Modeling with Rules Using Semantic

Knowledge Engineering 1st Edition

Grzegorz J. Nalepa (Auth.)
Visit to download the full and correct content document:
https://textbookfull.com/product/modeling-with-rules-using-semantic-knowledge-engin
eering-1st-edition-grzegorz-j-nalepa-auth/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Knowledge Engineering and the Semantic Web 5th

International Conference KESW 2014 Kazan Russia
September 29 October 1 2014 Proceedings 1st Edition
Pavel Klinov
https://textbookfull.com/product/knowledge-engineering-and-the-
semantic-web-5th-international-conference-kesw-2014-kazan-russia-
september-29-october-1-2014-proceedings-1st-edition-pavel-klinov/

Taxonomy Matching Using Background Knowledge Linked

Data Semantic Web and Heterogeneous Repositories 1st
Edition Heiko Angermann

https://textbookfull.com/product/taxonomy-matching-using-
background-knowledge-linked-data-semantic-web-and-heterogeneous-
repositories-1st-edition-heiko-angermann/

Knowledge Graph and Semantic Computing Language

Knowledge and Intelligence Second China Conference CCKS
2017 Chengdu China August 26 29 2017 Revised Selected
Papers 1st Edition Juanzi Li
https://textbookfull.com/product/knowledge-graph-and-semantic-
computing-language-knowledge-and-intelligence-second-china-
conference-ccks-2017-chengdu-china-august-26-29-2017-revised-
selected-papers-1st-edition-juanzi-li/

The Essentials of Data Science Knowledge Discovery

Using R 1st Edition Graham J. Williams

https://textbookfull.com/product/the-essentials-of-data-science-
knowledge-discovery-using-r-1st-edition-graham-j-williams/
Multiphysics Modeling with Application to Biomedical
Engineering 1st Edition Z. Yang

https://textbookfull.com/product/multiphysics-modeling-with-
application-to-biomedical-engineering-1st-edition-z-yang/

Chemical and Biomedical Engineering Calculations Using

Python 1st Edition Jeffrey J. Heys

https://textbookfull.com/product/chemical-and-biomedical-
engineering-calculations-using-python-1st-edition-jeffrey-j-heys/

Knowledge Graph and Semantic Computing Semantic

Knowledge and Linked Big Data First China Conference
CCKS 2016 Beijing China September 19 22 2016 Revised
Selected Papers 1st Edition Huajun Chen
https://textbookfull.com/product/knowledge-graph-and-semantic-
computing-semantic-knowledge-and-linked-big-data-first-china-
conference-ccks-2016-beijing-china-september-19-22-2016-revised-
selected-papers-1st-edition-huajun-chen/

Knowledge Management and Engineering with Decisional

DNA Edward Szczerbicki

https://textbookfull.com/product/knowledge-management-and-
engineering-with-decisional-dna-edward-szczerbicki/

Space Engineering Modeling and Optimization with Case

Studies 1st Edition Giorgio Fasano

https://textbookfull.com/product/space-engineering-modeling-and-
optimization-with-case-studies-1st-edition-giorgio-fasano/
Intelligent Systems Reference Library 130

Grzegorz J. Nalepa

Modeling with
Rules Using
Semantic
Knowledge
Engineering
Intelligent Systems Reference Library

Volume 130

Series editors
Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland
e-mail: kacprzyk@ibspan.waw.pl

Lakhmi C. Jain, University of Canberra, Canberra, Australia;

Bournemouth University, UK;
KES International, UK
e-mail: jainlc2002@yahoo.co.uk; jainlakhmi@gmail.com
URL: http://www.kesinternational.org/organisation.php
About this Series

The aim of this series is to publish a Reference Library, including novel advances
and developments in all aspects of Intelligent Systems in an easily accessible and
well structured form. The series includes reference works, handbooks, compendia,
textbooks, well-structured monographs, dictionaries, and encyclopedias. It contains
well integrated knowledge and current information in the field of Intelligent
Systems. The series covers the theory, applications, and design methods of
Intelligent Systems. Virtually all disciplines such as engineering, computer science,
avionics, business, e-commerce, environment, healthcare, physics and life science
are included.

More information about this series at http://www.springer.com/series/8578

Grzegorz J. Nalepa

Modeling with Rules Using

Semantic Knowledge
Engineering

123
Grzegorz J. Nalepa
AGH University of Science and Technology
Kraków
Poland

ISSN 1868-4394 ISSN 1868-4408 (electronic)

Intelligent Systems Reference Library
ISBN 978-3-319-66654-9 ISBN 978-3-319-66655-6 (eBook)
https://doi.org/10.1007/978-3-319-66655-6
Library of Congress Control Number: 2017950017

© Springer International Publishing AG 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To DeXa, my truest companion
To my Father, the first engineer
To my Mother, the first researcher
Contents

Part I Domain of Rule-Based Systems

1 Rules as a Knowledge Representation Paradigm . . . . . . . . . . . . . . . 3
1.1 Rules in AI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Selected Knowledge Representation Methods . . . . . . . . . . . . . . . 5
1.3 Expert Systems and Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4 Formal Calculi for Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.5 Introduction to Attributive Logics . . . . . . . . . . . . . . . . . . . . . . . . 19
1.6 Rule Languages and Formalization . . . . . . . . . . . . . . . . . . . . . . . 21
1.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2 Knowledge Engineering with Rules . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.1 Knowledge Acquisition and Rule Base Modeling . . . . . . . . . . . . 28
2.2 Automated Inference with Rules . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.3 Structure of Rule Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.4 Knowledge Base Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.5 Rule Interchange and Interoperability . . . . . . . . . . . . . . . . . . . . . 39
2.6 Architectures for Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3 Selected Applications of Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.1 Business Rules Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.2 Rules in Business Process Management . . . . . . . . . . . . . . . . . . . 54
3.3 Rules on the Semantic Web . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.4 Rules in Context-Aware Systems . . . . . . . . . . . . . . . . . . . . . . . . 61
3.5 Rules in Software Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.6 Rules as Programming Paradigm . . . . . . . . . . . . . . . . . . . . . . . . 69
3.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

vii
viii Contents

Part II Formal Models for Rules

4 Formalism for Description of Decision Rules . . . . . . . . . . . . . . . . . . 85
4.1 Attributive Logic with Set Values over Finite Domains . . . . . . . 85
4.2 Inference in ALSV(FD) Formulae . . . . . . . . . . . . . . . . . . . . . . . 90
4.3 Formulation of XTT2 Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.4 Structure of XTT2 Rule Base . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.5 Inference Control in Structured Rule Bases . . . . . . . . . . . . . . . . 100
4.6 Formalized Detection of Anomalies in Rules . . . . . . . . . . . . . . . 103
4.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5 Formalized Integration of Rules and Processes . . . . . . . . . . . . . . . . . 109
5.1 Formal Description of BPMN Process Model . . . . . . . . . . . . . . . 110
5.2 General Business Logic Model . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.3 SKE-Based BPMN Process Model . . . . . . . . . . . . . . . . . . . . . . . 124
5.4 Description of Example Using the Model . . . . . . . . . . . . . . . . . . 126
5.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
6 Prototyping Structure of Rule Bases . . . . . . . . . . . . . . . . . . . . . . . . . 133
6.1 Main Concepts of ARD+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
6.2 ARD+ Method Formalization . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
6.3 Prototyping Structure of the Rule Base . . . . . . . . . . . . . . . . . . . . 142
6.4 Design of Business Processes with Business Rules. . . . . . . . . . . 144
6.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
7 Handling Uncertainty in Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
7.1 Uncertainty in Mobile Systems . . . . . . . . . . . . . . . . . . . . . . . . . . 156
7.2 Improving Uncertainty Handling. . . . . . . . . . . . . . . . . . . . . . . . . 159
7.3 Modified Certainty Factors Algebra for XTT2 Rules . . . . . . . . . 162
7.4 Certainty Factors on Table Level . . . . . . . . . . . . . . . . . . . . . . . . 166
7.5 Time-Parametrised Operators for XTT2 . . . . . . . . . . . . . . . . . . . 168
7.6 Probabilistic Interpretation of XTT2 Models . . . . . . . . . . . . . . . . 170
7.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
8 Formalizing Interoperability in Rule Bases . . . . . . . . . . . . . . . . . . . . 179
8.1 Formalized Approach to Rule Interoperability . . . . . . . . . . . . . . 180
8.2 Data Types, Objects and Attributes . . . . . . . . . . . . . . . . . . . . . . 182
8.3 Taxonomy of Formulae and Operators . . . . . . . . . . . . . . . . . . . . 192
8.4 Rule Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
8.5 Modules and Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
Contents ix

8.6 Complete Knowledge Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206

8.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208

Part III Practical Studies in Semantic Knowledge Engineering

9 Semantic Knowledge Engineering Approach . . . . . . . . . . . . . . . . . . . 211
9.1 Introduction to the Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
9.2 Design Process for Rule-Based Systems . . . . . . . . . . . . . . . . . . . 215
9.3 Architecture of Rule-Centric Applications . . . . . . . . . . . . . . . . . 218
9.4 Overview of the HADEs+ Tool Framework . . . . . . . . . . . . . . . . 221
9.5 Visual Design of a Modularized Rule Base . . . . . . . . . . . . . . . . 224
9.6 HMR Rule Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
9.7 HEART and HEARTDROID Rule Engines . . . . . . . . . . . . . . . . . . . 230
9.8 Rule Analysis with HALVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
9.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
10 Rule Interoperability with Expert System Shells . . . . . . . . . . . . . . . . 245
10.1 Semantically Equivalent Features of Rule Languages . . . . . . . . . 245
10.2 Analysis of the CLIPS Rule Language . . . . . . . . . . . . . . . . . . . . 247
10.3 Analysis of the DROOLS Rule Language . . . . . . . . . . . . . . . . . . . 249
10.4 Formalization of Selected Features of Rule Languages . . . . . . . . 252
10.5 Rule Translation Using a Case Study . . . . . . . . . . . . . . . . . . . . . 260
10.6 Translation of Module Structure . . . . . . . . . . . . . . . . . . . . . . . . . 266
10.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
11 Visual Software Modeling with Rules . . . . . . . . . . . . . . . . . . . . . . . . 275
11.1 Integration of Rules with UML Design . . . . . . . . . . . . . . . . . . . 276
11.2 Representation of ARD+ and XTT2 with UML Diagrams . . . . . 277
11.3 Selected Model Translations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
11.4 Evaluation Using a Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . 291
11.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
12 Using Rules to Support Software Testing . . . . . . . . . . . . . . . . . . . . . 299
12.1 Software Unit Testing with Rules . . . . . . . . . . . . . . . . . . . . . . . . 299
12.2 Decision Table-Based Testing. . . . . . . . . . . . . . . . . . . . . . . . . . . 300
12.3 Framework for Test Unit Generation . . . . . . . . . . . . . . . . . . . . . 303
12.4 Evaluation of the Testing Approach . . . . . . . . . . . . . . . . . . . . . . 308
12.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
x Contents

