Re: URIQA, schema namespaces etc.

On Mar 16, 2004, at 16:12, ext Danny Ayers wrote:

> Hi Patrick,
> Just a thought - what is/should be considered the Concise Bounded 
> Resource Description if the URI identifies an RDF schema/OWL ontology 
> namespace?

It depends on (a) what is really being denoted by the URI (i.e. is it 
the vocabulary/ontology,
or the particular serialization (document) expressing the 
vocabulary/ontology, and (b) what
the owner of the URI want's to say about it.

One could have distinct URIs for a vocabulary/ontology, it's RDF/XML 
expression, it's N3
expression, it's XTM expression, etc. so one first has to be sure about 
what the URI is
actually intended to denote.

If it denotes e.g. an RDF vocabulary or OWL ontology, then one could 
include in the
description statements defining its title, description, owner/creator, 
restrictions, version, relation to other vocabularies/ontologies, etc. 
I.e. statements
describing the vocabulary/ontology.

One would not expect to find an entire definition of the 
vocabulary/ontology itself
in the description of the vocabulary (nor could you, given the 
definition of a concise
bounded description).

> e.g. what should an MGET on return?

It would return the concise bounded description of the resource denoted
by that URI. It's really as simple as that.

If the owner of the URI isn't clear about what that URI actually 
and ends up with a confusing and even self-contradictory description
then s/he has to be more precise/consistent in the use of that URI in
statements so that the denotation is consistent and unambiguous.

If I had to guess, I'd expect that
denotes a vocabulary, and a description of that vocabulary would
indicate the terms in includes, its relation to other vocabularies,
who owns/defines the vocabulary, perhaps a title and description,

E.g. something along the same lines as the description for the
vocabulary denoted by



Patrick Stickler
Nokia, Finland

Received on Wednesday, 17 March 2004 05:30:53 UTC