Browse Books

Reviewer: Andreas Schürr

As one of the first attempts to present a complete methodology for the object-oriented (OO) development of software systems, this book is not just another book about a new OO modeling language with some remarks about its systematic usage. On the contrary, it combines the so-called contract-driven life cycle with an extensive list of task descriptions, which are the atomic steps of this life cycle model. These tasks cover all software development activities, ranging from initial requirements identification and business modeling activities to physical database design and user interface design activities. The contract-driven life cycle model is basically a refined version of the fountain life cycle model of Hendersen-Sellers and Edwards. It combines the idea of iterative or incremental software development with the following task concept: any software engineering task supports a distinguished set of process model activities, such as impact analysis or team structuring. It may be executed using certain techniques; needs some resources; and produces a well-defined set of deliverables. The process specification (process model) presented was developed by the OPEN consortium, a nonprofit organization that includes about 30 distinguished researchers from the OO software engineering community. The consortium developed the object-oriented modeling language (and metamodel) OML, which was submitted to the Object Management Group (OMG) as a competitor to the now-accepted Unified Modeling Language (UML) standard. The book complements the OML reference manual [1] and explains how to use OML for the development of distributed information systems. The book starts with a brief introduction to the OPEN consortiums activities. It then clarifies important terms such as software engineering process and discusses the importance of seamlessness (the smooth and reversible progression from one life cycle phase or task to the next) for software development. Next, the authors use approximately a hundred pages to sketch the main activities and tasks of the process specification presented, including such activities as the development of reusable software component libraries. The most important appendix, Appendix A, contains a systematic description of all identified OPEN tasks, in alphabetical order. The book is a must-read for any process modeling engineer who designs or tailors OO development methodologies for a company or for a specific project. It is probably not well-suited as a text for an introductory course on OO development methodologies, for the following reasons: The process specification is rather eclectic, combining more than 100 different modeling techniques (including CRC cards, PERT charts, and Petri nets). The text is full of acronyms, such as SEPA, TOM, and BOM, which forces the reader to make extensive use of the list of abbreviations. Almost all explanations are abstract and not accompanied by any concrete examples. The first concrete example, a context model diagram, appears on page 90, without any real explanation of its purpose. Nevertheless, the book is an invaluable resource on currently available object-oriented software engineering techniques and their merits for the execution of specific activities. To some extent, the approach is independent of the OML language , so it will also be useful for software developers who are using the OMG modeling language standard UML.