security, and multitasking;
however, it required more expensive hard-
WIMP /wimp/ (windows, icons, mouse, ware and was too slow for some uses, es-
and pointers) A GUI (graphical user in- pecially computer games. It was succeeded
terface) as used in the Macintosh and Win- by Windows 2000 and then by the current
dows operating systems. It was first version, Windows XP, which finally uni-
developed in the 1970s at PARC (Xerox fied the two lines of development. Win-
Palo Alto Research Center).
dows CE is a smaller version of Windows
for use on hand-held and palm computers.
Winchester disk drive /win-chess-ter/ A
See tables overleaf.
type of DISK DRIVE introduced by IBM in the
early 1970s. It was adopted by many man-
Windows 3.x See Windows.
ufacturers and used in large and small com-
puters. It uses one or more hard MAGNETIC
DISKS – known as Winchester disks – that
Windows 95 See Windows.
are mounted inside a hermetically sealed
container together with the read/write
Windows 98 See Windows.
heads and their supporting mechanism. All
modern fixed hard disk units are Winches- Windows 2000 See Windows.
ter disk drives.
Windows CE See Windows.
window 1. A rectangular area on a dis-
play screen inside which a portion of a Windows ME See Windows.
stored image or file can be displayed. The
window can be of any size up to that of the Windows NT See Windows.
screen; more than one window can usually
be displayed at once. The process of mak- Windows open services architecture
ing a portion of an image or file visible (WOSA) A set of application program-
through the window is called windowing. ming interfaces in Microsoft Windows to
The visible window formed by the process enable Windows applications from a vari-
can be moved to allow examination of the ety of sources to communicate with each
different portions making up the entire other.
stored image or file.
2. In a GUI, a similar area on a display Windows XP See Windows.
screen used to display a FORM.
Winsock /win-sok/ The programming
Windows Trademark A Microsoft GUI interface in a Windows operating system
operating system introduced in 1983. Early that handles input and output requests for
versions were in fact MS-DOS applications Internet applications. A comparable inter-
that extended its capabilities and increas- face exists for Macintosh computers.
ingly bypassed its limitations. Windows
3.x became the dominant PC operating Wintel /win-tel/ An Intel-based PC run-
239
WinZip
WINDOWS VERSIONS
ning a Microsoft Windows operating sys- number of triangular faces. See also ren-
tem. dering.
240
word
WINDOWS VERSIONS
Windows 98 June 1998 This integrated Internet Explorer with the operating
system, gave improved USB support and introduced
ACTIVE desktop.
Windows ME September Easier for users to maintain with system restore and
2000 auto update. A fully functional media player and
scanner and camera wizards.
word (machine word; computer word) What the contents of a word signify is
A collection of bits that is treated as a sin- determined entirely by its context. A given
gle unit by the hardware of a computer. bit sequence could represent a MACHINE IN-
The number of bits in a word is called the STRUCTION in one part of main store, while
word length. The word length is fixed in in another part it might represent an IN-
most computers, and in present-day ma- TEGER number, a REAL NUMBER, or per-
chines is usually 32 bits. Main store is di- haps one or more CHARACTERS, depending
vided into either words or BYTES. In the on word length. A 32-bit word, for ex-
former case, each storage location holds ample, could hold integers between
one word, and the word is thus the small- –2 147 483 648 and +2 147 483 647, or
est addressable unit in the computer sys- up to four characters, or a FLOATING-POINT
tem, i.e. the smallest unit to be identified by NUMBER with a PRECISION of about seven
an ADDRESS. digits, or a machine instruction.
241
Word
242
WYSIWYG
wp (WP) See word processing; word write protection A system for the pre-
processor. vention of recording (writing) information
usually on a disk.
wrap-around A feature of a VDU or
printer, allowing it to output lines of text
write-protect notch A small opening in
that would otherwise be too long to be dis-
played or printed in their entirety. The line the jacket of a FLOPPY DISK that is used to
appears as two or more successive lines on prevent the disk from being written to.
the screen or paper. There is a tab in a hole in the corner of the
disk that, when removed, means the disk is
write To record data in some form of write-protected.
storage, usually either in a magnetic
medium or in semiconductor RAM. WWW See World Wide Web.
write head See tape unit.
WYSIWYG /wiz-ă-wig/ (what you see is
write instruction A MACHINE INSTRUC- what you get) What is seen on a screen is
TION that causes an item of data to be an accurate representation of how text,
recorded on some form of storage. graphics, etc., will appear when printed.
243
X
244
XSL
is used to specify rules that manipulate and used to select which parts of the input doc-
restructure all or selected parts of an XML ument are subject to each transformation
document. The output can be an XML or formatting rule. The specification of all
document, but can also be PLAIN TEXT, the rules necessary to achieve a particular
HTML, etc.: a common use of XSLT is to transformation and/or presentation is con-
change XML to HTML so it can be down- tained in a separate file called an XSL
load over the Web. XSL-FO (XSL Format- stylesheet, which is itself a valid XML doc-
ting Objects) is used with XSLT to specify ument. XSL was first proposed in 1997
presentational details (typeface, type size, and was released as a W3C recommenda-
color, page layout, etc.) for the various tion in 2001. XSLT was quickly adopted
parts of an XML document. XSL-FO because its transformation capabilities met
processors typically use these specifications a need among XML developers. XSL-FO,
to produce output in a format that can be however, was taken up much more slowly
viewed on screen or printed, such as POST- and is still not widely used because were al-
SCRIPT or PDF (see Acrobat). The third part ready other methods of formatting XML,
of XSL is the XPATH language, which is e.g. CASCADING STYLE SHEETS.