13 Integrating Business Process Models with Rules . . . . . . . . . . . . . . . . 313

13.1 Motivation and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
13.2 Design and Execution Using the Integrated Approach . . . . . . . . 315
13.3 Model-Based Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
13.4 Complexity Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
13.5 Practical Application of the Model . . . . . . . . . . . . . . . . . . . . . . . 326
13.6 Evaluation of the Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
13.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
14 Rule-Based Systems and Semantic Web . . . . . . . . . . . . . . . . . . . . . . 339
14.1 Integrating SKE with the Semantic Web . . . . . . . . . . . . . . . . . . 339
14.2 DAAL Rule Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
14.3 Hybrid Reasoning with PELLET-HEART . . . . . . . . . . . . . . . . . . . . 347
14.4 DAAL in Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348
14.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
15 Collaborative Knowledge Engineering with Wikis . . . . . . . . . . . . . . 355
15.1 Semantic Knowledge Engineering in LOKI . . . . . . . . . . . . . . . . . 356
15.2 Case Study Implementation in LOKI . . . . . . . . . . . . . . . . . . . . . . 363
15.3 Collaborative Rule Authoring with SBVR . . . . . . . . . . . . . . . . . 369
15.4 Collaborative Modeling of BPMN Models . . . . . . . . . . . . . . . . . 373
15.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
16 Designing Robot Control Logic with Rules . . . . . . . . . . . . . . . . . . . . 381
16.1 Robot Prototyping with Mindstorms NXT . . . . . . . . . . . . . . . . . 381
16.2 PLNXT Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
16.3 Rule-based Control with XTT2 and HEART . . . . . . . . . . . . . . . . 388
16.4 Examples of Control Algorithms . . . . . . . . . . . . . . . . . . . . . . . . 390
16.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
17 Rules in Mobile Context-Aware Systems . . . . . . . . . . . . . ......... 403
17.1 Challenges for Context-Aware Systems on Mobile
Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
17.2 Overview of the KNOWME Architecture . . . . . . . . . . . . . . . . . . . 406
17.3 Knowledge Modeling in KNOWME . . . . . . . . . . . . . . . . . . . . . . . 410
17.4 KNOWME Toolset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415
17.5 Feedback Loop in System Development . . . . . . . . . . . . . . . . . . . 421
17.6 Evaluation Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
17.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 428
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429
Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
Abbreviations

AI Artificial Intelligence
ALSV(FD) Attributive Logic with Set Values over Finite Domains
AMVCBC Adaptable Model, View, Context-Based Controller
ARD Attribute Relationship Diagrams
BPM Business Process Management
BPMN Business Process Model and Notation
BPMS Business Process Management System
BR Business Rules
BRA Business Rules Approach
BRMS Business Rules Management System
CAS Context-Aware Systems
CEP Complex Event Processing
CKE Collaborative Knowledge Engineering
CLIPS C Language Integrated Production System
DAAL Description And Attributive Logic
DL Description Logics
DMN Decision Model and Notation
DSS Decision Support System
DSL Domain Specific Language
DT Decision Table
ES Expert Systems
FOL First Order Logic
HADES HEKATE Design Environment
HALVA HEKATE Verification and Analysis
HATHOR HEKATE Translation framework for Rules
HEART Hybrid Rule RunTime
HEARTDROID HEART for Android
HEKATE Hybrid Knowledge Engineering
HMR HEKATE Meta Representation
HQED HEKATE Qt Editor

xi
xii Abbreviations

IS Intelligent System
KB Knowledge Base
KBS Knowledge-Based Systems
KE Knowledge Engineering
KR Knowledge Representation
LHS Left-Hand Side
LOKI Logic-based Wiki
MDA Model-Driven Architecture
MDD Model-Driven Development
MDE Model-Driven Engineering
MOF Meta Object Facility
MVC Model View Controller
OCL Object Constraint Language
OMG Object Management Group
OPS5 Official Production System
OWL Web Ontology Language
PLWIKI Prolog-based Wiki
PL Predicate Logic
RDF Resource Description Framework
RHS Right-Hand Side
RBS Rule-Based Systems
SE Software Engineering
SKE Semantic Knowledge Engineering
UML Unified Modeling Language
VARDA Visual ARD+ Rapid Development Alloy
XMI XML Metadata Interchange
XTT2 eXtended Tabular Trees, Version 2
List of Figures

Fig. 1.1 Example of frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Fig. 1.2 Example of a semantic network . . . . . . . . . . . . . . . . . . . . . . . . . 7
Fig. 1.3 Example of a decision table . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Fig. 1.4 Example of a decision tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Fig. 1.5 Architecture of a rule-based expert system, after [24] . . . . . . . . 10
Fig. 2.1 Knowledge engineering process . . . . . . . . . . . . . . . . . . . . . . . . . 28
Fig. 2.2 Example of a decision tree corresponding to the decision
table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 32
Fig. 3.1 Decision process modeled using a BPMN gateway and
sequence flow guards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Fig. 3.2 Decision process of the model modeled using a BR task . . . . . 56
Fig. 3.3 Visualization of a rule in BPMN . . . . . . . . . . . . . . . . . . . . . . . . 57
Fig. 4.1 An example of an XTT2 knowledge base . . . . . . . . . . . . . . . . . 97
Fig. 4.2 DDI inference mode for XTT2 . . . . . . . . . . . . . . . . . . . . . . . . . 102
Fig. 4.3 TDI inference mode for XTT2. . . . . . . . . . . . . . . . . . . . . . . . . . 102
Fig. 4.4 GDI inference mode for XTT2 . . . . . . . . . . . . . . . . . . . . . . . . . 103
Fig. 5.1 Conditional sequence flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Fig. 5.2 Conditional (start and intermediate) events . . . . . . . . . . . . . . . . 119
Fig. 5.3 Conditional (non-interruptive and interruptive) boundary
events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Fig. 5.4 Event subprocesses with a conditional start event . . . . . . . . . . . 120
Fig. 5.5 Business rule task (a standard and a call activity task) . . . . . . . 120
Fig. 5.6 Exclusive, inclusive (multi-choice) and complex diverging
gateways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Fig. 5.7 Converging complex gateway . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Fig. 5.8 Gateway after the BR task . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Fig. 5.9 Gateway preceded by a subprocess . . . . . . . . . . . . . . . . . . . . . . 123
Fig. 5.10 Event-based exclusive gateways (non-instantiating and
instantiating) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Fig. 5.11 The BPMN model for the PLI case study . . . . . . . . . . . . . . . . . 128

xiii
xiv List of Figures

Fig. 6.1 Decision table schema (left) and table filled in with rules
(right), after [11] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Fig. 6.2 Possible subgraphs in the ARD+ structure . . . . . . . . . . . . . . . . . 143
Fig. 6.3 Prototyping process model with rules . . . . . . . . . . . . . . . . . . . . 145
Fig. 6.4 Prototyping steps presented in a process model . . . . . . . . . . . . . 147
Fig. 6.5 The “Develop Business Rule task” subprocess . . . . . . . . . . . . . 146
Fig. 6.6 First phase: Generating Business Rule tasks . . . . . . . . . . . . . . . 150
Fig. 6.7 The BPMN model for the PLI case study with forms
and rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Fig. 7.1 Visualization of epistemic and aleatoric sources
of uncertainty [1] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Fig. 7.2 Classification of sources of uncertainty in mobile
context-aware systems [1] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Fig. 7.3 Fragment of an XTT2 model for context-aware system
for applications recommendation and phone profile
switching [1] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
Fig. 7.4 Bayesian network representation of a considered model [1] . . . 172
Fig. 7.5 Bayesian network representation of a time-parametrised
operator from the previous model [1] . . . . . . . . . . . . . . . . . . . . 174
Fig. 9.1 Design process for the Semantic Knowledge Engineering . . . . . 216
Fig. 9.2 Heterogeneous architecture for integrating RBS. . . . . . . . . . . . . 220
Fig. 9.3 The complete HADEs+ environment . . . . . . . . . . . . . . . . . . . . . 221
Fig. 9.4 HQED cell editor dialog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Fig. 9.5 Cashpoint XTT2 model exported from HQED . . . . . . . . . . . . . . 226
Fig. 9.6 XTT2 model in HQED with anomalies detected . . . . . . . . . . . . 227
Fig. 9.7 Example of GUI for HEART callbacks . . . . . . . . . . . . . . . . . . . . 233
Fig. 9.8 HQED executing the Cashpoint model in HEART . . . . . . . . . . . 234
Fig. 9.9 Inconsistency between conditions . . . . . . . . . . . . . . . . . . . . . . . 237
Fig. 9.10 Inconsistency between condition and conclusion . . . . . . . . . . . . 237
Fig. 9.11 Inconsistency between a pair of rules . . . . . . . . . . . . . . . . . . . . 238
Fig. 9.12 The verification process of the model in the HQEd editor . . . . . 239
Fig. 9.13 Combination of states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Fig. 9.14 State tree of the system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Fig. 9.15 Subsumption of a pair of rules . . . . . . . . . . . . . . . . . . . . . . . . . 240
Fig. 10.1 Order of module evaluation in the PLI use case . . . . . . . . . . . . 267
Fig. 10.2 Translation of the bonus-malus module structure . . . . . . . . . . . . 269
Fig. 10.3 Model of PLI case study in XTT2 . . . . . . . . . . . . . . . . . . . . . . . 271
Fig. 10.4 Complete translation of the rule base structure
for the PLI case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Fig. 11.1 Example of UML representation of ARD+ diagram . . . . . . . . . 278
Fig. 11.2 Representation of a TPH diagram corresponding to the
previous ARD model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
Fig. 11.3 Metamodel for the UML representation of ARD+ diagrams . . . 279
Fig. 11.4 Metamodel for the UML representation of TPH diagrams . . . . . 280
List of Figures xv

Fig. 11.5 Example of an XTT2 table . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Fig. 11.6 UML representation of the XTT2 table from Fig. 11.5 . . . . . . . 281
Fig. 11.7 The lower level more complex model example . . . . . . . . . . . . . 282
Fig. 11.8 The higher level model example . . . . . . . . . . . . . . . . . . . . . . . . 283
Fig. 11.9 Metamodel for XTT2 diagrams . . . . . . . . . . . . . . . . . . . . . . . . . 283
Fig. 11.10 Introducing the Output Pin to the metamodel of XTT2 . . . . . . . 284
Fig. 11.11 The flowchart presenting the translation from XMI
to XTT2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Fig. 11.12 The process of simplification of the table representation . . . . . . 287
Fig. 11.13 The process of simplification of the network of tables
representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Fig. 11.14 The translation from TT2 to XMI for the lower level
representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
Fig. 11.15 UML representation of the Cashpoint XTT2 Model . . . . . . . . . 292
Fig. 11.16 UML representation of the domain for the Cashpoint XTT2
Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Fig. 11.17 UML model of the View for Cashpoint . . . . . . . . . . . . . . . . . . 293
Fig. 11.18 Qt GUI for the Cashpoint application . . . . . . . . . . . . . . . . . . . . 293
Fig. 11.19 UML behavioral model of the Controller for Cashpoint
application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Fig. 12.1 Architecture of testing framework based on HaDEs
and Eclipse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
Fig. 12.2 Communication diagram for testing process . . . . . . . . . . . . . . . 305
Fig. 12.3 Rule-based testing perspective in Eclipse. . . . . . . . . . . . . . . . . . 307
Fig. 12.4 List of attributes for testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
Fig. 12.5 Full specification as a XTT2 model . . . . . . . . . . . . . . . . . . . . . . 310
Fig. 12.6 Specification verification results . . . . . . . . . . . . . . . . . . . . . . . . . 311
Fig. 12.7 Resulting JUnit code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Fig. 13.1 Overview of the design approach for BP with BR. . . . . . . . . . . 316
Fig. 13.2 Outline of the Oryx editor’s architecture, after [16] . . . . . . . . . . 318
Fig. 13.3 Screenshot from the prototype Oryx GUI for XTT2,
after [12] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
Fig. 13.4 Exemplary process model in BPWiki . . . . . . . . . . . . . . . . . . . . 319
Fig. 13.5 Fragment of the MOF metamodel abstract syntax,
after [35] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
Fig. 13.6 Business Logic Model in the MOF Modeling
Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
Fig. 13.7 Steps of the approach in the context of the MDE
development process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
Fig. 13.8 The BPMN model for the PLI case study with forms
and rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
Fig. 13.9 The JUnit tests for the PLI model . . . . . . . . . . . . . . . . . . . . . . . 333
Fig. 14.1 Hybrid system combining HEART and Pellet, after [5] . . . . . . . 347
Fig. 15.1 LOKI architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
xvi List of Figures

