- From: Arnold, Curt <Curt.Arnold@hyprotech.com>
- Date: Tue, 22 May 2001 08:12:50 -0600
- To: "'www-rdf-interest@w3.org'" <www-rdf-interest@w3.org>
- Cc: "'www-dom-ts@w3.org'" <www-dom-ts@w3.org>
I have suggested to the DOM test suite WG that it might be beneficial to use RDF to describe the tests to be formulated and their relations to the recommendations, approval status, etc. Unfortunately, my lack of knowledge with RDF has made it difficult for me to effectively advocate that approach. I was wondering if there was anyone here would might be able to participate the the www-dom-ts mailing list or at least be able to provide some guidance on using RDF to describe DOM tests. Here is a synopsis: The definitive definition of a DOM tests is in XML and the test is identified by a fragment URI. A test suite is a collection of tests and other suites and is identified by a URI. Tests will need to identify their source, creation date, versioning information, and relationship to particular passages in the corresponding recommendations. The "big" questions now are: Whether to use RDF or a custom metadata approach? What immediate benefits would using RDF have? For example, could some browser access the RDF file and guide you through the maze. Should the metadata be embedded in the test definitions or in a separate file or both? Is DCMES (http://dublincore.org/documents/2001/04/11/dcmes-xml/) the approach to start with? What, if any, custom properties need to be defined. For example, should the relationship between a test and a suite be expressed using a generic <dc:relation/> tag or should we define more specialized properties like <domts:childOf/>. I took my first baby steps with RDF with a trivial example of DCMES describing some example tests at http://home.houston.rr.com/curta/domtest/sample.dcxml I would appreciate any comments.
Received on Tuesday, 22 May 2001 10:13:43 UTC