245
Y
246
Z
247
APPENDIXES
Appendixes
Symbols and Notation
251
Appendixes
Symbols and Notation (continued)
252
Appendixes
Symbols and Notation (continued)
253
Appendixes
Symbols and Notation (continued)
Calculus
increment of x x or δx
limit of function of
x as x approaches a Lim f(x)
x→a
254
Appendixes
File Extensions
There are a large number of different file extensions in use. Some of the more common
extension names are listed below.
255
Appendixes
File Extensions (continued)
256
Appendixes
File Extensions (continued)
.stm HTML file with SSI .wav WAV format audio file
257
Appendixes
Organizational Top-Level Domain Names
Internet top-level domain names are of two types: organizational names and geographic
names. Organizational domain names indicate the type of Internet site and are used as
follows:
.aero air transport industry
.biz businesses
.cat relating to Catalan language or culture
.com any purpose (originally for commercial organizations)
.coop cooperative associations
.edu colleges and universities (mainly US)
.gov US government (nonmilitary)
.info any purpose
.int international organizations established by treaty
.jobs businesses advertising jobs
.mil US military; this has three subdomains:
.af.mil Air Force
.army.mil Army
.navy.mil Navy
.mobi mobile devices and related services
.museum museums
.name individuals
.net any purpose (originally for network-related
organizations)
.org any purpose, but especially nonprofit-making
organizations
.pro professionals (doctors, lawyers, accountants)
.travel travel industry
258
Appendixes
Geographic Top-Level Domain Names
Geographic names indicate the country. When a geographic name is the top-level do-
main, the second-level subdomain in the US may indicate an organization or a state (see
page 264). For other countries, it may indicate an organization. For example
.co.uk a company in the United Kingdom
.ed.uk an educational establishment in the United Kingdom
259
Appendixes
Geographic Top-Level Domain Names (continued)
260
Appendixes
Geographic Top-Level Domain Names (continued)
261
Appendixes
Geographic Top-Level Domain Names (continued)
262
Appendixes
Geographic Top-Level Domain Names (continued)
263
Appendixes
United States Subdomain Names
There are various subdomains in use with the top-level domain name .us. A few
of these refer to organizations, for example:
.lib.us United States library
.state.us state government
Most common subdomain names are used for states, for example:
.ak.us Alaska
.ar.us Arkansas
.ca.us California
.fl.us Florida
.ga.us Georgia
.il.us Illinois
.in.us Indiana
.la.us Louisiana
.ma.us Massachusetts
.md.us Maryland
.mi.us Michigan
.mn.us Minnesota
.ms.us Mississippi
.nc.us North Carolina
.ne.us Nebraska
.nh.us New Hampshire
.ny.us New York (state)
.oh.us Ohio
.ok.us Oklahoma
.or.us Oregon
.va.us Virginia
.vt.us Vermont
.wv.us West Virginia
.wy.us Wyoming
A third-level subdomain may be used for a town or city, as in:
.atl.ga.us Atlanta, Georgia
.nyc.ny.us New York City, New York
.sf.ca.us San Francisco, California
264
Appendixes
Number Conversions
1 1 1 1
2 10 2 2
3 11 3 3
4 100 4 4
5 101 5 5
6 110 6 6
7 111 7 7
8 1000 10 8
9 1001 11 9
10 1010 12 A
11 1011 13 B
12 1100 14 C
13 1101 15 D
14 1110 16 E
15 1111 17 F
16 10000 20 10
17 10001 21 11
18 10010 22 12
19 10011 23 13
20 10100 24 14
21 10101 25 15
22 10110 26 16
23 10111 27 17
24 11000 30 18
25 11001 31 19
26 11010 32 1A
27 11011 33 1B
28 11100 34 1C
29 11101 35 1D
30 11110 36 1E
31 11111 37 1F
32 100000 40 20
33 100001 41 21
34 100010 42 22
35 100011 43 23
36 100100 44 24
37 100101 45 25
38 100110 46 26
39 100111 47 27
40 101000 50 28
265
Appendixes
Number Conversions (continued)
266
Appendixes
Number Conversions (continued)
267
Appendixes
Number Conversions (continued)
268
Appendixes
Number Conversions (continued)
269
Appendixes
Number Conversions (continued)
270
Appendixes
Number Conversions (continued)
271
Appendixes
Web Sites
272
Bibliography
Bibliography
Collins, H. & Pinch, T. The Golem at large: what you should know about technology.
Cambridge, U.K.: Cambridge University Press, 1998.
Cormen, T.H., Leiserson, C.D., Rivest, R.L., & Stein, C. Introduction to Algorithms.
Cambridge, MA: MIT Press, 2001.
Nissanke, N. Introductory Logic and Sets for Computer Scientists. Boston: Addison-
Wesley, 1999.
Patterson, D. & Hennessy, J. Computer Organisation and Design. San Francisco, CA:
Morgan Kaufmann, 1998.
Silberschatz, A., Peterson, J.L., & Galvin, P.C. Operating Systems Concepts. New York:
Addison-Wesley, 1998.
Simon, H.A. The Sciences of the Artificial. Cambridge, MA: MIT Press, 1996.
273