Fig. 15.2 PIWIKI architecture, after [13] . . . . . . . . . . . . . . . . . . . . . . . . . . 361

Fig. 15.3 Bookstore example in LOKI: semantic annotations . . . . . . . . . . . 364
Fig. 15.4 Bookstore example in LOKI: RDF/XML export . . . . . . . . . . . . . 365
Fig. 15.5 Rule base for the BOOKSTORE system . . . . . . . . . . . . . . . . . . . . . 367
Fig. 15.6 Bookstore example in PLWIKI: wikitext with embedded
HMR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368
Fig. 15.7 User interaction with SBVRwiki . . . . . . . . . . . . . . . . . . . . . . . . 371
Fig. 15.8 Diagram visualization with PlantUML . . . . . . . . . . . . . . . . . . . . 371
Fig. 15.9 EU rent rule syntax highlighting . . . . . . . . . . . . . . . . . . . . . . . . 372
Fig. 15.10 EU rent fact visualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
Fig. 15.11 BPMN model for the Order Fulfillment Process . . . . . . . . . . . . 375
Fig. 15.12 BPMN model of PLI in BPwiki . . . . . . . . . . . . . . . . . . . . . . . . 377
Fig. 16.1 LEGO Mindstorms NXT brick hardware, after [5] . . . . . . . . . . 383
Fig. 16.2 PlNXT layered architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
Fig. 16.3 HeaRT/PlNXT runtime architecture . . . . . . . . . . . . . . . . . . . . . . 388
Fig. 16.4 Third conceptual level of the line following robot . . . . . . . . . . . 390
Fig. 16.5 History of the design of the line following robot . . . . . . . . . . . . 391
Fig. 16.6 Last conceptual level of the line following robot . . . . . . . . . . . . 391
Fig. 16.7 XTT logic model of the line following robot . . . . . . . . . . . . . . . 391
Fig. 16.8 Initial state definition of the line following robot . . . . . . . . . . . . 392
Fig. 16.9 Third conceptual level of the evader robot . . . . . . . . . . . . . . . . 393
Fig. 16.10 History of the conceptual design transformations
of the evader robot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
Fig. 16.11 Last conceptual ARD level of the evader robot . . . . . . . . . . . . . 394
Fig. 16.12 XTT2 logic model of the evader robot . . . . . . . . . . . . . . . . . . . 395
Fig. 16.13 Design of the third and the forth conceptual level
of the cleaner robot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396
Fig. 16.14 Last conceptual level of the cleaner robot . . . . . . . . . . . . . . . . . 396
Fig. 16.15 History of the conceptual design transformations
of the cleaner robot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
Fig. 16.16 State diagram of the cleaner robot . . . . . . . . . . . . . . . . . . . . . . . 398
Fig. 16.17 XTT logic model of the cleaner robot . . . . . . . . . . . . . . . . . . . . 396
Fig. 17.1 Trade-off between different modeling and architectural
approaches with respect to the mobile CAS requirements,
after [5] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
Fig. 17.2 Adaptable Model, View, Context-Based-Controller
(AMVCBC) Architecture [5] . . . . . . . . . . . . . . . . . . . . . . . . . . . 408
Fig. 17.3 Example of a decision tree from a question forest . . . . . . . . . . . 415
Fig. 17.4 State interpretation in the HEARTDROID inference engine. . . . . . 418
Fig. 17.5 Phases of building mobile CAS supported by KNOWME . . . . . . 422
Fig. 17.6 Decision tree for the prediction of mobile network usage . . . . . 424
Fig. 17.7 Decision tree generated with uncertain data . . . . . . . . . . . . . . . . 424
Fig. 17.8 XTT2 table generated from the uncertain decision tree . . . . . . . 425
Fig. 17.9 XTT2 model for intelligent personal assistant system . . . . . . . . 426
Fig. 17.10 Time efficiency results for the most popular rule engines . . . . . 427
Introduction

Knowledge is a deadly friend

when no one sets the rules

—KING CRIMSON

Sorrow never comes too late

and happiness too swiftly flies (…)
where ignorance is bliss
it is folly to be wise

—THOMAS GRAY

Motivation

This book is dedicated to a systematic presentation of my research results in the

area of intelligent rule-based systems. I conducted this work with my Team,
GEIST1 together with colleagues, students at AGH UST in Kraków, with collab-
oration with many colleagues in Europe, in the years 2008–2016. My original
research background combines several areas. As I have graduated from automation
and robotics, I have a basic knowledge in automatic control, dynamic systems, as
well as different hardware platforms, including embedded systems. On the other
hand, in my Ph.D. thesis and afterwards it, I worked mainly in computer science
and artificial intelligence. I focused on symbolic knowledge representation, rea-
soning, and knowledge engineering. In the group that I worked in, there was also
strong research in the fields of data bases, software engineering, and formal
methods. After my Ph.D., with my team I explored several new areas, including
Semantic Web, Business Process Management, and more recently context-aware

1
See the webpage http://geist.re.

xvii
xviii Introduction

systems. These topics influenced my research, emphasizing aspects of modeling,

design and analysis of intelligent systems.
The core of my research in these years was related to rule-based knowledge
representation methods. This was the main research thread, that I explored and
extended in the areas mentioned above. Therefore, the single most important focus
of this book is on rules. As such, this book aims at providing an academic, yet
practical perspective on knowledge engineering with rule-based systems. The focus
of the book is on the Semantic Knowledge Engineering approach (SKE) I devel-
oped. In the next section I introduce the structure and contents of the book. As the
book contains the synthesis of my most important works in last eight years, I
discuss specific results and publications that were the foundation for the specific
chapters. These results were often delivered within the scope of the research pro-
jects that I was involved in. After that, I summarize the original and specific
perspective of this book and the targeted audience. The book is enriched with
selected examples of designs of intelligent systems that I briefly describe after-
wards. After 20 years in research and development, it is clear to me that one of the
most important factors in doing it, are the right people to cooperate with. I discuss
in more detail particular collaborators and contributors in the last section of this
introduction. Without these persons, this book might not have come into existence.

Structure of the Book

Organizing this material was not a trivial task for me. When I worked on specific
topics there has always been a “big picture” of how to possibly incorporate research
threads in a diverse, yet single and coherent structure. However, not all research
produced the results that I wished, and clearly the resulting structure was different
to what I had originally envisioned. While working on this book I took my
habilitation monograph [1] as a starting point. I selected and then reworked the
presentation of the most important results that provided the foundation for my post
habilitation work.
I decided to partition the contents of the book into three different parts. The first
part presents selected aspects of the state of the art in the area of rules in AI and
software engineering (in a broad sense). In Chap. 1, I aimed at emphasizing the
roots of my work in classic AI and knowledge representation. Chapter 2 identifies
challenges as well as methods relevant for the engineering practice. Then, in Chap.
3. I discuss these recent applications of rules which are interesting for me, and
relevant for the rest of the book. These are related to software engineering, business
systems, and web applications.
The core part of the book includes several related models presented in the second
part. Clearly, the most important one is XTT2, based on ALSV(FD) presented in
Chap. 4. This main model was augmented with several others. The inference
control in the XTT2 decision network can be delegated to a BP engine, as discussed
Introduction xix

in Chap. 5. The design of the structure on the XTT2 knowledge base can be
efficiently supported and simplified by the ARD+ method, as presented in Chap. 6.
Two important extensions of the base XTT2 formalism were developed. The first is
related to uncertain (or incomplete) knowledge, and is covered in Chap. 7. The
second one is oriented at providing knowledge interoperability, and is discussed in
Chap. 8.
The models from the second part were developed and used in several specific
domains discussed in the third part. This part begins with Chap. 9 which presents
the SKE approach, including the specific design process for intelligent systems, and
tools supporting it. After it, a number of case studies where SKE was used, are
presented. I begin with the discussion of knowledge interoperability techniques in
Chap. 10. This work shows how XTT2 -based systems can be translated to the
languages used in classic rule-based system shells. The translation uses the model
from Chap. 8. Then, I move to software engineering, to demonstrate in Chap. 11,
how XTT2 can be used with UML to design software. In this theme, in Chap. 12.
I show the application of XTT2 to the automation of test cases in software testing.
In Chap. 13 I remain in the area of business applications where I present practical
integration of business processes with rules encoded with XTT2. To this end the
model previously introduced in Chap. 5 was used. The two next chapters consider
knowledge representation on the Web. In Chap. 14 I discuss the DAAL language
that allows for an integration of ALSV(FD) with Description Logics, opening up
opportunities for using XTT2 rules in the Semantic Web applications. In Chap. 15
the LOKI platform is introduced which provides a Collaborative Knowledge
Engineering environment. It is a web-based engineering environment based on the
concept of Semantic Wikis. The last two chapters describe case studies closer to
specific hardware platforms. In Chap. 16 I discuss the use of the SKE design and
application of XTT2 rules to the modeling of control logic for mobile robots. Then
in Chap. 17 I demonstrate how XTT2 rules can be put in operation on mobile
platforms to support reasoning in context-aware systems. I use the model previously
introduced in Chap. 7 to handle uncertainty which is omnipresent in such systems.
The contents of the book form a synthesis of selected important results that were
partially made available in some previous works described next.

Previous Related Research

In this section I mention the works conducted, and papers published after the
habilitation monograph [1], that are related to this book.
The first part of the book gives an introduction to the domain, as well as a certain
perspective on it. The starting point for Chaps. 1–3 was the Chap. 2 of [1]. Clearly,
the discussion was much extended and focused. It was also enriched and illustrated
by selected examples and analysis from [2–4].
xx Introduction

The second part presents the formal models I developed for rule-based systems.
Chapter 4 is a reworked and extended version of the Chap. 4 of [1]. Chapter 5
discusses the model I developed with Dr. Krzysztof Kluza in his Ph.D. thesis [3]. It
is based on the discussion in [5]. The results of these works were also presented in
our papers [6–8]. Chapter 6 presents the formalized version of the ARD+ method.
This chapter is partially based on parts of Chap. 5 of [1], our report [9], Chap. 6 of
[3], and paper [10]. Chapter 7 discusses the model I developed with Dr. Szymon
Bobek in his Ph.D. thesis [4]. It is based on the results presented in [11], and in
[12]. Chapter 8 discusses the model I built with Dr. Krzysztof Kaczor in his Ph.D.
thesis [2]. This chapter uses and extends the results from papers [13–15].
The discussion of case studies in the third part begins with the presentation of the
SKE approach in Chap. 9. That chapter extends the Chap. 5 of [1], and includes
more recent work on the HaDEs + toolset. Chapter 10 contains the presentation of
selected practical results of the thesis [2]. It is also partially based on the paper [13].
Then Chap. 11 is a reworked version of Chap. 7 of [1]. Chapter 12 is a synthesis of
results regarding software, preliminary published in [16–18]. In Chap. 13 selected
practical results of the thesis [3] are discussed, also partially presented in [5].
Chapter 14 is partially based on the results from Chap. 6 of [1]. Chapter 15 presents
a synthesis of work on semantic wikis, also partially published in [19–22]. Chapter
16 is based on the thesis [23] and on an extended version of [24]. Finally, the work
in Chap. 17 is related to research discussed in [25–28]. The Chapter presents
selected original results discussed in the Ph.D. Thesis [4]. Moreover, it presents
some results also discussed in [11, 12].
The results presented in the book were often achieved in the scope of research
projects I was involved in together with my team and my collaborators.

