(I ran out and bought the book and just skipped to chapter 19, the one that discusses cryonics and I haven't finished the final one on mind uploading yet; this isn't a comprehensive review but more of a quick first impression.)
The last two chapters bring explicit religious comparisons. The chapter on cryonics starts out comparing Alcor to Pascal's Wager and takes the comparison seriously (and ignores the whole issue of the Pascal's Wager Fallacy Fallacy), and the chapter on mind uploading starts with talking about human's conception of heaven throughout the ages. "Late in the second millennium, a radically new [conception of heaven] emerged: Heaven is a really powerful computer."
These chapters have passages that I feel are at least mildly confused/confusing/insufficiently Bayesian:
Philosophers can argue until they're blue in the face, and scientists can uncover all the evidence they want, but they can never completely convince us that the body and the brain are machines. The final proof will come only when engineers manage to construct machines that are just as complex and miraculous.
Perhaps I've not been properly exposed to the wider cryonics community, but his first description doesn't describe anyone I know who is signed up with Alcor, including myself:
Many Alcor members might not be eager to see the results of such a test. They may prefer blind belief as a means of consolation about their impending demise. If a scientific test has the potential to uncover factual information refuting their beliefs, they might prefer that the test not be conducted. There may be other thought, who want evidence over faith, and would demand tests of connectome integrity.
The opening to the paragraph immediately before that:
At the present time, cryonics is closer to religion than to science, because it is based on faith rather than evidence.
STILL. Despite all the above, this is a better treatment of cryonics than I expected out of a popular mass market science book and I'm glad Luke made mention of it. That said, without a better reason to continue, I'm going back to finishing Cochran and Harpending's The 10,000 Year Explosion instead of starting on the rest of this.
Philosophers can argue until they're blue in the face, and scientists can uncover all the evidence they want, but they can never completely convince us that the body and the brain are machines. The final proof will come only when engineers manage to construct machines that are just as complex and miraculous.
This statement may be interpreted as
We shouldn't update until we see this final proof - bad thinking
A large number of people ("us") won't update until they see this final proof - probably a true statement
My impression was that he was leaning on the first more than the second, but on second thought that may just be me being uncharitable.
What positive things the book says about cryonics? It asserts that personal identity is contained in the connectome but, from what's been said here, it seems to follow the standard anti-cryonics tropes on other counts. While that would be useful to reference in heated intellectual debates with cryonics opponents (just refute the standard anti-cryonics tropes as per the usual procedure), I'm not sure it would help cryonics' standing among the general public.
Well, I know precious little about neuroscience, but from my EE background I can certainly agree that it's the connections between gates that make a circuit do what it does. Only it's called by a less fancy term, schematics. Maybe I should start calling it "connectome".
There's an interview with him on the Think Atheist Podcast this week. Here:
http://www.blogtalkradio.com/thinkatheist/2012/02/13/episode-45-dr-sebastian-seung-feb-12-2012
Sebastian Seung's new book Connectome: How the Brain's Wiring Makes Us Who We Are is very well-written, and aimed at a broad audience.
The penultimate chapter explains why cryonics might make sense given our current understanding of the brain.
The final chapter does the same for mind uploading.
Transhumanism continues its march into the mainstream.
Despite its flaws, I recommend the book.