| The Semantic Web is a vision of a web of machine processable data, encoding knowledge, that agents can reason about to process complex tasks. This vision relies on large amounts of data describing interrelated resources. One approach to producing such data is the "Web 2.0" approach, of collaborative production of content by end users.;Building on earlier work where this paradigm was applied to structured documents, we extend this approach to support a graph data model. We introduce a formal model for schema-based file sharing systems, then extend this model with links between shared documents. We propose a query language for this graph of interlinked documents. Finally, we describe the design of our prototype implementation.;We seek an alternative to the centralized control model of web-based solutions. Our approach uses the principles of peer-to-peer file sharing, where data is also contributed by end users, with decentralized control, and a data circulation model where files are replicated by downloads, causing popular or useful files to spread in the network. |
