Sharing work, experience, interpretation, and maybe even meanings between natural and artificial agents

Multi-user virtual environments (MUVEs) are becoming increasingly popular as environments where humans and artificial agents can share collaboratively the use of a variety of different information processing and visualization tools. The paper builds on a new concept, first proposed at the NIST 1996 Semiotics conference that MUVEs provide an exciting testbed for developing new semiotic concepts and mathematical foundations. The approach is to provide a computer-mediated, instrumented environment in which all agents´ actions are observable, permitting an analytic basis for recording and analyzing all agents interactions and results. The hope is that this `shared experience´ and common interpretations within a defined context will allow the development of notations that represent the `meanings´ among agents in that environment