Issue No. 01 - January (2002 vol. 35)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/2.976919
<p>With the Interspace, the global information infrastructure will, for the first time, directly support interaction with abstraction. This infrastructure uses technologies that go beyond searching individual repositories to analyze and correlate knowledge across multiple sources and subjects. The Interspace will offer distributed services to transfer concepts across domains, just as Arpanet used distributed services to transfer files across machines and the Internet uses distributed services to transfer objects across repositories. Standard protocols for the emerging information infrastructure will support searching knowledge collections maintained and indexed by specialized communities and residing directly on users' personal machines. These protocols will automatically interconnect related logical spaces, letting individuals navigate across community repositories rather than searching for interlinked objects within physical networks. The Community Architectures for Network Information Systems Laboratory has developed a working Interspace prototype that uses scalable technologies for concept extraction and navigation. They have successfully tested these technologies, which compute contextual frequency of document phrases within community repository, on discipline-scale, real-world collections. Within the next decade, semantic indexing will extend beyond concepts and categories to perspectives, which relate concepts within categories, and situations, which relate categories within collections. These more abstract semantic levels will lead to a closer matching of the meanings in the user's mind to the world's objects.</p>
B. R. Schatz, "The Interspace: Concept Navigation Across Distributed Communities," in Computer, vol. 35, no. , pp. 54-62, 2002.