Projects

The first project that gave early ideas for my work was Regulus (KBN 8 T11C 019
17), where tabular knowledge representation methods were considered. An
important first project was Mirella (KBN 4 T11C 027 24), as it provided the very
first version of the XTT method. The primary motivation for this work and initial
results were first delivered in the HeKatE project (N516 024 32/2878). HeKatE
(Hybrid Knowledge Engineering) resulted in number of methods, including ALSV
(FD) and ARD+. The objectives were to investigate the possible bridging of
knowledge engineering and software engineering. The second important project
was BIMLOQ (N516 422338). The main aim of the project was to build a
declarative model for business processes, including business rules specification,
with an emphasis on the analysis and optimization of those processes. In that
project preliminary work on BP integration with semantic wiki was carried out. The
next project was Parnas (N516 481240) which was aimed at developing tools for
Introduction xxi

inference control and quality analysis in modularized rule bases, and provided the
final formalized version of the XTT2 method.
All of the above mentioned projects regarded fundamental research in the area of
computer science. An important step in the application of selected results was the
Prosecco project (NCBiR PBS1/B3/14/2012). In this project my team had the
opportunity to apply methods and tools we developed in the previous projects. This
included the work on business processes, business rules, and semantic web
methods.
Finally, I supervised three Preludium grants from NCN supporting work related
to the theses of my Ph.D. students. This included the SaMURAI project
(UMO-2011/03/N/ST6/00886) regarding the work on semantic interoperability of
rules, that supported the thesis [2]. Then the HIBUPROBURUL Project
(UMO-2011/03/N/ST6/00909) supported the work on the integration of business
rules and processes considered in the thesis [3]. Finally, the KnowMe Project
(UMO-2014/13/N/ST6/01786) supported the thesis [3].
More information on the mentioned projects can be found on the webpage of the
GEIST team at http://geist.re. I wish to thank NCN and NCBR for financing these
research projects.

Perspective and Intended Audience

The domain of rule-based systems is a rich research field with a long tradition.
Clearly, there are now many textbooks in this area. There is also a lot of important
original research conducted in last 40 years. I discuss a selection of these in the first
part of the book.
The perspective of this book, is on developing intelligent systems based on rules.
Today this area is on the intersection of original knowledge engineering and
software engineering, we aim at combining these approaches by appropriate models
and tools. I am interested in the practical design of rule bases, using visual repre-
sentations. Moreover, my focus is on formal models knowledge for the design and
analysis of rule-based systems. Finally, a very important aspect, is the integration of
rule-based systems with selected current methods and tools. I consider mainly
selected areas in software engineering and business systems.
From my perspective, rules can serve as a high level decision and control logic.
Systems can incorporate such logic in a heterogeneous manner. It can be combined
with business processes for inference control, and with ontologies for extending
vocabulary or type systems. The rule-based core can be considered as a dynamic
system, where rules describe the transition between system states, thus capturing the
dynamics of the system.
The intended audience of this book are researchers and practitioners in the field
of intelligent systems and software engineering. From a conceptual point of view,
the second part of the book contains formal models, which are the results of
xxii Introduction

fundamental research. The applied research results, including practical applications

of these models are described in the third part of the book. The book could also
prove to be useful for Ph.D. students in the areas related to intelligent systems,
software engineering, and business systems.

Examples in the Book

In the book I use a number of examples to illustrate the main methods and concepts.
Some of them are general enough to be used in several chapters, while others are
more specific so are used to demonstrate only selected features. Most of the
examples are related to three main system cases described below.
The BOOKSTORE system case is first introduced in Chap. 4, and then used in
Chaps. 14, and 15. The BOOKSTORE system is a simple rule-based book recom-
mender. It is used to demonstrate the features of the ALSV(FD) and XTT2 .
Furthermore, I use it to discuss applications of the SKE approach to the Semantic
Web and CKE . The case was developed together with Weronika T. Adrian and first
used in [1].
The PLOC system case is the extension of the previous case PLI used in [1].
The PLI case was first modeled by Szymon Książek. It is a rule-based decision
support system that calculates car insurance rate, based on historical data in Poland.
PlOC was extended and formalized by Krzysztof Kaczor and Krzysztof Kluza in
their Ph.D. theses. The case is used in several chapters, including Chaps. 5, 6, 8,
and 10, 13, to demonstrate the ARD+ design method, then the integration of rules
and processes, and finally rule base interoperability.
The Cashpoint system case was also previously modeled with XTT2 and used in
[1]. It is a simple model of an ATM system for cash withdrawal. The specification
considered by us is based on the system presented by Denvir et al. in [29] and by
Poizat and Royer in [30]. In this book it is used to demonstrate HaDEs + tools in
Chap. 9, and the integration of the SKE design process with UML in Chap. 11.
Moreover, in Chaps. 7, and 17 several small cases from [4], developed with
Szymon Bobek, are used in the discussion of context-aware systems.
Finally, during the HeKatE project, several cases for the SKE approach were
developed. These are available online at http://ai.ia.agh.edu.pl/wiki/hekate:cases:
start.

Acknowledgments

This book contains a synthesis of results of my research in last eight years. After all
these all these years of doing research, I very much realize, that besides research
topics and results, what matters a lot are the people I work with. I was very lucky in
Introduction xxiii

these years to meet and cooperate with many unique and talented persons.
Moreover, research is done by individuals. Even if they form teams, their indi-
viduality is what makes these teams successful and their research fruitful. Without a
number of individuals my research and this book would not be possible.
This books summarizes almost a decade of research, started a few years after my
Ph.D. [31] in 2004. First of all I wish to thank Prof. Antoni Ligeza, who was the
supervisor of my Ph.D. thesis, who has been a good colleague of mine since than.
His concept of formalized algebra for rules laid foundation for the W-trees and the
original XTT method. Our shared work then included the ALSV(FD) logic which
allowed me to formalize XTT2 in [1] and in [32]. We worked together on many
of the previously mentioned projects, including Regulus, HeKatE, and BIMLOQ.
These projects would not be possible without a great number of talented young
people. My special thanks and gratitude go to my Ph.D. students. I would like also
thank them for their permission to include some examples and studies from their
theses in this book. Krzysztof Kaczor worked for many years on the HQEd editor
for XTT2 and we formulated a formalized model for rule interoperability [2].
Krzysztof Kluza extended my early work on integration of business rules and
processes [33]. We provided a formalized model for such an integration in [3]. With
Szymon Bobek, together we boldly went where no one in our team had gone
before. We successfully developed research in context-aware systems, and learned
many completely new methods and tools from the area of machine learning and
data mining. Building on our previous work on XTT2 and the HEART engine, we
delivered KNOWME in [4] and several important research papers. Szymon together
with Mateusz Ślażyński developed the HEARTDROID engine. Krzysztof Kutt is
finishing his thesis on collaborative knowledge engineering. Szymon and Krzysztof
also gave me many valuable remarks on this book.
So far, I have supervised over 40 master and bachelor theses. Most of them were
closely related to my research. A number of my students developed valuable results
and we wrote papers together. Weronika T. Adrian worked with me on DAAL and
other topics related to the Semantic Web. Piotr Hołownia developed PlNXT mid-
dleware, and then Maciej Makowski and Błażej Biesiada developed it to support
XTT2 for programming Mindstorms NXT robots. Krzysztof Kotra developed the
prototype of LOKI/PLWIKI, then Janusz Kamieński and Mirosława Ozgowicz
enhanced it towards semantic wikis. Marta Woźniak developed the SBVR plugin
for Loki and Urszula Ciaputa provided the BPMN plugin. Finally, Olgierd Grodzki
worked on context-aware systems, including in-door micro-localization.
I am in debt with many distinguished professors that supported me in the last
decade, and I wish to thank them for their support. Prof. Ryszard Tadeusiewicz
originally attracted me to AI with his lectures and research, and for all of my years
at AGH UST offered me his guidance and continuous support. Prof. Tomasz Szmuc
shared knowledge about from software engineering and supported we when I
needed it, including the Prosecco project. Prof. Mariusz Flasiński gave me valuable
remarks on the “big picture on AI”. Prof. Stan Matwin has supported me strongly
since the first days of the Polish AI Society (PSSI), and also gave many important
ideas about machine learning and my research. Prof. Jerzy Stefanowski, not only
xxiv Introduction

supported me with PSSI, but also shared many valuable comments regarding my
research and its development towards machine learning. Prof. Paweł Gryboś has
been encouraging me to finish writing this book and related papers for a long time
—I am very grateful for his support. Last, but very much not least I want to thank
Prof. Janusz Kacprzyk, who motivated and supported me since my habilitation, and
made this book possible.
I wish to thank my good colleagues who supported me on many levels and from
whom I learned a lot.2 I would like to thank Prof. Marcin Szpyrka. We have been
working together for over 17 years. I have always had a great respect and admi-
ration for his passion for research, and endless motivation for his hard work. I also
wish to thank him for all his valuable remarks he gave me on this book and the Ph.
D. theses of my students. My thanks also go to Prof. José Palma Mendez for his
friendliness and openness during all my stays in Spain. This included my recent
stay in Murcia where he created a friendly working atmosphere and shared many
books that helped me in improving this book. I wish to thank Prof. Jerzy Pamin for
all his support and friendship. I very much appreciate all of our meetings, talks, and
email conversations. Finally, I very much wish to thank my good friend, Prof.
Joachim Baumeister, PD. We have been colleagues and friends for over a decade
now. What started as a small collaboration on a startup workshop, turned out to be a
great collaboration on the topics of knowledge engineering and software engi-
neering. We developed KESE (http://kese.ia.agh.edu.pl) into a recognized scientific
event and built a vibrant scientific community. This has been a very important
professional and personal experience for me.
Thank you all!
Kraków 2014—Murcia 2016—Kraków 2017

References

1. Nalepa, G.J.: Semantic Knowledge Engineering. A Rule-Based Approach. Wydawnictwa AGH,

Kraków (2011)
2. Kaczor, K.: Knowledge formalization methods for semantic interoperability in rule bases. PhD
thesis, AGH University of Science and Technology (2015) (Supervisor: Grzegorz J. Nalepa)
3. Kluza, K.: Methods for Modeling and Integration of Business Processes with Rules. Ph.D.
thesis, AGH University of Science and Technology (March 2015) Supervisor:
Grzegorz J. Nalepa
4. Nalepa, G.J., Bobek, S.: Rule-based solution for context-aware reasoning on mobile devices.
Comput. Sci. Inf. Syst. 11(1), 171–193 (2014)
5. of Business Processes Integrated with Business Rules, F.M.: Krzysztof kluza and grzegorz j.
nalepa. Information Systems Frontiers (2016) submitted.
6. Kluza, K., Nalepa, G.J., Lisiecki, J.: Square complexity metrics for business process models. In
Mach-Król, M., Pełech-Pilichowski, T. (eds.): Advances in Business ICT. Advances in
Intelligent Systems and Computing, vol. 257, pp. 89–107. Springer (2014)

2
On a more private note, I want to warmly thank Dorota and Chris for their support, and their help
in improving my English in this book.
Introduction xxv

