Well, everyone else is talking about it...
I should start by saying that Watchmen has never been my favourite of Alan Moore's works; I much prefer League of Extraordinary Gentlemen, V for Vendetta, and The Annoyingly Unfinished Ballad of Halo Jones, although I do think it is a strong piece of work, and probably the best English-language superhero graphic novel ever produced.
The film is a surprisingly faithful portrayal of the plot (except for the ending; more below) and characters of the book. From that perspective, it's a complete success and Zack Snyder should be pleased with his achievement, although he may want to consider making something original at some point. However, I think that what's great about Watchmen is not the plot, or the characters, but something else, something in the way that Moore (and Gibbons, but artists working with Alan Moore tend to get bumped down to "illustrator" from what I've heard) explores the very language of comics. I find it difficult to explain what I mean here; I tried with Meg, and she thought I was talking about pacing, but that's only part of it. It's the rhythm of the panels, the juxtaposition of words and images, the recurring visual themes. It's stuff like the covers of the individual comics (or the chapter dividers in the book) doubling as the first panels of the issues themselves, or the bits with the parallel narratives in chapter eight, pages ten to fifteen, or the infamous pirate stuff which is far more clever than most people realise.
For me, it's these bits that make Watchmen what it is, and they didn't make it into the film. Some of them could have been translated into cinematic techniques, but some of them are just features of the medium and can't be carried over, and I think that in an important sense, the book is more about these "mechanics" than it is the plot and characters. Terry Gilliam said the book was unfilmable, and I suspect he was talking about all this under the hood stuff; Snyder passed it over completely, and as such, the film ends up, for me, competent, but ultimately superficial and empty.
This superficial approach also harmed the ending. I have no problem, in theory with swapping the Cthulhoid space squid out for another threat, but I think it was a mistake to go ahead without taking into consideration the rest of the story. Spoilers follow. By involving Doctor Manhattan in the final attack, there is an appearance of cleverness, of drawing things together, but rather it fractures the story somewhat. Veidt formulates an elaborate plan to convince the world that Manhattan has turned against humanity, a plan which hinges at certain moments on predicting how Manhattan will act, and a plan which Manhattan greets with a shrug and vague approval when it's explained to him at the end of the film. So why doesn't Veidt (a man who loves the simple solution, as his discussion of the Gordian Knot attests, although to be fair this isn't in the film) just go to the good Doctor right at the start and convince him to do it himself? This is not a question which can be asked of the book's Veidt, as he doesn't involve Manhattan in the plan at any point, but here, he's integral to it, and in an apparently stupid way. Not to mention that Manhattan, a supposed genius, spends the first half of the film building bombs somehow without realising it, like some big blue thickie. I also find it hard to be convinced by claims that the space squid would be "too silly" for audiences when a purple cgi tiger bounces around without explanation or comment.
I wouldn't call the film a disappointment, as it was well made and entertaining, but because it's a different medium, it misses the deep structures of the book, and for me, it is those deep structures that make the book what it is. In the end, Watchmen the film strikes me as good but ultimately pointless, a triumphant failure. I give it a big Moviewatch 7 out of 10.