Ingeniouser and ingeniouser

Roy Tennant, user services architect at the California Digital Library, was the speaker here at the Library of Congress last Thursday. In his talk on "Life beyond MARC: The case for revolutionary change in library systems and services," Roy treated his large audience to a series of demonstrations of innovative catalog interfaces that in many cases owe more to the examples set by Google and Amazon than to traditional OPAC design. Roy's presentation provoked much discussion, and one of the questions raised (take a bow, Winton!) was about the extent to which current resource discovery systems are taking advantage of classification data, such as the DDC numbers that are assigned to works cataloged by libraries worldwide. As well as highlighting OCLC's FirstSearch and Curiouser, Roy had shown us RLG's RedLightGreen service, which automatically divides the results of any search into smaller groups of records that share the same LC subject heading, author, or language; could a similar approach be taken so that results are clustered by DDC number? Such a feature would be especially beneficial to the searcher since the clusters themselves could be arranged and browsed hierarchically rather than simply alphabetically. In fact, this is precisely the thinking behind OCLC's DeweyBrowser, which we pointed to a few weeks ago. Thom Hickey of OCLC's Office of Research took up Roy's theme in a recent blog entry -- "Why our catalogs don't work" -- and advocates further Amazonization as the key to satisfying library users' needs. Will the process of Amazonization involve the exploitation of DDC numbers? Can readers point to other examples of library search systems that use classification data in new and interesting ways?