Browse Books

Reviewer: Murat M. Tanik

Cook and Daniels are longtime users and developers of object technology. As such, they give their book an industrial and European flavor. The emphasis on precision in the introduction and use of object-oriented terminology is one of the strengths of this book. Studying most object-oriented design and methodology books is frustrating because of inconsistency between books and informal presentation of notions. It is not easy to introduce rigor to an area lacking basic formal foundations. The authors take the middle-of-the-road engineering approach by paying attention to precision and the necessary degree of formalism to achieve a practical objective. Therefore, this can be considered a book for software practitioners. I like this book because it does not try to present a rigid set of rules to be precisely followed as a step-by-step methodology of software development. Like any sound engineering approach, it provides a set of techniques and tools and suggests that you can use them in any order to get the job done. The authors recognize that there are at least two schools of thought on software development: “there are two communities which at first hand seem to have remarkably little in common, and yet which are actually trying to solve the same problem. The first community is that of methodology authors who invent informal analysis and design notations. The second is the formal methods community.” The preface states that one of the goals of writing the book, in addition to contributing to the philosophy and practice of object-oriented software development, is “to help build some bridges between these two communities.” The modeling method described in the book, the Syntropy method, is a synthesis of several approaches, including the methods of Rumbaugh and Booch. The book is divided into five parts and three appendices. Parts 1 and 5 are each composed of one chapter. These chapters describe the authors' philosophy of object-oriented method and its practical implications and limitations. The authors believe that effective modeling requires discussing design from three viewpoints—those of an observer, a software specifier, and a software implementor. Following their ecology of software, Part 2 (chapters 2 to 5) covers aspects of modeling the world. Part 3 (chapters 6 to 9) covers techniques for modeling software, and Part 4 (chapters 10 to 12) covers relationships between models, domains, and reuse, which are collectively called system architecture. Three appendices provide summaries of notation, value types (built-in and user-defined), and object identification techniques, respectively. Prospective readers should be prepared to spend considerable time reading and mastering the techniques. This book is not written for the casual reader.