7. OMG: Business Process Model and Notation (BPMN): Version 2.0 specification. Technical
report formal/2011-01-03, Object Management Group (January 2011)
8. Kluza, K., Kaczor, K., Nalepa, G.J.: Enriching business processes with rules using the
Oryx BPMN editor. In Rutkowski, L., et al. (eds.): Artificial Intelligence and Soft Computing:
11th International Conference, ICAISC 2012: Zakopane, Poland, April 29–May 3, 2012.
Lecture Notes in Artificial Intelligence, vol. 7268, pp. 573–581. Springer (2012)
9. Nalepa, G.J., Wojnicki, I.: ARD+ a prototyping method for decision rules. method overview,
tools, and the thermostat case study. Technical Report CSLTR 01/2009, AGH University of
Science and Technology (2009)
10. for Generation, A.M., of Business Processes with Business Rules, D.: Krzysztof kluza and
grzegorz j. nalepa. Information and Software Technology (2016) submitted.
11. Bobek, S., Nalepa, G.J.: Uncertain context datamanagement in dynamic mobile environments.
Future Gener. Comput. Syst. 66, 110–124 (2017)
12. Bobek, S., Nalepa, G.J.: Uncertainty handling in rule-based mobile contextaware systems.
Pervasive and Mobile Computing (2016)
13. Kaczor, K., Nalepa, G.J.: Enabling collaborative modeling of rule bases by semantic
interoperability methods. Future Generation Computer Systems (2015) Submitted.
14. Kaczor, K., Nalepa, G.J.: Formalization of production rule representation model for multilevel
approach to rule interoperability. Data and Knowledge Engineering (2015) submitted.
15. Kaczor, K., Nalepa, G.J.: Encapsulation-driven approach to interchange of knowledge base
structure. Lect. Notes Softw. Eng. 4(1), 66–72 (2016)
16. Nalepa, G.J., Kaczor, K.: Proposal of a rule-based testing framework for the automation of the
unit testing process. In: Proceedings of the 17th IEEE International Conference on Emerging
Technologies and Factory Automation ETFA 2012, Kraków, Poland, 28 September 2012.
(2012)
17. Kutt, K.: Proposal of a rule-based testing framework. Master's thesis, AGH University of
Science and Technology (July 2013) Supervisor: G. J. Nalepa.
18. Nalepa, G.J., Kutt, K., Kaczor, K.: Can the generation of test cases for unit testing be
automated with rules? In Rutkowski, L., [et al.], eds.: Arti cial Intelligence and Soft
Computing: 13th International Conference, ICAISC 2014: Zakopane, Poland. Volume 8468 of
Lecture Notes in Arti cial Intelligence., Springer (201 548–599
19. Nalepa, G.J., Kluza, K., Kaczor, K.: Sbvrwiki a web-based tool for authoring of business rules.
In: Rutkowski, L., et al. (eds.) Artificial Intelligence and Soft Computing: 14th International
Conference, ICAISC 2015. Lecture Notes in Artificial Intelligence, pp. 703–713. Springer,
Zakopane, Poland (2015)
20. Nalepa, G.J., Kluza, K., Ciaputa, U.: Proposal of automation of the collaborative modeling and
evaluation of business processes using a semantic wiki. In: Proceedings of the 17th IEEE
International Conference on Emerging Technologies and Factory Automation ETFA 2012,
Kraków, Poland, 28 Sept 2012. (2012)
21. Nalepa, G.J.: Collective knowledge engineering with semantic wikis. J. Univers. Comput. Sci.
16(7), 1006–1023 (2010)
22. Nalepa, G.J.: Loki – semantic wiki with logical knowledge representation. In Nguyen, N.T.
(ed.) Transactions on Computational Collective Intelligence III. Lecture Notes in Computer
Science, vol. 6560, pp. 96–114. Springer, Berlin (2011)
23. Biesiada, B.: lnxt enhancements proposal. Master's thesis, AGH University of Science and
Technology (2011)
24. Nalepa, G.J., Biesiada, B.: Declarative design of control logic for mindstorms NXT with XTT2
method. In Jedrzejowicz, P., Nguyen, N.T., Hoang, K., eds.: Computational Collective
Intelligence. Technologies and Applications - Third International Conference, ICCCI 2011,
Gdynia, Poland, September 21-23, 2011, Proceedings, Part II. Volume 6923 of Lecture Notes
in Computer Science., Springer (2011) 150 159
25. Bobek, S.: Methods for modeling self-adaptive mobile context-aware systems. Ph.D. thesis,
AGH University of Science and Technology (April 2016) Supervisor: Grzegorz J. Nalepa
xxvi Introduction

26. Bobek, S.,Nalepa,G.: Compact representation of conditional probability for rule-basedmobile

context-aware systems. In: Bikakis, A., Fodor, P., Roman, D. (eds.) Rules on the Web. From
Theory to Applications. Lecture Notes in Computer Science. Springer International Publishing
(2015)
27. Bobek, S., Dziadzio, S., Jaciów, P., ‘la»y«ski, M., Nalepa, G.J.: Understanding Context with
ContextViewer Tool for Visualization and Initial Preprocessing of Mobile Sensors Data. In:
Modeling and Using Context: 9th International and Interdisciplinary Conference, CONTEXT
2015, Lanarca, Cyprus, November 2- 6,2015. Proceedings. Springer International Publishing,
Cham (2015) 77–90
28. Bobek, S., Nalepa, G.J.: Incomplete and uncertain data handling in context-aware rule-based
systems with modified certainty factors algebra. In: Bikakis, A., Fodor, P., Roman, D. (eds.)
Rules on the Web. From Theory to Applications. Lecture Notes in Computer Science, vol.
8620, pp. 157–167. Springer International Publishing (2014)
29. Denvir, T., Oliveira, J., Plat, N.: The cash-point (ATM) ‘Problem’. Form. Asp. Comput. 12(4),
211–215 (2000)
30. Poizat, P., Royer, J.C.: Kadl specification of the cash point case study. Technical report,
IBISC, FRE 2873 CNRS - Universite d’Evry Val d’Essonne, France, Genopole Tour Evry 2,
523 place des terrasses de l’Agora 91000 Evry Cedex (2007)
31. Nalepa, G.J.: Meta-Level Approach to Integrated Process of Design and Implementation of
Rule-Based Systems. PhD thesis, AGH University of Science and Technology, AGH Institute
of Automatics, Cracow, Poland (September 2004)
32. Nalepa, G.J., Lige˛za, A., Kaczor, K.: Formalization and modeling of rules using the XTT2
method. Int. J. Artif. Intell. Tools 20(6), 1107–1125 (2011)
33. Nalepa, G.J.: Proposal of business process and rules modeling with the XTT method. In Negru,
V., et al. (eds.) Symbolic and numeric algorithms for scientific computing, 2007. SYNASC
Ninth international symposium. September 26–29, Los Alamitos, California, Washington,
Tokyo. IEEE Computer Society. IEEE, CPS Conference Publishing Service, pp. 500–506
September 2007
 Part I
Domain of Rule-Based Systems

In this first introductory part of the book, we give an overview of selected topics
related to practical and current knowledge engineering with rule-based systems.
Clearly, we do not to provide a thorough textbook-like treatment of this domain. In
fact, a number of very good books are available. A short and concise introduction to
expert systems is given in general Artificial Intelligence (AI) books [1–3]. A more
detailed treatment of these topics was given in [4–8]. Important books on Intelligent
Systems also provide a nice introduction to rules [9, 10] while giving a broader
perspective. More recently, the book [1] provided a detailed treatment of formalized
approaches to rule-based systems in general. Moreover, the edited book [12] gave a
broad overview of recent important challenges in the domain of rules as well as
applications of rule-based systems. We use them as a reference and starting point
for this part of book, which was partitioned into three chapters.
Chapter 1 starts from the classic perspective of symbolic AI, where rules are
treated as one of the main knowledge representation methods [13]. Among simple
rules found in the production rule systems a number of other important methods can
be identified [14]. In fact, when considering a formalized description some of them
(e.g. decision tables and decision trees) can be converted to rules [11]. They will be
in our focus in the remaining part of the book.
Chapter 2 addresses the selected important topics in practical knowledge engi-
neering with rules [6, 15]. This includes issues of knowledge acquisition and
modeling [16]. Since we focus on formalized methods for rule representation, it
becomes possible to formally analyze the contents of rule bases [17]. Another
important issue introduced in this chapter is the interchange of rule base knowledge
between different systems.
The objective of Chap. 3 is to present important and recent applications of
rule-based technologies. A much more detailed review is given in the book [12]. Here,
we focus on applications related to business systems in a very broad sense. This
includes business rules and processes, as well semantic representations. Moreover, we
2

indicate the importance of rule-based approach in software engineering and

context-aware systems.
To summarize, our perspective in this book, including this part, will be on
reasoning with rules in intelligent systems. We are interested in the practical design
of rule bases with different visual representations. Moreover, we focus on a formal
description of knowledge for design and analysis. Finally, we consider the inte-
gration of rule-based systems with selected current methods and tools considered in
software engineering and business systems.

References

1. Nilsson, N.J.: Artificial Intelligence: A New Synthesis, 1st edn. Morgan Kaufmann Publishers
Francisco, CA, USA (1998) Inc., San Francisco (1998)
2. Negnevitsky, M.: Artificial Intelligence. A Guide to Intelligent Systems. Addison-Wesley,
Harlow (2002). ISBN 0-201-71159-1
3. Mariusz, F.: Wstęp do sztucznej inteligencji. WN PWN (2011)
4. Giarratano, J.C., Riley, G.D.: Expert Systems. Thomson, Toronto (2005)
5. Genesereth,M.R., Nilsson, N.J.: Logical Foundations for Artificial Intelligence.Morgan
Kaufmann Publishers Inc., Los Altos (1987)
6. Gonzalez, A.J., Dankel, D.D.: The Engineering of Knowledge-Based Systems: Theory and
Practice. Prentice-Hall Inc, Upper Saddle River (1993)
7. Waterman, D.A.: A Guide to Expert Systems. Addison-Wesley Longman Publishing Co. Inc.,
Boston (1985)
8. David, J.M., Krivine, J.P., Simmons, R. (eds.): Second Generation Expert Systems. Springer,
Secaucus (1993)
9. Torsun, I.S.: Foundations of Intelligent Knowledge-Based Systems. Academic Press, London
(1995)
10. Hopgood, A.A.: Intelligent Systems for Engineers and Scientists, 2nd edn. CRC Press, Boca
Raton (2001)
11. Ligęza, A.: Logical Foundations for Rule-Based Systems. Springer, Berlin (2006)
12. Giurca, A., Gašević, D., Taveter, K. (eds.) Handbook of Research on Emerging Rule-Based
Languages and Technologies: Open Solutions and Approaches. Information Science
Reference. Hershey, New York (May (2009)
13. Hendler, J., van Harmelen, F.: The semantic web: webizing knowledge representation. Hand
book of Knowledge Representation. Elsevier, New York (2008)
14. van Harmelen, F., Lifschitz, V., Porter, B. (eds.): Handbook of Knowledge Representation.
Elsevier Science, Amsterdam (2007)
15. Buchanan, B.G., Shortliffe, E.H. (eds.): Rule-Based Expert Systems. Addison-Wesley
Publishing Company, Reading (1985)
16. Scott, A.C.: A Practical Guide to Knowledge Acquisition, 1st edn. Addison-Wesley Longman
Publishing Co., Inc, Boston (1991)
17. Vermesan, A.I., Coenen, F. (eds.): Validation and Verification of Knowledge Based Systems.
Theory, Tools and Practice. Kluwer Academic Publisher, Boston (1999)
Chapter 1
Rules as a Knowledge Representation
Paradigm

Rules are a commonly used and natural way to express knowledge. They have been
used for decades in AI, Computer Science, Cognitive Science and other domains.
We start by discussing the AI roots of rules in Sect. 1.1 where we elaborate on dif-
ferent kinds and types of rules. We then focus on a more careful treatment of rules in
symbolic AI. There, they constitute an approach which allows for the representation
of knowledge and basic automated reasoning, see Sect. 1.2. Originally, one of the
most important areas for rule applications were expert systems. While this perspec-
tive is much broader, we will briefly discuss it in Sect. 1.3. What makes rule-based
representation and reasoning particularly interesting is the opportunity for the for-
malization of rule languages. Selected logic-based formalizations are considered in
Sect. 1.4. Then in Sect. 1.5 we present in more detail a family of so-called Attributive
Logics which play an important role in the remaining part of the book. Based on
these concepts we introduce important requirements for a formalized description of
rule-based systems in Sect. 1.6. We summarize the chapter in Sect. 1.7.

