Professional Atl Com Programm Ing

Rating:
Summary: Detailed
Review: It's a decent book. Definitely worth the money and much better than "Beginning ATL COM" in that this book is not all about how to use ATL Object Wizard. But just so you know, it is one boring book to read. It takes time n patience to understand how ATL works.

Rating:
Summary: A very tough book.
Review: This is a great book but it is a very painful read. I cannot explain why it is but I certainly learned a lot from it but it took a very long time to read it. Examples are fantastic. Other than the fact that it is hard to read, it certainly is another quality product from Wrox. If you are a little confused with this review, you will be even more confused after read this book. Not bed time reading.

Rating:
Summary: Worth the price (if you work with ATL)
Review: Unquestionably, a strong side of this book is that it is very comprehensive (which is, btw, consistent with his previous books--"Dcom "and "Beginning ATL...") If you work with ATL I'd say this new one is worthy enough to purchase. There are downsides though: he doesn't write very well--that's one, and another is that he sometimes overdoes with comprehensiveness--in the Dcom book he made a few plainly wrong statements--like with Corba providing implementation inheritance, for example. Although, I must add, I bought this brick-book for its ATL content, and this part is ok, except it's a difficult read at times. Finally, someone here mentioned that this book belongs on the same shelf with Brockschmidt, Box, and Petzold. Personally, it pains me to see Petzold mentioned in the same phrase with the other two dim-wits. Grimes doesn't quite make it to Petzold's level. But then, neither is his writing nearly as bad as anything by Brockscmidt (and his goofy koalas). Well, except for the book size, of course . On the yet another hand, I'm not sure ATL is such a great thing--it's MFC all over again, this time--a template-infested version of the same bloatware saga. If you care to really know what you're writing and, btw, make it smaller, faster, and perform according to your wishes (and not to what MS thinks everyone needs) then you might just try to write directly to the Com api, or throw together your own thin helper library. Do not let yourself be lead to believe that you necessarily _need_ ATL. It is an improvement compared to MFC (since you don't need to attach 4meg worth of dlls with every little piece you produce), but still the code it makes is very far from "thin" or "small" .

Rating:
Summary: Good topic, bad contents
Review: Will any book be more frustrating when the codes can't compile, or compiles but won't run?

<< 1 2 >>