Earlier this year, a small group of people gathered at Internet Archive's San Francisco office to discuss whether this was possible. Could we build something so grand? We concluded that we could. We located a copy of the Library of Congress card catalog, phoned publishers and asked them for their data, created a brand new database infrastructure for handling millions of dynamic records, wrote a new type of wiki that lets users enter structured data, set up a search engine to look through it all, and made the resulting site look good.
We hooked it up to the Internet Archive's book scanning project, so that you can read the full text of all the out-of-copyright books they've made available. And we hope to add a print-on-demand feature, so that you can get nice paper copies of these scanned books, as well as a scan-on-demand feature, so you can fund the scanning of that out-of-copyright book you've always loved.
This is all Aaron Swartz, by the way. I told you you should have voted for him. (Also, check out the UI overview for an endless fount of coolness.)