Google Book Search lets you see the actual text of public domain books, not just the scanned images. If you click on "view plain text" in the sidebar, you'll only see the pages that contain text. This is useful if you want to copy some text: just select the text and press Ctrl+C. But it also opens up "hundreds of thousands of books to people who use adaptive technologies such as speech output, screen readers, and Braille displays". It's strange that Google only offers the option to download the books as PDFs and not as text files or HTML files, like many other sites (Project Gutenberg's search features are pretty bad). Google doesn't make it easy to find public domain books either: not all the books that are available in "full view" are in the public domain.
Another enhancement is an AJAX-less version of the site, that should work better on older browsers and mobile devices.
Feedly Cloud launches, freeing it from Google Reader
5 minutes ago