I feel bad for the authors; it sounds like they're taking the heat for all of the negative vibes surrounding this book, when in all likelihood the responsibility lies with the editors and/or publishers. It definitely smacks of "get it out the door quick" marketing.
I won't mention the atrocious code samples in the book (it's true...they're horrible. At least the downloadable codes seems to have fixed that -- thank you, authors).
I won't mention the spelling and grammar mistakes (again, though, it's true...the editor(s) must have just verified that the manuscript was, in fact, about ActionScript and called it good).
I will mention that I found the book generally useless. To be fair, I read Moock's Essential ActionScript 2.0 first, which I personally found much more informative and worthwhile. I followed it up with OOPAS2, thinking that I could only solidify my OOP knowledge, but would up feeling that it was a darn good thing that I read Moock's book first, or else I would have been totally lost in OOPAS2. I just don't feel that the book is well organized or that it presents ideas effectively.
At some points conepts are belabored until they bleed, and other concepts are just sort of casually mentioned and never heard from again.
The example files and exercises aren't exactly useful. Contrast that with Essential AS2, which provides a useful class for nearly every full example. Let's see, from Moock you get an design pattern frameworks, an ImageViewer, a Logger, and several other rather useful class files. In OOPAS2 you get a Loan application and a Magazine subscription application.
To make another comparison with Essential AS2, that book closes with an excellent introduction to Design Patterns, which was the first time MVC made sense to me. OOPAS2 doesn't talk about Design Patterns.
OOPAS2 does spend the middle third of the book talking about XML, Web Services, Flash Remoting, and Flash Communication Server. Those are all extremely great things to talk about, but the connection to Object-Oriented Programming was weak at best. That whole section just seemed tacked on. In all, it did the book a huge disservice by taking pages away from "real" OOP stuff, and consequently trying to cram a huge thing like FlashCom or Remoting into a single chapter. Come on...write a book on Remoting, don't shove it into a book on OOP.
I need to stop an remember that there were actually a few redeeming qualities about the book, but they are far outweighed by the negative qualities.
If you can only afford one OOP for AS book, make it Moock's book (I haven't read the new OOP for Flash 8 book from Freinds of Ed, that one looks worthwhile, too). I wish I could even recommend getting both, but really, the book is so scattered that it's really not worth it. If you're serious about OOP, you won't care about the middle section too much. And if you want to learning Remoting and the like, then you're much better off buying a dedicated book(s) on the subject. And the last section where they build the sample application is just, well, "eh." I rarely don't finish something that I start, but they just lost my interest in the last section. I couldn't see any point in actually building the app that they were building.
Go with Moock's book or the new Friends of Ed one (hopefully).