Using captchas to proofread digitized texts. This is sheer brilliance. Everybody wins by harnessing the time that people are expending filling out captchas, which up until now had been largely wasted time, to do something productive that would be cost-prohibitive to do in any other way.
This article about re-imagining electronic publishing not just as recreating the words of the book on a screen, but also recreating the social, “coffeehouse” aspects of reading and discussing texts online, is pretty brilliant too. (Hat tip: ACRLog.) I cannot wait for publishers to take full advantage of the benefits of e-books, and even though this is a benefit that I hadn't really spent much time considering before I read this article, it might just become my favorite new feature of online books. And I bet it will work fairly well for scholarly books, where the limited number of people reading any given book (and the fact that most of those people will have professional reputations to maintain) ought to keep the quality of the comments pretty high.
Apropos my Cult of the Amateur rant: Tyler Cowen talks about the economics of blogging, and a response by FP Passport.
The down side of open information: People may use freely available online information to do evil things.
Apparently Google Book Search is now making use of Library of Congress Subject Headings data for the books it has digitized. (Hat tip: Cataloging Futures.)
Somebody else is having fun with Amazon's data.