- From: Peter F. Patel-Schneider <pfps@research.bell-labs.com>
- Date: Fri, 14 Dec 2001 12:17:20 -0500
- To: www-webont-wg@w3.org
- Cc: www-archive@w3.org, hendler@cs.umd.edu, connolly@w3.org
- Message-Id: <20011214121720U.pfps@research.bell-labs.com>
Providing Syntax for SWOL The syntax for the Semantic Web Ontology Language (SWOL) may be more contentious than its semantics. In particular, there may be several syntaxes, to support several modes of interaction. I think that is would be a good idea to get the details of one syntax ironed down. I've picked on an XML syntax for SWOL. The Basic Problem: SWOL is supposed to be an extension of RDF. As such, the semantics of SWOL should be upward compatible with the semantics of RDF. That is, that SWOL entailment reduces to RDF entailment on RDF knowledge bases. However, if the syntax of SWOL is just RDF triples, then there will be RDF knowledge bases where this is not true, namely RDF knowledge bases that contain triples that encode SWOL syntax. Therefore, I have developed a draft syntax that distinguishes between the RDF and non-RDF parts of SWOL. The Approach: I have tried to develop a syntax that allows SWOL syntax processing to use standard XML tools as much as possible. To that end I have specified SWOL syntax in terms of the XQuery 1.0 and XPath 2.0 Data Model. This makes the syntax harder to develop and harder to read, but should have important benefits for implementers. An initial stab at the syntax specification is given in an attachment. Examples of the syntax are given in another attachment. Note: The syntax is much less developed than the semantics. I welcome any help in determining better ways of doing this. In particular the syntactic names are taken from DAML+OIL and there are undoubtably better ones available. Action Items for the Clean-Up Task Force: [Again, relevant comments from other WG members are welcome.] 1/ Review the attachment. 2/ Determine if you disagree with the following: - Upward semantic compatibility with RDF is important. - The proposed solution allows this upward semantic compatability. - The proposed solution does not introduce other problems. 3/ Make suggestions to improve the syntax or fill in holes in the syntax. Schedule: morning of 20 December 2001: comments on the approach back to me 7 January 2002: draft document for face-to-face discussion
Attachments
- Text/Plain attachment: syntax-examples.text
- Text/Plain attachment: syntax-data-model.text
Received on Friday, 14 December 2001 12:19:11 UTC