8 of 8 people found the following review helpful
on December 25, 2001
Believe it or not, we really do need a book on event handling in Java. Events are a pervasive form throughout the SDK classes, not just to AWT and Swing. They inform the structure of JavaBeans, Java Messaging Service (JMS), and are important to EJB, Jini, and Jiro. An OO methodologist may say events are "just" an expression of the Observer design pattern, but from that simple pattern spring at least four major variants used in Java, each one worth careful study. Once you understand event-handling in Java, the structure of many classes become quite clear.
But 590 pages? Why so many? Oh...it's a reference book, another API treatment that re-documents some classes, and adds some light code examples. I'd estimate 460-70 pages of the book do that and only that. The opening text tells the reader it will reveal the power of event handling -- which to me is like learning a meal will be best when it's hot -- but then concludes that learning by doing is the only way, and delegates many important side-trips to other sources.
There's little graphic relief, and no visual modelling at all. Discounting the lists of class methods, there are 2-3 class hierarchy diagrams in the book, and one screen shot. The latter is located somewhere after the bulky reference section -- bad placement, in my opinion.
There is no mention of how events drive JavaBeans, nor any satisfying description of local versus remote event-handling. That's not a crime, but that's what I wanted in place of another javadoc rewrite. The Jini overview is only two pages; and it's now somehow been covered? Even the Jini code samples themselves are borrowed from Core Jini. The reader is referred to that book for an explanation; it's unreasonable to think the reader can make that leap easily.
An example using RMI would be simpler and more appropriate. Remote event-handling has been covered for years in the periodicals. Why not cross the tee on those discussions with a full-length treatment here?
This book feels like it lost its way through the tedious business of describing classes. The event handling book I hope for is suggested in these pages, but doesn't materialize. I think the real work got left off until the end, when there was no time to think and the deadline was in sight.
If events were "new," this book might have reason to be an API reference. But they aren't new, and certainly not mysterious. To write API treatments, you need an eager, information-deprived audience, and it's essential in my opinion that the author show a reading of the source behind the API in order to comment with authority. javadoc is not completely helpful, sure, but the shortcomings aren't so egregious we need to publish every "better" set of class definitions either.
3 of 3 people found the following review helpful
on August 31, 2001
Java Event Handling has to be the most complete collection of documentation in one book on the Delegation Event API provided by java. Not only an easy reference with lots of excellent examples to demonstrate, the explanitory texts are well written, easy to follow, well organized and gives more than enough information to model effective event handling and management routines in your application. You won't find a better source of information on this topic anywhere else.