Adaptive collaboration for wired and wireless platforms

The Disciple and Manifold (I. Marsic, 1999; 2001) frameworks, developed at Rutgers Center for Advanced Information Processing (CAIP), take a data-centric approach to the development of collaborative multimedia applications. (Disciple is an abbreviation for Distributed System for Collaborative Information Processing and Learning.) This approach represents data and modifies it in a generic way via a standard representation medium and a standard communications protocol. Each user can obtain a subset of the shared data that can be visualized differently for different users (polymorphic views). The amount of data presented and the visualization technique reflect a user's interests and, more important, a device's computing and communications capabilities. The article begins by introducing a data-centric architecture that abstracts collaborative tasks as editing of data repositories, followed by descriptions of the role of XML in managing heterogeneity and intelligent software agents in discovering network and computing environment conditions. Finally, it presents an instantiation of a furniture arrangement application used to test the approach.