The Semantics of Multi-user Virtual Environments

Multi-user Virtual Environments (VE) are used today for a number of different application domains. Traditionally, their presentation and interaction designs focus on human users, and are rarely suitable for machine access. This is complicated by the large number of different model formats and network protocols in use. Accordingly, it is difficult to develop new VE software, such as agents, services, and tools. Recent research has introduced the idea of combining traditional VE techniques with emerging Semantic Web techniques, resulting in Semantic Virtual Environments (SVE). In this paper we investigate the application of this idea to the particular problems of multi-user VE modeling and dynamics. We show how a machine-understandable semantic abstraction can form the basis for VE software that is independent from particular VE systems and could even operate across VE system boundaries.