Foreword to the Third Edition

Bruce Douglass' book has only improved with time. The main change, of course, is that it now caters to the new version of the the language, UML 2.0. As with the first edition, this edition too is one of the clearest and most valuable texts for engineers who want to model and specify systems using the UML, especially reactive and real-time ones. Hence, I applaud Bruce for updating the text and presenting to the public another valuable product of his prolific pen (keyboard?...).

Still, I should say a few words about the UML itself, especially relating to the following two passages from the earlier foreword one a prediction and one an opinion:

The recent wave of popularity that the UML is enjoying will bring with it not only the official UML books written by Rational Corporation authors, but a true flood of books, papers, reports, seminars, and tools, describing, utilizing, and elaborating upon the UML, or purporting to do so. Readers will have to be extra careful in finding the really worthy trees in this messy forest. I have no doubt that Bruce's book will remain one of those....

Despite all of this, one must remember that right now UML is a little too massive. We understand well only parts of it; the definition of other parts has yet to be carried out in sufficient depth to make crystal clear their relationships with the constructive core of UML (the class diagrams and the statecharts)....

As to the first of these quotes, it wasn't too hard to predict the flood, which has materialized above all expectations. Here is one small statistic: A search at amazon.com for books with "UML" in the title results in 213 items, and the same search limited to 1998 and on yields 198 items. That is, there were 15 UML books when the first edition of this book was published, and there are some 200 more now! Nevertheless, I maintain that Bruce's book indeed remains one of the few really worthy ones.

As to the second remark, about the UML being a little too massive, things have not really improved much. With version 2.0 almost ready to be launched, a fact that is doubtless a milestone in the development of the UML, we may ask ourselves whether it has become leaner and meaner, or larger and messier. Many people hoped that a new version of of something that was so multifaceted and complex, but which had been adopted as a standard to be used worldwide, would concentrate on its most important aspects. It would improve and sharpen them and narrow down or discard those things that turned out to be inessential or less well-defined. This could have resulted in a language that was easier to learn, easier to use, easier to implement responsibly, and thus would carry a lot more "punch." While UML 2.0 contains several exciting new features, especially for the realm relevant to this book real-time and reactive systems the new version of UML is larger and more complex.

As mentioned in the 1997 foreword, object-orientation is here to stay and so is the UML, probably in a big way. Let us thus hope that version 3.0 of the language will remove, intergrate, clarify, and solidify more than it adds. In any case, good books about a language are almost as important as the language itself, and in this respect the present book is one of only a handful that can be heartily recommended.

David Harel
The Weizmann Institute of Science
Rehovot, Israel
November 2003