Hiam Kilov - Böcker

This text encourages builders of complex information systems to accelerate their move to using the approach of a scientific discipline in analysis rather than the approach of a craft. The focus is on understanding customers' needs and on precise specification of understanding gained through analysis. Specifications must bridge any gaps in understanding about business rules among customers, subject matter experts, and "computer people", must inform decisions about reuse of software and systems, and must enable review of semantics over time. Specifications need to describe semantics rather than syntax, and to do that in an abstract and precise manner, in order to create software systems that satisfy business rules. The papers in this book show various ways of designing elegant and clear specifications which are reusable, lead to savings of intellectual effort, time, and money, and which contribute to the reliability of software and systems. This text offers a treatment of the object-oriented paradigm by examining the limitations of traditional OO methodologies and by describing the significance of competing trends in OO modeling.It builds on four years of successful OOPSLA workshops (1991-1995) on behavior semantics. This book deals with precise specifications of `what' is accomplished by the business and `what' is to be done by a system. It includes descriptions of successful use of abstract and precise specification in industry. It draws on the experience of experts from industrial and academic settings and benefits from international participation. Collective behavior, neglected in some treatment of the OO paradigm, is addressed explicitly in this book. The book does not take "reuse" of specifications or software for granted, but furnishes a foundation for taking as rigorous an approach to reuse decisions as to precise specifications in original developments.