Goodreads, the largest social networking site for readers, launched its new recommendation engine today, which draws on ratings made by its nearly six million members on the 190 million books in its database. In March 2011, Goodreads bought the smaller site Discovereads to obtain its recommendation algorithm, which forms the basis of Goodreads' recommendation engine. The system appears similar to those on sites such as the DVD rental giant Netflix or book retailer Amazon. Books are recommended for site members, based on star ratings and user-generated genre categorizations. New users are asked to rate a series of popular books to generate recommendations immediately (pictured). Otis Chandler, CEO of Goodreads, told LJ that "down the road, we want to be doing a lot more with this algorithm," including suggesting books for Goodreads-based book clubs, as well as recommending book clubs for members to join. Last June, Goodreads struck a deal with EBSCO's NoveList, a readers' advisory service used by libraries, in which NoveList would integrate Goodreads' 11 million-plus book reviews and 100 million ratings. Chandler said that he didn't know at this point whether NoveList would be syndicating the new Goodreads recommendations, though he did say that Goodreads was planning to make the information accessible through its application programming interface (API). (NoveList did not immediately respond when contacted for comment.) Many libraries, including the San Diego Public Library and Salt Lake County Library Services (SLCLS), use Goodreads for book clubs and discussion groups. Jennifer Fay, assistant manager at SLCLS's Whitmore Library and the moderator of the system's Goodreads book discussion group, told LJ that its members were always asking for recommendations, so the group would likely make use of those provided by Goodreads. She also said that she would welcome Goodreads making book club suggestions, as it could help grow the SLCLS group's membership. "Any way to get our name out there would be great," she said.