1.1 Rules in AI

Artificial Intelligence is often considered to be a field that aims to study and design
Intelligent Systems (IS) [1, 2]. Since the 1970s, a number of successful paradigms
for the engineering of such systems have been developed. From the application point
of view, some of the most important are those that have focused on decision support.
Expert Systems (ES) [3, 4], especially those that are rule-based proved to be the most
successful. This is mainly due to the fact that rules allow for a powerful and declarative
specification of knowledge that is easy to understand and use. The application areas
of rules have been expanding. In last decades, they have been used in the area of
business where they allow us to define and constrain aspects of business operations;
© Springer International Publishing AG 2018 3
G.J. Nalepa, Modeling with Rules Using Semantic Knowledge Engineering,
Intelligent Systems Reference Library 130,
https://doi.org/10.1007/978-3-319-66655-6_1
Another random document with
no related content on Scribd:
One of the delights of our wanderings is to stop at a strange post
office, and have a whole handful of letters respond to our call.
Chester responded very generously, for here the truant letters, which
were each time a little behind, and had been forwarded and
reforwarded, met the ever prompt ones and waited our arrival. A few
miles from Chester we found lovely maidenhair ferns by the
roadside, and were gathering and pressing them, when an old man,
in a long farm wagon, stopped and asked if we were picking
raspberries. We told him it was rather late for raspberries, but we
had found pretty ferns. To our surprise this interested him, and he
talked enthusiastically of ferns and flowers, saying he had one
hundred varieties in his garden, and asking if we ever saw a certain
agricultural journal which was a treasure-house of knowledge to him.
Still he was not a florist, but a vegetable gardener, and we learned
ever so much about the business, and for a while could talk glibly of
Angel of Midnight corn and Blue-eyed (?) pease and so on. He gave
quite a discourse, too, on the advantages of co-operation and
exchange of ideas. He told us how much he enjoyed a fair at the
New England Institute Building, and was interested to know that we
saw it when in flames. Our pleasant chat was brought to a sudden
stop, just as he was telling us of his ambitious daughter and other
family details, by other travelers, for whom we had to clear the road.
We spent a night pleasantly at Saxton’s River, and received the
courtesies of friends, then on through Bellows Falls and Keene
towards Monadnock. We wanted to go to the Mountain House for the
night, but it was several miles out of our way, and we were tired as
well as Charlie, with thirty miles’ driving in the heat, so contented
ourselves with recollections of two delightful visits there, and stopped
at Marlboro, five miles from Keene.
When we were packing up in the phaeton, the next morning, a lady
brought us three little bouquets, the third and largest for Charlie, we
fancy. It was a very pleasant attention to receive when among
strangers and gave us a good send-off for our last day’s drive. Forty
miles is a long drive at the end of a long journey, but Charlie seemed
fully equal to it, and all went well as we journeyed along the familiar
route through Troy, Fitzwilliam, Winchendon, Ashburnham and
Fitchburg. We dined at Winchendon and visited the friends in
Fitchburg from whom we have a standing invitation for our last tea
out. The five miles from Fitchburg to Leominster Charlie never
counts. He knows his own stall awaits him. Our last day, which
began so pleasantly with a floral testimony from a stranger, ended
with a night-blooming cereus reception in our own home.
“Did you take Summer Gleanings,” do I hear some friend ask? Oh
yes, we took it, but not one sketch did we add to it. The fever for
sketching ran high last year and spent itself, but every day of the
July pages is radiant with pressed flowers and ferns. One more trip
and the book will be full, “a thing of beauty,” which will be “a joy
forever.”
 CHAPTER VIII.
NARRAGANSETT PIER AND MANOMET POINT.
“Think on thy friends when thou haply seest
Some rare, noteworthy object in thy travels;
Wish them partakers of thy happiness.”
We thought of omitting our annual letter to the Transcript, believing
that vacations in everything are good; but, even before the journey
existed, except in mind, a report of it was assumed as a matter of
course, as the part belonging to our friends, who have not found
opportunity to travel in our gypsy fashion. Then, too, we remembered
the lines above, quoted by Andrew Carnegie, as we journeyed with
him in his “Four in Hand through Britain,” and still more delightful
“Round the World,” all in a hammock in those scorching July days,
without a touch of fatigue or sea-sickness. Even a carriage journey
on paper has some advantages, no dust, no discomfort of any kind;
but we prefer the real thing, and enjoyed it so much we will change
our mind and tell you a little about it. The places are all so familiar,
and so near the “Hub” of the universe, that when you get to the end
you may feel, as we did, as if you had not been anywhere after all.
We did, however, drive four hundred miles, and had a very delightful
time.
Before we really start, we must introduce to you the new member of
our party. With deep regret and many tender memories we tell you
we parted with our Charlie last spring, and a big, strong Jerry came
to take his place. A friend in cultured Boston said, “Why, how will
Jerry look in the Transcript?”
We did not go until September, and, like every one else, you may
wonder why we waited so late, when we have often started as soon
as the “crackers” were fired off. Well, Jerry had not become used to
our climate, although July was hot enough for any Southerner. Then
the company season came, and various things made it advisable to
wait until September. We were quite reconciled, because you know
all those “conjunctions” of the planets were to culminate in August,
and it seemed likely the world was to be turned upside down. We
thought it would be so much pleasanter to be swallowed up by the
same earthquake, or blown away by the same cyclone as our home
friends.
Jerry waxed in strength, the world still stood, the last summer guest
had departed, and on the afternoon of Sept. 8, we started for Stow.
“What on earth are you going there for?” and similar comments
reveal the impressions of our friends; but we knew why, and do not
mind telling you. We were going to Boston to begin our journey, and
we could not go beyond Stow that afternoon, without going farther
than we liked to drive Jerry the first day, for he is young and we were
determined to be very considerate of him. We knew we should be
comfortable at the little, weather-beaten hotel, and that Jerry would
have the best of care.
How lovely that afternoon drive! It was the day after those terrific
storms and gales, the final “conjunction,” probably, and there was an
untold charm in everything. As we drove leisurely along, gathering
flowers to press for “Summer Gleanings,” we thought of our friends
who were speeding their way back to New York just at the time when
the country is loveliest, and knew they were envying us. Still,
somehow it did not seem as if we were traveling, but only going to
drive as we had been doing all summer. Perhaps we missed the July
heat and dust!
“Still as Sunday” gives no idea of the quiet of Stow. It seemed as if
one might live forever there, and perhaps one could, if permitted, for
just as we were leaving the hotel for a little stroll, our landlady was
saying to some “patent medicine man,” “We don’t have any
rheumatism here, nobody ever dies, but when they get old they are
shot.”
We had not walked far before we came to a cemetery, and,
remembering the landlady’s remark, we went in to read the
inscriptions. No allusion was made to shooting, but if it was a familiar
custom the omission is not strange. We noted a few epitaphs which
interested us:
“When I pass by, with grief I see
My loving mate was taken from me.
Taken by him who hath a right
To call for me when he sees fit.”

“A wife so true there are but few,

And difficult to find,
A wife more just and true to trust,
There is not left behind.”

“A while these frail machines endure,

The fabric of a day,
Then know their vital powers no more,
But moulder back to clay.”
“Friends and physicians could not save

My mortal body from the grave.”

