I read this book about 30-Nov-2005. This is the first time I've read this book. The book is copyright 2006. This note was last modified Thursday, 01-May-2014 17:34:44 PDT.
This note contains spoilers for the book.
This is an uncorrected bound manuscript—an advanced reading copy from the publisher (Tor books). Amazon thinks it will be released in May 2006. (Amazon also thinks it's a "zones of thought" book; it's actually a near-future Terrestrial thriller, with no mention of singularity, aliens, or the zones of thought, or any obvious connection to that universe. It could of course be set in that universe before Earth discovers anything about it.)
It relies more on character, and does more with character, than recent Vinge books. It relies less on tech, and has less interesting tech to it, than recent Vinge books. It doesn't, for me, have the feeling of a fresh wind blowing through that the last two have given me.
Instead of the singularity (with uploading and AI) we get wearable computing and advanced user interfaces, and ubiquitous networking, and the Secure Hardware Environment (pervasive hardware-based DRM and software control), and of course people who can get around all that.
We also get an old poet and English professor, treated successfully for Alzheimer's and having to find a way to fit into the new society. He's been sent to Fairmont High, along with a bunch of slow kids and some fast kids. It's sort of a reverse coming of age story in some ways—an old guy learning to do better than he did last time.
One thing I think he's doing is showing a near-singularity situation, and showing just how far beyond comprehension it is for a good number of the adults living and working then.
It's a good solid work, even better than just "solid", but doesn't pack the kind of punch his last two books have. I'm afraid people will see it as a severe disappointment (and those who hate the "zones" and the singularity will almost have to see it as a great improvement).
The really clever tech thing in the book is the idea that software continues to get fast faster than physical devices, leading to a situation where the efficient way to scan a bunch of books is to shred them into little squares, blow them down a tube, and have high-res cameras capture a lot of images of them, which the software then reassembles and OCRs.
But it's used stupidly. For no apparent reason the only choice available is to shred-and-scan everything in a library, or nothing. This leads to artificial conflict. Rational people will quickly realize that most of the works in any given library are relatively modern works issued in large quantities, and the loss of one physical copy is trivial compared to the gain of a good digital scan of it—but that a few of the works in the library are rare or physically valuable in some way and should not be destroyed for this purpose. So, everybody bands together to raise the funds to scan the rare books without destroying them, right? But no, nobody is that smart, it doesn't even get proposed. WTF?