Since canceling my OpenGL ES 2.0 for iOS 4 book, I've had several people request book recommendations to use instead of my book. Honestly, I didn't really have one to give before today, partially because I intentionally avoided reading competing books while working on mine.
Today, while stuck on a train, I checked out Philip Rideout's iPhone 3D Programming by O'Reilly. Now, it's never easy to give an objective opinion on a book that competes with one you're writing, and even though my book is currently not on a production schedule and I have no time to work on it, I tend to still think of myself as working on an OpenGL ES book. Although there are several things that I would have done differently, I also know it's unfair to judge a book that way. Deciding what to include and what not to include in a book is a horribly difficult thing with an expansive topic like OpenGL ES and no two authors are going to make the same exact decision in that regard, so I'm not going to criticize Philip Rideout for, well, for not being me. Hell, normally not being me is taken as a bit of a compliment anyway.
Yet, though it's not the book I would have written, I can wholeheartedly recommend it. The information is solid and useful and it's presented in a way that doesn't assume or require a Ph.D in math. One thing that Rideout does do that I opted not to even attempt, is to give both the OpenGL ES 1.1 and OpenGL ES 2.0 approach to tasks that can be done in both environments, which is actually interesting on multiple grounds. One, it's a great transition for people who know 1.1 but not 2.0, but also, it shows the interesting ways people have figured out to work around the limitations of the fixed pipeline. These kinds of solutions fascinate me because they're the result of people refusing to accept technical limitations.
Much (all?) of the material from this book is available free online, but I'd recommend buying the book, which gives you access to the book's source code and also will also help an angel get his wings. Well, I'm not 100% sure on the latter, but it seems likely this time of year.
If you're floundering a bit with OpenGL ES, this book is well worth your time.