<< 1 >>
Rating:  Summary: Thinly-veiled advertisement for BoundsChecker Review: As you can probably deduce from the title of the book, the book is a thinly-veiled advertisement for Nu-Mega's BoundsChecker product. Mind you, I do believe BoundsChecker is an outstanding product, but at the same time I believe that this book should have provided me something more than I got for my money.First, the coverage of the Visual C++ debugger itself is not covered "in depth" -- in fact, it is only covered in enough detail to discuss its so-called "serious flaws" -- i.e., reasons you should buy BoundsChecker. Second, Spy++ is only discussed in enough detail to describe its flaws -- personally, I find Spy++ much more useful than this book describes. Third, BoundsChecker itself is not covered in depth at all -- it is only described by way of a few anecdotes about certain classes of bugs. You can get this same information for free by going to www.numega.com and spending a half hour reading their promotional material. Finally, the SoftIce tool, which is indeed a very powerful debugger, worthy of a detailed treatise on advanced debugging techniques, gets a total of one page of discussion in this book. Not only that, but you can download the documentation for Nu-Mega's products for free, and this documenation is much more thorough and up-to-date than what you get from the book being reviewed. The book spends nearly a third of its pages documenting elements of the Visual C++ environment that have very little to do with debugging. This sort of effort is largely futile, since of course the Visual C++ environment has changed since the book was written (as has BoundsChecker). In general, I was expecting the book to consist of more general information that would stand the test of time -- not information that would be out of date before I read it. The book does not discuss many of the subjects that it claims it does. Specifically, it does not discuss "when to fix fugs", except to recommend that if you have found more than three bugs in a debug session, you should end the debug session and fix them. I was expecting a discussion of release cycles, configuration management, etc., in relation to bug fixes. The book does not discuss "how to track bugs" -- it just says that you should be able to re-create any released build, by using a CM tool (no further details provided.) The book does not describe how to design a good beta test at all. The book does not tell you how you know when to release a product. The book assumes in many places that you are using the MFC. In fact, in many places the term "Windows" is used synonymously with "MFC". This book does have four or five short chapters towards the end (each about a ten-minute read, similar to a magazine article) on various classes of program bugs. There are some helpful tips on how to find and diagnose these bugs. This is the most useful section of the book, but it does not constitute a sizable percentage of the book's pages. In summary, I found this book to be almost entirely "fluff". The book's description suggests that it has much more substance than is actually present (much like a typical resume!
Rating:  Summary: NuMega advertisement Review: Extremely thin coverage of an important topic. This is a few hundred pages of advertisement for NuMega's BoundsChecker combined with some simple excerpts from the Visual C++'s debugger manual.
<< 1 >>
|