Information Retrieval and the Virtual Document

After many years building the foundation of the digital library, information retrieval has emerged from its roots in bibliographic retrieval to establish its role in hypertext and Web applications. We are now in the midst of a “seismic” shift of digital content and context from the global repository [i.e., memex (Bush, 1945)] through docuverse (Nelson, 1981) to the global resource. From a repository of archived documents with embedded links we have seen the emergence of CGI, JavaScript, Java, ActiveX, and XML, all tuned to providing dynamic and application oriented use of information resources. The digital library is merging with the digital workplace and the digital marketplace. Many of the “documents” retrieved are virtual documents. That is they are created on the way to the user rather than copied from a digital repository. This has a significant impact on the notion of information storage and retrieval. What does it mean to search nonrepository data? How is information authenticated or referenced when it comes from different sources and processes? Who owns it? Although the wide availability of Web resources makes this problem obvious, it is not restricted to the Web.