suggested response re proposed rdfs:Schema class

In http://lists.w3.org/Archives/Public/www-rdf-comments/2003JanMar/0384.html
Daniel Krech asks that we define a class whose members are RDF Schemas. Others
have previously suggested the same, though I couldn't find any in LC issue list.

My suggested response is basically that OWL is the place for such added extras.

Propose:

[[

The WG notes:
(i) many rdf implementors have noted (to the WG and in IG discussions) that
there is some utility in having a class whose members are RDF schema documents
(and/or vocabularies, or namespaces; practice varies).

(ii) that deployed rdf vocabularies often 'self describe' by including rdf 
statements keyed off the vocabularies namespace URI (@@examples?) and that
non-W3C namespaces (eg. Dublin Core) are applicable to that task.

(iii) that theRDF Schema design is very minimalistic, there are many 
characteristics of RDF vocabularies which it does not
provide built-in facilities to describe.

(iv) that W3C created the Web Ontology WG to address these needs, and that the OWL
language (currently in Last Call) provides richer facilities for RDF vocabulary 
description, including a class OWL:Ontology.

(v) that, to our knowledge, there are no RDF vocabularies which
are not also OWL full ontologies (@@check!!); and no OWL full ontologies which are
not also RDF vocabularies.

(vi) that the terminological mismatch ('Ontology' vs 'Schema') re creating a 
class to represent RDF/OWL vocabularies may be regrettable, but
is outweighed by the value of having the RDF user community explore the extensions
afforded by W3C's new OWL vocabulary.
...ie OWL:Ontology is applicable. RDF can encourage vocabulary creators to 
augment their vocabulary with OWL, beginnin with the 'Ontology' construct.
]]


Dan

Received on Friday, 25 April 2003 09:53:57 UTC