There were six stones in close proximity bearing these familiar lines
—
“Stop, traveler, as you pass by,
As you are now, so once was I.
As I am now, so you must be.
Prepare for death and follow me.”
All that night was lost, for we never woke once. Was it the stillness?
or was it that cosy, bright room, with its very simple, but effective,
“homey” touches? Be that as it may, we were fresh as the morning,
and ready to enjoy every mile of the drive to Boston, gladdening our
hearts with the sight of friends as we tarried now and then. We in
Boston and our Boston friends in the country was something new,
but a room at the B. Y. W. C. A. is next to home, and we heartily
recommend it to homeless ladies traveling as we were, or on
shopping expeditions. The night, with the unceasing din of the horse
cars, and the thousand and one noises peculiar to the city, was a
marvelous contrast to Stow, but in time we became adjusted to our
environments, and were lost in sleep.
How delightful to be in Boston, and know that there were only two
things in the whole city we wanted—a Buddhist catechism and a
horn hairpin. These procured, we went for Jerry and began the day,
which was to be devoted to making calls. We went spinning along
over the smoothly paved Columbus avenue on our way to the
Highlands, and rattled back on cobble-paved Shawmut avenue.
Dinner over, off we started for Allston, Somerville and Cambridge,
and as it was not yet five o’clock when we came back over the Mill-
dam, we could not resist turning off West Chester Park, and hunting
up some friends in Dorchester, returning in early evening. Jerry
seemed perfectly at home; perhaps he has been used to city life in
Kentucky. The day was long and full of pleasant things, but the diary
record was brief; for just this once we will confess we were tired.
Secured the catechism and hairpin, and oh! we forgot, a bit of
embroidery we got at Whitney’s, and mailed to a friend who asked us
to do so if we “happened to be near there,” drove eighteen miles and
made twelve calls, that was all.
During the day we decided to stay over Sunday, as a cousin we
wanted to see was coming. Jerry rested all day, and we did, except
the writing of many letters, dining with a friend, and attending service
at the only church we saw lighted on the Back Bay in the evening.
We thought of many things to do and places to go to, and wondered
how we should like to take a carriage journey and spend all the
nights in Boston. There would be no lack of pleasant driving, and if
we missed the variety in hotels, we could easily remedy that by
going from one to another. Boston would supply that need for a
while, and we are sure Jerry would be more than glad to find himself
at Nims’s in Mason street, day or night. But we had other things in
view for this journey, and, the cousin’s whereabouts being wrapped
in mystery, we left Boston early Monday morning.
Now, we will take you by transit, hardly excelled in rapidity by the
feats of occultism, to Narragansett Pier, and while you are taking
breath in our charming room in that vine-covered hotel at the
jumping-off place, with the surf rolling up almost under the windows,
we will just tell you a bit about the journey as we had it; driving all
day in the rain on Monday and enjoying it, making hasty doorstep
calls, spending the night at Lake Massapoag House in Sharon, and
on through the Attleboros to Pawtucket the next day, dining
Wednesday with friends in Providence, then on to East Greenwich
for the night. A drive of twenty-one miles Thursday morning, and we
are with you again at the Pier, where our first exclamation was, “Oh!
let’s stay here!” We like the mountains, but the ocean is quite
satisfying if we can have enough of it, and as our host said, here
there is nothing between us and Europe, Asia and Africa. We wrote
letters all the afternoon, with one eye on the surf, and the next
morning we drove to Point Judith, where we investigated the wrecks,
went to the top of the light-house, and were much interested in
hearing all about the work at the life-saving station. We took a long
walk, and visited the Casino in the afternoon.
We were still enthusiastic about the Pier, but the next morning was
so beautiful it seemed wise to enjoy it in Newport. The captain could
not take our horse across from the Pier, and we drove twelve miles
back to Wickford to take the ferryboat. It was quite cool, but with
warm wraps it was just right for a brisk drive. We had time for dinner
before going to the boat. The hour’s sail was very delightful, and at
half after two we were in Newport, with nothing to do but drive about
the city until dark. We saw all there was to be seen, even to the
hydrangea star described in the Transcript by “M. H.” We did not
know which was Vanderbilt’s and which Oak Glen, but that mattered
little to us, for to all intents and purposes they all belonged to us that
bright afternoon, and are still ours in memory. We fell into the grand
procession of fine turnouts on the prescribed ocean drive, but the
people generally did not look as if they were having a good time.
They had a sort of “prescribed” look, except one young lady we met
several times, perched in a high cart, with a bright-looking pug for
company; she really looked as if she was enjoying herself.
The charm of Newport fled when we were inside the hotel. The
fountain in the park below our window was very pretty, but it could
not compete with our ocean view at the Pier, and we had to sit on the
footboard of the bed, too, in order to see to read by the aspiring
gaslight.
We walked around the Old Mill and went into the Channing Church
and then left Newport for Fall River. There we called on several
friends, then inquired for some place to spend a night, on our way to
Plymouth, and were directed to Assonet. We had never heard of
Assonet before, but we did not mind our ignorance when the widow,
who “puts up” people, told us the school committee man where her
daughter had gone to teach had never heard of it. Our good woman
thought at first she could not take us, as she had been washing and
was tired, but as there was no other place for us to go, she
consented. When she saw our books, she asked if we were traveling
for business or pleasure, and as F—— drove off to the stable she
remarked on her ability; she thought a woman was pretty smart if
she could “turn round.” We had a very cosy time. People who always
plan to have a first-class hotel lose many of the novel experiences
which make a pleasant variety in a journey It is interesting
occasionally to hear the family particulars and be introduced to the
pet dogs and cats, and walk round the kitchen and backyard, where
the sunflowers and hollyhocks grow from old-time habit, and not
because of a fashion.
The Samoset House at Plymouth seemed all the more luxurious
after the modest comforts at Assonet. We “did” Plymouth once more,
this time taking in the new monument, and having plenty of time, we
drove down to Manomet Point for a night. The Point is quite a resort
for artists, but as we have given up sketching, we did not delay
there, but returned to Plymouth and on to Duxbury. We did not ask
Jerry to travel the extra miles off the main route to take in Brant Rock
and Daniel Webster’s old home, as this was our second drive in this
vicinity, and rather than drive two miles to a hotel possibly open, we
took up with the chances near by. We found oats at a grocery store,
but it was too cold to camp; indeed, we did not have one of our
wayside camps during the entire journey. There was no hotel, no
stable, no “put-up” place or available barn, but the grocer,
appreciating our dilemma, said he could easily clear a stall back of
his store, and while he was helping us unharness, we saw a large
house perched on a high bluff not far away. Although it was a private
boarding-house we made bold to cross the fields, mount the many
flights of steps and ask for dinner, which was willingly granted.
You will surmise we are bound for Boston again, and will not be
surprised to find us with friends on the Jerusalem Road, after
enjoying the beauties of this road from Cohasset to Hingham, where
we went for a handful of letters only equalled by that parcel at
Providence.
Oh, how cold it was the next day! The thought of Nantasket Beach
made us shiver, and preferring to think of it as in “other days,” we
turned our faces inland and drove a pretty back way to South
Hingham. Of course we could have driven right into Boston, but it
was Saturday, and we thought we would have a quiet Sunday
somewhere and go into the city Monday. After protracted
consultation we agreed on a place, but when we got there there was
no room for us, as a minstrel troupe had taken possession. Hotels
four, eight and nine miles distant were suggested. In consideration of
Jerry we chose the four miles’ drive. We will not tell you the name of
the town, suffice it to say we left immediately after breakfast. It was a
beautiful morning—far too lovely to be spoiled by uncongenial
surroundings. We intended to drive to the next town, where we had
been told there was a hotel. We found none, however, but were
assured there was one in the next. So we went on, like one in pursuit
of the end of the rainbow, until the last man said he thought there
was no hotel nearer than the Norfolk House!
Here we were almost in Boston, Sunday, after all the miles we had
driven to avoid it. “All’s well that ends well,” however, and a little visit
with the “Shaybacks” at home, not “in camp,” could not have been on
Monday, and before we reached the Norfolk House we were taken
possession of for the night by a whole household of hospitable
friends.
Monday morning we drove into the city proper, and hovered in its
vicinity several days, calling on friends we did not see before and
driving here and there, among other places to Middlesex Fells, so
often spoken of. We ended our journey as we began it, searching for
our clerical cousin, but all in vain. We did see so many of our friends
of the profession, however, from first to last, that privately we call it
our “ministerial” journey.
Everything must have an end, but we did wish we could go right on
for another month. The foliage was gorgeous and the yellowish haze
only made everything more dreamy and fascinating. We prolonged
our pleasure by taking two days to drive home, straying a little from
the old turnpike, and driving through Weston, spending the night in
Framingham, and then on through Southboro to Northboro, Clinton
and Lancaster to Leominster. The country was beautiful in contrast
with flat, sandy Rhode Island. We gathered leaves and sumacs until
our writing tablet and every available book and newspaper was
packed, and then we put a great mass of sumacs in the “boot.”
Finally our enthusiasm over the beauties along the way reached
such a height that we spread our map and traced out a glorious trip
among the New Hampshire hills, and home over the Green
Mountains, for next year.
“Summer Gleanings” is now complete, and the last pages are fairly
aglow with the autumn souvenirs of our sixteenth annual drive.
 CHAPTER IX.
BOSTON, WHITE MOUNTAINS AND VERMONT.—A SIX
HUNDRED MILE DRIVE.
In self-defence we must tell you something of our seventeenth
annual “drive,” for no one will believe we could have had a good
time, “on account of the weather;” and really it was one of our finest
trips. We regret the sympathy, and pity even, that was wasted on us,
and rejoice that now and then one declared, “Well, I will not worry
about them, for somehow they always do have a good time, if it does
rain.”
If two friends, with a comfortable phaeton and a good horse,
exploring the country at will, gladly welcomed and served at hotels
hungry for guests, with not a care beyond writing to one’s friends,
and free to read to one’s heart’s content, cannot have a good time,
whatever the weather may be, what hope is there for them?
Why has no one ever written up the bright side of dull weather? The
sun gets all the glory, and yet the moment he sends down his
longed-for smiles, even after days of rain, over go the people to the
other side of the car, the brakeman rushes to draw your shutter, the
blinds in the parlor are closed, and the winking, blinking travelers on
the highway sigh, “Oh, dear, that sun is blinding,” and look eagerly
for a cloud. Then, if the sun does shine many days without rain, just
think of the discomfort and the perpetual fretting. Clouds of dust
choke you, everything looks dry and worthless, the little brooks are
moping along, or there is only a dry stony path that tells they once
lived, and the roadsides look like dusty millers. Now, fancy a drive
without the sunshine to blind your eyes, no dust (surely not, when
the mud fairly clogs the wheels), every tree and shrub glistening and
all the little mountain streams awakened to life and tearing along,
crossing and recrossing your path like playful children; indeed, all
Nature’s face looking like that of a beautiful child just washed. Really,
there is no comparison.
Perhaps you are thinking that is a dull day drive. Now, how about a
drive when it pours. Oh, that is lovely—so cosy! A waterproof and
veil protect you, and the boot covers up all the bags and traps, and
there is a real fascination in splashing recklessly through the mud,
knowing you have only to say the word and you will come out spick
and span in the morning.
We have purposely put all the weather in one spot, like “Lord”
Timothy Dexter’s punctuation marks, and now you can sprinkle it in
according to your recollection of the September days, and go on with
us, ignoring the rain, as we did, excepting casual comments.
Our journey was the fulfilment of the longing we felt for the
mountains, when we were driving home from our Narragansett Pier
and Newport trip one year ago. Perhaps you remember those hazy,
soft-tinted days, the very last of September. The air was like
summer, as we drove along through Framingham to Southboro,
gathering those gorgeous sumacs by the wayside, and wishing we
could go straight north for two weeks.
The morning of Sept. 6th, 1888, was very bright, just the morning to
start “straight north,” but with our usual aversion to direct routes we
turned our faces towards Boston. We could not stop at Stow this
time, for the old hotel, where we slept so sweetly our first night one
year ago, is gone, and only ashes mark the spot. Waltham had a
place for us, however. A cold wave came on during the night, and we
shivered all the way from Waltham to Hull, except when we were
near the warm hearts of our friends on the way.
The ocean looked cold, but nothing could mar that quiet drive of five
miles on Nantasket Beach just before sunset. We were lifted far
above physical conditions. We were just in season to join in the last
supper at The Pemberton, and share in the closing up. We were
about the last of the lingering guests to take leave in the morning,
after dreaming of driving through snowdrifts ten feet deep, and
wondering if we should enjoy the mountains as well as we had
fancied. The weather, however, changed greatly before noon, and it
was very sultry by the time we reached Boston. Prudence prompted
us, nevertheless, to add to our outfit, against another cold wave. We
found all we wanted except wristers. Asking for them that sultry
afternoon produced such an effect that we casually remarked, to
prove our sanity, that we did not wish them to wear that day.
Night found us at Lexington, pleading for shelter at the
Massachusetts House. Darkness, rain and importunity touched the
heart of the proprietor, and he took us into the great hall, which
serves for parlor as well, saying all the time he did not know what he
should do with us. We wanted to stay there, because we do not often
have a chance to stay in a house that has traveled. The signs are
over the doors just as when it stood on the Centennial grounds, and
many things seem quite natural, although we did not chance to be
among the distinguished guests entertained under its roof when in
Philadelphia.
Our stay there was made very pleasant by a lady who gave us
interesting accounts of her journeys by carriage with “Gail Hamilton”
and her sister.
Here ended our one hundred miles preliminary, and bright and early
Monday morning we were off for the mountains. The day was just
right for a wayside camp, and just at the right time we came to a
pretty pine grove, with seats under the trees. We asked a bright
young woman in the yard opposite if we could camp there, and were
given full liberty. She said Jerry might as well be put into the barn,
then helped unharness and gave him some hay. Jerry was happy.
He does not have hay—which is his “soup,” I suppose—when he
camps. We went to the grove with our little pail filled with delicious
milk, and a comfortable seat supplied by our hospitable hostess.
When we went to pay our bill, everything was refused but our thanks.
We said then, “If you ever come to Leominster you must let us do
something for you.”
“Oh, do you live in Leominster? Do you know ——?”
“Oh, yes, she is in our Sunday-school class.”
This is only one of the many pleasant incidents of our wanderings.
We spent that night at Haverhill and had one more camp, our last for
the trip, this time on the warm side of a deserted barn.
Two and a half days’ driving up hill and down to Dover, and over a
good road through Rochester and Farmington, brought us to Alton
Bay, where we all went on board the Mt. Washington for the sail of
thirty miles to Centre Harbor. Jerry was tied in the bow, and as we
got under way the wind was so strong we should have had to wrap
him up in our shawls and waterproofs if the captain had not invited
him inside. We braved it on deck, for Lake Winnipiseogee is too
pretty to lose.
We “did” Centre Harbor some years ago, so drove on directly we
landed. At Moultonboro we stopped to make some inquiries, and
while waiting, the clouds grew very mysterious, looking as if a
cyclone or something was at hand, and we decided to spend the
night there. The people were looking anxiously at the angry sky; and
the Cleveland flag was hastily taken down; but no sooner were we
and the flag under cover than the sun came out bright, dispelling the
blackness. We wished we had gone on as we intended, and looked
enviously on the Harrison flag, which waved triumphantly, not afraid
of a little cloud.
We saw a large trunk by the roadside as we drove through the
woods next morning. We gave all sorts of explanations for a good-
looking trunk being left in such an out-of-the-way place, but, not
being “reporters,” we did not “investigate” or “interview,” but
dismissed the matter with, “Why, probably it was left there for the
stage.” We do not feel quite satisfied yet, for why any one should
carry a trunk half a mile to take a stage when we had no reason to
think there was any stage to take, is still a mystery.
We got all over our disappointment at stopping early for the cloud,
for the drive, which was so lovely that bright morning, would have
been cold and cheerless the night before. It seemed as if we went on
all sides of Chocorua, with its white peak and pretty lake at the base.
Why has somebody said—
“Tired Chocorua, looking down wistfully into
 A land in which it seemed always afternoon.”

