Isn't chat or search or many of the Peer to Peer services fit into the research area of Networked Virtual Environments? Meaning that much can be learned from research that has been done to setup a shared networked environment with scalability and performance. A peer doesn't have to represent a simple node on the network. A framework like JXTA can be extended to represent the more advanced 3D Java API's as a peer and you could start to simulate simple tasks in a visual environment. My point is stressing on the framework. With the right framework all of the other technologies can evolve around the framework and the system as a whole will look spectactular in about 10 years. Thanks for listening.