- From: Pierre-Antoine CHAMPIN <champin@bat710.univ-lyon1.fr>
- Date: Wed, 15 Nov 2000 10:02:28 +0100
- To: James Tauber <JTauber@bowstreet.com>, www-rdf-interest@w3.org, Jay Roberts <jayr@adrenaline.com>
James Tauber wrote: > XML Schema is about constraining the surface XML. RDF Schema is > about constraining triples (the relationship to XML Schema depends entirely > on the serialization syntax) I agree with that view. See also the Cambridge communique from the W3C, stating that XML-Schema and RDF are distinct - though both WG should work hand in hand... http://www.w3.org/TR/1999/NOTE-schema-arch-19991007 > My view, as I have expressed before, is that RDF should not provide a > serialization syntax the way it currently does. It should provide: > > 1. the triple model [done] ahem... lots of discussion on that list would not take place if it were absolutely true. ;P But well, this is the aim of RDF to provide it, and globally, we have it. > 2. a schema for constraining the model [done] I'm not sure this is a primary aim of RDF. 'RDF Schema' is a spec for describing schemas, but other independant specs should be possible... Although there would be interoperability issues to deal with, my guess is that we could share vocabularies without necessarily sharing the schemas constraining them. > 3. a means for describing, in an XML schema, how instances of that > schema map to RDF triples Once again, I'm not sure that RDF *has* to provide that... It is a shame that RDF M&S provided such a cumbersome syntax with which we have to deam now (since it is the *official* syntax), while the important stuf was the model. A lot of syntaxes may be read as RDF triples, and describing how it can be done is the job of the syntax developers, more than the RDF WG's. IMO Pierre-Antoine -- Sometimes I think the surest sign that intelligent life exists elsewhere in the universe is that none of it has tried to contact us. (Bill Watterson -- Calvin & Hobbes)
Received on Wednesday, 15 November 2000 04:03:13 UTC