One might spend a whole summer amid the charming surroundings

of North Conway, but we had only a night to spare. There were many
transient people about, as is usual in the autumn. The summer
guests had departed, and now some of the stayers-at-home were
having a respite. We wished all the tired people could try the
experience of an old lady there, who said she “could not make it
seem right to be just going to her meals and doing nothing about it.”
Oh, how lovely that morning at North Conway! This was the day we
were to drive up Crawford Notch; and what about all the prophecies
of our seashore friends? Where were the snowdrifts we dreamed of?
The air was so soft we put aside all wraps, and, as we leisurely
drove along the bright, woodsy road, I wonder how many times we
exclaimed, “This is heavenly!” We fairly drank in the sunshine, and
fortunately, for it was the last we had for a full week.
We dined at the hotel in Bartlett, and strolled about the railway
station near by, so tempting to travelers, having a pretty waiting-
room like a summer parlor, with its straw matting and wicker
furniture. We took our time so leisurely that we found we could not
get to the Crawford House in season to walk up Mt. Willard, as we
had planned, so stopped at the old Willey House, this side. It was
quite too lovely to stay indoors, and, after we had taken possession
of the house, being the only guests, we took the horn our landlady
used to call the man to take care of Jerry, and went down the road to
try the echo, as she directed us. It was very distinct, and after we got
used to making such a big noise in the presence of those majestic
mountains, we rather liked it. We gathered a few tiny ferns for our
diaries, and took quite a walk towards the Notch, then came “home,”
for so it seemed. We had chosen a corner room in full view of Mt.
Webster, Willey Mountain, and the road over which we had driven,
and where the moon would shine in at night, and the sun ought to
look in upon us in the morning. The moon was faithful, but the sun
forgot us and the mountains were veiled in mists.
Will there ever be another Sunday so long, and that we could wish
many times longer? We had the warm parlor to ourselves and just
reveled in a feast of reading, watching the fluffy bits of mist playing
about Mt. Webster, between the lines. Just fancy reading “Robert
Elsmere” four hours on a stretch, without fatigue, so peaceful was it
away from the world among the mountains. After dinner we drove to
the Crawford to mail a letter and back to the Willey, having enjoyed
once more in the short one hour and a half one of the grandest
points of the whole mountain region, the White Mountain Notch. We
were now fresh for another long session with Robert and Catherine.
It was raining again, and steadily increased through the night until it
seemed as if there would not be a bridge left of the many we had
crossed the day before.
We were interested in the fate of the little bridges, for we were to
retrace our steps, seventeen or eighteen miles, to Glen station. We
had driven up through the Notch because—we wanted to; and we
were going back all this distance because we wanted to go on the
Glen side of the mountains; for with all our driving, we had never
been there. What a change from the drive up on Saturday! How
lively the streams; and the little cascades were almost endless in
number.
The foliage looked brighter, too. The roads were washed, but the
bridges all stood. We dined once more at Bartlett, then on to
Jackson via Glen Station. We had not thought of Jackson as so
cosily tucked in among the mountains.
Again we were the only guests at the hotel, and the stillness here
was so overpowering, that it required more courage to speak above
a whisper in the great empty dining-room than it did to “toot” the horn
in Willey Notch.
We usually order our horse at nine, but when it pours, as it did at
Jackson, we frequently dine early and take the whole drive in the
afternoon. These rainy stop-overs are among the pleasant features
of our journeys. Who cannot appreciate a long morning to read or
write, with conscience clear, however busy people may be about
you, having literally “nothing else to do”? It does not seem to trouble
us as it did the old lady at North Conway. It was cool in our room,
and we took our books down stairs, casually remarking to the clerk,
who apparently had nothing to do but wait upon us, that we had
been looking for the cheery open fire we saw in the reception room
the evening before. He took our modest hint, and very soon came to
the parlor, saying we would find it more comfortable in the other
room, where there was a fire.
Early in the afternoon we were off, full of anticipation of a new drive,
and by many the drive from Jackson to Gorham through Pinkham
Notch and by the Glen House is considered the finest of all. The
foliage was certainly the brightest and the mud the deepest of the
whole trip, and we enjoyed every inch of the twenty miles. We fully
absorbed all the beauty of the misty phases of the mountains, and
did not reject anything, thinking instead how we would some time
reverse things and drive from Gorham to Jackson on a pleasant day.
Another famed drive is the one from Gorham to Jefferson. Part of
this was new to us, too, and we must confess that the “misty phases”
were too much for our pleasure that time. Not a glimpse of the peaks
of the Presidential range was to be had all that morning. Even the
Randolph Hills were partly shrouded in mists. We dined at
Crawford’s at Jefferson Highlands, and one of the guests said Mr.
Crawford had promised a clear sunset, but what his promise was
based on we could not imagine.
It does not seem as if anything could entirely spoil the drive from the
Highlands to the Waumbek at Jefferson, and from Jefferson to
Lancaster the views are wonderfully beautiful. The clouds relented a
little as we slowly climbed the hills, and just as we reached the
highest point we turned back once more for a last look at the entire
White Mountain range, and we had a glimpse of the peak of Mt.
Washington for the first time since the morning we left North
Conway.
A moment more, and the Summit House glistened in sunlight, a stray
ray from behind a cloud. As we began to descend, what a change of
scene! Sun-glinted Washington was out of sight behind the hill, and
before us were threatening clouds, black as midnight, and the
mountains of northern New Hampshire looked almost purple. The
sky foreboded a tempest rather than Mr. Crawford’s promised
sunset, but while we were thinking of it there was a marvelous
change. Color mingled with the blackness, and as we were going
down the last steep hill into Lancaster, there was one of the most
gorgeous sunset views we ever witnessed. We drove slowly through
the broad, level streets to the outer limit of the town, and then turned
back, but did not go to the hotel until his majesty dropped in full glory
below the horizon.
The sun set that night for the rest of the week, and the clouds were
on hand again in the morning. We went to Lancaster just for a look
towards Dixville, but we made this our turning-point. The drive to
Whitefield is very like the one just described, only reversed. There
were no sun-glints this time, but memory could furnish all the clouds
refused to reveal, for that ride was indelibly photographed on our
minds.
From Whitefield we drove to Franconia, and as we went through
Bethlehem street we thought it seemed pleasanter than ever before.
The gray shades were becoming, somehow.
Having driven through Franconia Notch five times and seen the
“boulder” before and after its fall, we did not fret about what the
weather might be this time. We had been through in rain and
sunshine, in perfect, gray, and yellow days, and never failed to find it
charming. This time it poured in torrents. We dined at the Flume
House, and watched those who were “doing” the Notch for the first
time, and almost envied them as they gayly donned their waterproofs
and were off for the Pool and Flume. One party declared they had
laughed more than if it had been pleasant, and all in spite of that
ruined Derby, too, which the gentleman of the party said he had just
got new in Boston, and intended to wear all winter. They had passed
us in the Notch in an open wagon, with the rain pelting their heads.
The drive to Campton that afternoon was one of those “cosy” drives.
It never rained faster, and the roads were like rivers. Memory was
busy, for it is one of the loveliest drives in the mountains. It was dark
when we reached Sanborn’s, at West Campton, but it is always
cheery there, and the house looked as lively as in summer.
One might think we had had enough of mountains and mists by this
time, but we were not yet satisfied, and having plenty of time, we
turned north again, just before reaching Plymouth, with Moosilauke
and the Green Mountains in mind. A happy thought prompted us to
ask for dinner at Daisy Cottage in Quincy, and unexpectedly we met
there one of the party who braved Franconia Notch in winter a few
years ago, and who told the tale of their joys and sorrows in the
Transcript. We mailed our cards to the friends whose house was
closed, and then on to Warren, near Moosilauke. We experienced
just a shade of depression here, perhaps because the hotel, which
had been full of guests all summer, was now empty and cold, or
possibly the sunshine we absorbed at North Conway—“canned”
sunshine, Mr. Shayback calls it—was giving out. Be that as it may,
our enthusiasm was not up to the point of climbing a mountain to see
what we had seen for eight successive days,—peaks shrouded in
white clouds. The sun did shine in the early morning; but it takes
time to clear the mountains, and the wind blew such a gale we
actually feared we might be blown off the “ridge” on Moosilauke if we
did go up. We waited and watched the weather, finished “Robert
Elsmere,” and began for a second reading, and after dinner gave up
the ascent. By night we were reconciled, for we had the most
charming drive of twenty miles to Bradford, Vt., crossing the
Connecticut at Haverhill, and saying good-by to New Hampshire and
its misty mountains.
A new kind of weather was on hand next morning, strangely like that
we have become accustomed to, but not so hopeless.
These dense fogs along the Connecticut in September are the
salvation of vegetation from frosts, we were told, but they are fatal to
views. We drove above and away from the fog, however, on our way
over the hills to West Fairlee, but it rested in the valley until nearly
noon. It was encouraging to learn that fair weather always followed.
A “bridge up” sent us a little way round, but we reached West Fairlee
just at dinner time, and while Jerry was at the blacksmith’s we
strolled about the village with friends. The afternoon drive to Norwich
on the Connecticut—a pretty, old university town—was very
pleasant. We were directed to the hotel, but when a lady answered
the door bell, we thought we must have made a mistake, and were
asking hospitality at a private mansion. There was no sign; the yard
was full of flowers, and the big square parlor, with the fire crackling
under the high old mantel, the fan-decorated music-room through the
portieres—everything, in fact, betokened a home. And such in truth it
was, only, having been a hotel, transients were still accommodated
there, as there was no other place in Norwich. When the very gallant
colored boy ushered us into a room the size of the parlor below, with
all the homey touches, we felt really like company. The delicious
supper, well served from the daintiest of dishes, confirmed the
company feeling.
We started out in the densest of fogs from our luxurious quarters in
Norwich, but soon left it behind, and the drive along White River was
very lovely. We had to dine at a “putting-up” place, with another
fellow-traveler, in a kitchen alive with flies; and at Bridgewater, where
we went for the night, we were received by a woman with mop and
pail in hand—a little “come down” after our fine appointments. We
must not forget our pleasant hour in Woodstock that afternoon. We
drove through its pretty streets, called on friends, and took a look at
the fair grounds, for everybody was “going to the fair.”
Fine appointments are not essential to comfort, and when we were
all fixed in our little room, with a good book, waiting once more for it
to simply rain, not pour, we were just as happy as at Norwich. After
dinner we challenged the weather, and set forth for Ludlow. We
overtook the little Italian pedler, with what looked like a feather bed
on his back, who had sat at table with us, and was now ploughing his
way through the mud. His face was wreathed in the most
extravagant smiles in response to our greeting. The rain had spent
itself, and we enjoyed walking down the mountain as we went
through Plymouth. It seemed an unusual mountain, for there was no
“up” to it, but the “down” was decidedly perceptible.
Ludlow was as homelike as ever, and the Notch drive on the way to
Chester as interesting. The foliage, usually so brilliant at that season,