Marti Hearst, Professor in the School of Information at the University of California, Berkeley, who previously appraised Surf Canyon in a Technology Review article, recently released her new book, Search User Interfaces. In her sweeping review of the state of the art of search interface design, based on both academic research and deployment in commercial systems, Surf Canyon is honored to be mentioned in chapter 9.2.1, “Personalization in Search:”
“The use of implicit information from viewing of documents described in this section makes use of a user’s interaction history over many query sessions. An alternative is to automatically adjust the rankings in real time, during a single search session. Singh et al., 2008 use a version of the clickthrough inversion idea described in Chapter 5 to re-rank the hits returned for a single query and the user views those results. The terms associated with the links that have been clicked on so far are weighted positively, while those not clicked on are weighted negatively, to provide a real-time re-ranking. A similar idea is being promoted by the startup company SurfCanyon. This can be seen as an implicit version of the explicit re-rankings used in the Google search results personalization interface described above.”
The January 14th edition of the Mossberg Solution column in the Wall Street Journal also compares and contrasts Surf Canyon and Google’s SearchWiki.