I rarely review books, but this one was such a waste that I'd feel guilty if I didn't at least TRY to persuade you not to buy. The main problem with the book, as other reviewers have alluded to, is that it sort of tries to BE a design masterpiece rather than teach you. The design is interesting, but it is to the point that the content is subordinated to the design. There are maybe 5 pages worth of useful content.
Another irritant to me is that this is by far the most commercial book I've ever seen (more ads than a magazine). It seemed that every few pages I was being told to buy this book, or buy all of some other guys books. There is a two page section that is taken from Guy Kawasaki's blog. The whole thing is incredibly derivative (but shallow).
Finally, and this I can take responsibility for, it didn't meet my needs because I am not Al Gore, nor am I the CEO of a company. If I gave a presentation in the style of this book I'd be laughed out of the office. I'm an analyst, and the warm and fuzzy slides would not be good. Duarte's advice on data intensive slides? "Both [of Stephen Few's] books should be on your shelf along with everything Edward Tufte has written."
I have two Tufte books and they are virtually identical, I don't see myself buying his others (and they suffer from the same lack of relevance to corporate analysts). Disappointing.