Handbook of Software Engineering and Knowledge Engineering

Preface

The Handbook of Software Engineering and Knowledge Engineering is the first 
comprehensive handbook covering these two important areas that have become
interwoven in recent years. Many international experts contribute to this Handbook. 
Each article is written in a way that a practitioner of software engineering and
knowledge engineering can easily understand and obtain useful information. Each 
article covers one topic and can be read independent of other articles, providing
both a general survey of the topic and an in-depth exposition of the state-of-the-art. 
Practitioners will find this Handbook useful when looking for solutions to practical
problems in software engineering and knowledge engineering. Researchers in turn
can use the Handbook to quickly obtain background information, current trends and
the most important references on a certain topic. 

The Handbook consists of two volumes. Volume One covers the basic principles
and applications of software engineering and knowledge engineering.  Volume
Two expands the coverage of basic principles and also contains many articles
that specifically addresses visual and multimedia software engineering, and
emerging topics in software engineering and knowledge engineering such as
software patterns, data mining for software knowledge, etc.  The two volumes
form a complete set, but can be used separately for different purposes.


Turning Knowledge into Software

There is a growing awareness that the central issue in software engineering
and knowledge engineering is how to turn knowledge into software.  Traditionally
software engineering is concerned with the specification, design, coding, testing
and maintenance of software.  It also implicitly deals with the issues of
transforming knowledge into software in the sense that the gathering of
knowledge about the problem domain is incorporated into the requirements
analysis phase of the software life cycle.  Often, informal techniques of
knowledge acquisition are used. Thus in the past, the role of knowledge
engineering in the software process is an implicit one.

However it has long been recognized by many people that knowledge
engineering plays an increasingly important role in software design. Indeed
it is because of this conviction that the international conference series on
Software Engineering and Knowledge Engineering (SEKE) was founded in
1988, followed by the publication of the International Journal of Software
Engineering and Knowledge Engineering (IJSEKE) three years later. For
both the SEKE conference series and the IJSEKE journal, the basic
viewpoint is that the interdisciplinary area of software engineering and
knowledge engineering is concerned with the interplay between software
engineering and knowledge engineering - how software engineering can
be applied to knowledge engineering, and how knowledge engineering
can be applied to software engineering.

This viewpoint should now be modified and expanded because, both in
theory and in practice, more and more software engineers and knowledge
engineers are explicitly incorporating knowledge into the software process.
In editing this two-volume handbook, this expanded viewpoint - that software
engineering is concerned with the transformation of knowledge into software -
has been carefully taken into consideration to conceptually organize the
recent progresses in software engineering and knowledge engineering.


Software Patterns

Let us start with two distinct, yet complementary, viewpoints on software
engineering. The two viewpoints may seem completely different, but they
are but different ways of viewing the "elephant" that is software engineering.

The first viewpoint, as stated above, is that software engineering is
concerned with the transformation of knowledge into software. The
second viewpoint is somewhat more technical.  It says that software
engineering is concerned with the specification, design, transformation,
verification and validation of patterns.

Software is nothing but patterns.  A program is constructed from some
basic patterns, and the construction rules can in turn be expressed as
other types of patterns.  With grammars, formal languages and automata,
there are many approaches to describe the basic patterns and how
they are composed into programs.

Specifications are composed of patterns that are the basic building
blocks of formal, informal or visual specifications.  The specification,
in the ideal case, can then be automatically transformed into programs,
and verified and validated in the transformational process.

As mentioned above, knowledge used to be described informally,
but now there are formal techniques and more precise ways of
dealing with knowledge.  With advances in object oriented methods,
one comes to the inevitable conclusion that knowledge is also
composed of patterns. Knowledge is first acquired, then transformed
into formal/informal/visual specification, design and finally program.

Therefore, software engineering can now be viewed as the transformation
of knowledge into software through the transformation of patterns.  The
central issue of software engineering is how to turn knowledge into
software by means of the creation, composition and transformation
of various types of patterns.  A key question that can be asked repeatedly
for any topic or sub-topic is the following: how to turn what-kind-of knowledge
patterns into what-kind-of software patterns?


Overview of Volume One

As mentioned above, the Handbook of Software Engineering and Knowledge
Engineering is a comprehensive handbook providing the reader with both
useful overviews and detailed explanations of the methodologies, techniques
and current research issues in software engineering and knowledge engineering.
Volume One deals with the fundamentals of software engineering and knowledge
engineering.  Topics relevant to the traditional software life cycle are covered.
Current research in software engineering and knowledge engineering are also
surveyed.

The first group of articles deal with the basics in software engineering, including
such topics as requirements engineering, domain engineering, object technology,
software architecture, computer languages, program slicing techniques, 
incremental software development, technical reviews, formal verification,
software maintenance, software reliability engineering, management of
inconsistencies in software engineering, software configuration management,
reengineering, software measurement, software metrics for identifying
critical components in software projects, software engineering standards,
engineering access control, usability issues in the software life cycle, software
processes in software engineering and knowledge engineering, and message
sequence charts in the software engineering process.

The second group of articles deal with the basics in knowledge engineering,
including such topics as: conceptual modeling, case based reasoning, logical
abduction in software engineering, knowledge-level models of knowledge
systems, knowledge discovery and data mining, knowledge based
Information access, machine learning for software engineering, and neural
networks.

The third group of articles deal with the interplay of software engineering
and knowledge engineering, including such topics as: pattern based
software reengineering, agent-oriented software engineering, ontologies
in software design, rationale management in software engineering, learning
software organization, software engineering and learning theory, task models
for interactive software systems, software engineering and knowledge
engineering issues for web based education systems, and software
engineering and knowledge engineering issues in bio-informatics.


Overview of Volume Two

Volume Two expands the coverage of the basic principles of software
engineering.  However this volume also addresses many current topics in
software engineering and knowledge engineering such as visual patterns,
multimedia software engineering, etc.

A central issue is how to turn knowledge patterns into visual software
patterns and visual specifications such as UML. Techniques for knowledge
acquisition, software visualization and knowledge visualization, are also of
great interest.

The first group of articles deal with software engineering, including such 
topics as: formal description techniques, software specification and design,
software inspections, component-based software engineering, versions of
program integration, software reuse, assessing different testing strategies
for software engineering and knowledge engineering, verification
and validation, software cost estimation, data model metrics, uncertainty
management, software project management, and reverse engineering,

The second group of articles deal with visual and multimedia software
engineering, including such topics as: multimedia software engineering,
web engineering, object-oriented modeling of multimedia applications,
visual languages in software engineering, software engineering for
visual programming languages, assessing visual programming
languages, visual parallel programming, software visualization,
and visualization of knowledge structures.

The third group of articles deal with emerging topics in software
engineering and knowledge engineering, including such topics as:
software patterns, supporting software processes using knowledge
management, methods for knowledge elicitation, knowledge
elicitation from software code, nonmonotonic reasoning and
consistency management in software engineering, agent-oriented
software construction with UML, improving UML designs using
automatic design pattern detection, application of knowledge-based
systems for supervision and control of machining processes, 
system-level design notations for embedded systems,
situated computing for mobile users, and the synchronization of
interactive web documents.



In a rapidly expanding area such as software engineering and knowledge
engineering, no handbook can claim to cover all the subjects of
interest.  However it is hoped that this Handbook is comprehensive
enough to serve as a useful and handy guide to both practitioners
and researchers at least for a number of years to come.



Shi-Kuo Chang
University of Pittsburgh and
Knowledge Systems Institute