Former FIDE World Champion Alexander Khalifman has been writing opening repertoire books throughout this decade, in a number of multi-volume projects. There was "Opening for White According to Kramnik", "Opening for Black According to Karpov", and the most recent series, "Opening for White According to Anand". The players mentioned aren't collaborators in the series, and the games cited and analysis proposed in each book in the series extends beyond what those greats played and wrote. Still, the repertoire Khalifman proposes is well-based on what those champions played (and play), and as such it offers the reader a fundamentally stable repertoire based on main lines tested at the very highest level.
The "Anand" series is up to volume 11 (or XI, to use the numerals used by Chess Stars), and is based on 1.e4. (This was Anand's almost exclusive choice prior to the Kramnik match.) All 444 dense pages of volume XI are devoted to the Dragon and the Accelerated Dragon, and since his treatment of the Accelerated Dragon is based on 5.Nc3 approaches, it's pretty much a pure Dragon book.)
The book came out right at the start of the year, and given Chess Stars' estimable habit of speedy publication, readers can trust the book as expressing the state of the art as of the end of 2008. Likewise, the Khalifman books are consistently detailed, and this volume is no exception. In the opening books by him that I've examined closely, he invariably starts by looking at even poor moves before turning to the meat, and does so without giving short shrift to either the side- or the mainlines. Want proof? You can find the full table of contents (or more accurately, index of variations) here.
I would not consider myself a Dragon specialist for either side (one of the leading Dragon specialists, GM Mikhail Golubev, has reviewed and lauded the book for Chess Today), but there are some lines I have examined for my own use and in annotating games for the blog. One variation I spent a good deal of time on several years ago arises after 1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 g6 6.Be3 Bg7 7.f3 0-0 8.Qd2 Nc6 9.Bc4 Bd7 10.Bb3 Nxd4 11.Bxd4 b5. (This came up for me via an Accelerated Dragon move order: 2...Nc6 3.d4 cxd4 4.Nxd4 g6 5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 d6 9.f3 Bd7 10.Qd2 Nxd4 11.Bxd4 b5.)

White has two main moves here; ironically, they are rook-pawn pushes on opposite sides of the board. White can choose the prophylactic 12.a4 or the sharp and bloodthirsty 12.h4. I prepared the latter move for my intended opponent, found some nice new ideas, and won a nice game (though without getting to use the best part of my preparation). Some time later I saw Bologan successfully use a very similar idea in a game with Moldovan, and I've followed and discussed the path of this variation on the blog off and on the past few years.
Naturally, I was curious to see how well my old analysis held up in Khalifman's eyes, and was surprised to learn that he advocates 12.a4 instead. Even so, Khalifman's consistent thoroughness deserves praise, and he spends over half a page with dense analysis of 12.h4 to justify his rejection of that approach. (Readers should not fear that a large chunk of the book is wasted [from White's point of view] on lines he discards. When Khalifman discusses White alternatives, it's generally to explain why he prefers one major node over another.) His arguments for 12.a4 are plausible: he thinks White gains at least a slight edge with it (generally a clear edge) and that 12.h4 is both objectively good for Black and gives him the easier, more fun position to handle. (I take a look at his suggestions here - see the first game.)
Turning to a more topical variation, he spends a little over eight pages on the almost brand-new "Carlsen Variation": 1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 g6 6.Be3 Bg7 7.f3 0-0 8.Qd2 Nc6 9.Bc4 Bd7 10.Bb3 Rc8 11.0-0-0 Ne5 12.Kb1 a6

Khalifman's treatment includes discussion of the high-level 2008 games Dominguez Perez-Carlsen from Biel, Topalov-Carlsen from Bilbao, Anand-Carlsen from Mainz and Karjakin-Radajbov from Sochi and a whole lot more besides. Naturally, his coverage is deeper than what I offered on this blog (I'd hope so!), but it's still interesting to compare conclusions. (Which I do here - see the second game.)
To conclude this review, let me say that this book appears characteristic of the series. The coverage is thorough and the analysis is deep, but with a fair amount of verbal explanation to help the reader. In the past, I've noticed that Khalifman's analysis can sometimes be a little biased for the repertoire side and perhaps under-reliant on computers (even in tactical lines), but I didn't notice any problems this time around. (Granted, I've only checked a very small percentage of the book.) Even so, I'm happy to have the earlier volumes, and this is an obvious must-have for any serious Dragon player. Similarly, it's a must-have for the target audience as well; i.e. for anyone who faces main line Dragons (and Accelerated Dragons without 5.c4 - mention this, and that an earlier book in the Kramnik series covers that approach). I think 1800s and up can benefit from this book; players below that figure but not too far from it might try it as well, at least if they're devoted Dragoneers. Below that, and I think it's probably both a bit too much and a misallocation of resources - there are far more important things to do than learning heavy-duty Dragon theory.