- From: Robert E. Kent <rekent@ontologos.org>
- Date: Wed, 8 Mar 2000 14:21:56 -0800
- To: <www-rdf-interest@w3.org>
RDF/S Folk, I have a new short paper on "ontology sharing" that may be of interest to the RDF/S community: "The Information Flow Foundation for Conceptual Knowledge Organization." To appear in the Proceedings of the 6th International Conference of ISKO (the International Society for Knowledge Organization). [http://www.ontologos.org/Papers/ISKO6/ISKO6.PDF] This paper touches on some of the recent discussion topics on this list: constraints such as disjointness or covering, identification versus description, etc. In the paper I have tried to communicate on three different levels: intuitive, mathematical (standard), and mathematical (category-theoretic). As mentioned in the paper, there is a focus on the knowledge bases of communities. To enable ontology sharing in a distributed environment, several principles are used: Principle 1. A community owns its collection of instances: (a) it controls updates to the collection; (b) it can enforce soundness; (c) it controls access rights to the collection. Principle 2. Instances are linked through their types - in order to be able to compare instances of two specific ontologies, we must use the free logic of the generic ontology containing all of its formal instances. I stress that the approach for ontology sharing described in the paper is a "principled approach" in that it has a well-defined foundation, the distributed (hard) logic of Information Flow developed by Jon Barwise and Jerry Seligman. These principles and this approach can possibly be usefully compared and contrasted with the architecture of the Semantic Web. As always, your comments and suggestion will be greatly appreciated. Robert
Received on Wednesday, 8 March 2000 17:25:36 UTC