Used judiciously, the techniques of object-oriented analysis and design (OOAD) can improve the quality and thoroughness of the design of an automated system. The methodical, disciplined approach to system design is a good one. The various diagrams, used with restraint, can help the analyst or designer identify all the issues that must be addressed in the system to be implemented. In general, the approaches suggested in this book for the various diagrams, including format, notation, and symbology, form a good basis for working tools.
As the authors rightly point out early in the book, each text or reference on OOAD uses somewhat different terminology, espouses somewhat different rules, and uses different symbologies for the various OOAD diagrams. This lack of widely accepted standards for communicating OOAD information is a significant drawback for the practitioners to whom this book is targeted. As a result of the lack of standards, whenever a new project team is formed that will use OOAD, the members have to be trained in the particular set of rules, symbology, and so on that are to be used. Any materials sent outside the project group for review have to include sufficient reference information on the symbology to allow the reviewers to correctly interpret the design material.
The best use for this book would be as a working reference or cookbook for performing system analysis and design with the authors’ set of OOAD tools. This book is difficult to read, and I do not recommend it as an introduction to OOAD. I would not recommend it as a primary text for a course, but it could serve as a good supplementary text.
Some aspects of the book are done well, and others not well enough. The typeface, layout, and format of the illustrations are all good. The examples used to illustrate the concepts presented, while helpful in understanding individual concepts, do not build sufficiently on preceding examples, however. Given the complexity and breadth of the material covered, the examples should have operated as building blocks, each fitting together with previous examples in such a way as to create a coherent whole. After all the concepts were presented, the authors could have used other examples to show the generality of the concepts and tools.
The book attempts to cover too much material in too little space. In less than 200 pages, the authors introduce, define, and attempt to show how to create and use all of the following: domain charts, project matrices, subsystem relationship models, subsystem communications models, system access models, information models, object and attribute descriptions, relationship models, object communications models, event lists, object access models, state process tables, state models, access data flow diagrams, and process descriptions. As a result of attempting to cover so much material, some critical information gets too little attention. Thus, if readers are not already somewhat familiar with the concepts, they may end up confused and frustrated. In any case, several readings are needed to absorb the material.
The index has several page number errors; it looks as though part of the body of the book was repaginated and the index not adjusted accordingly. Some references simply are not included in the index (a key word is referenced on one page and not another). Appendix B devotes 21 pages to instructions on using object-oriented analysis documentation techniques to allow compliance with certain Department of Defense contract documentation standards. This topic does not seem appropriate. These pages could have been better used in presenting the core material of the book than in justifying some funding the authors received for their work. Finally, bound into the book is an advertisement for consulting services by the authors on the subjects covered by the book. If the book had been better, I might not have thought twice about the ad, but given the shallow coverage of the material, I thought it was inappropriate.
