W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > December 2002

Re: checked RDF semantics for XSD stuff, couldn't grok namespace entailment

From: Dan Connolly <connolly@w3.org>
Date: 16 Dec 2002 10:29:25 -0600
To: Patrick Stickler <patrick.stickler@nokia.com>
Cc: Brian McBride <bwm@hplb.hpl.hp.com>, ext pat hayes <phayes@ai.uwf.edu>, w3c-rdfcore-wg@w3.org
Message-Id: <1040056165.11337.35.camel@dirk.dm93.org>

On Mon, 2002-12-16 at 03:12, Patrick Stickler wrote:
> 
> [Patrick Stickler, Nokia/Finland, (+358 40) 801 9690, patrick.stickler@nokia.com]
> 
> > The big content issue is whether plain literals are XSD strings. That 
> > needs to be decided, soon, and may require changes in several 
> > documents. I plan to simply be conspicuously agnostic on this issue 
> > in this draft of the semantics, by deleting the paragraph that Dan 
> > objects to. However I don't think that we should go to last call with 
> > this an open issue, I'd like to get it clear one way or the other. 
> 
> For me, this issue is similar to whether xsd:nonNegativeInteger
> and UAProf prf:Number have equivalent value spaces. I think it's
> pretty clear they do. But it's not for RDF to say so.

Well, prf:Numbers aren't specified by the RDF specs.
Plain literals are.

> Just as Jeremy was able to justify equivalences and relations
> between various XML Schema datatypes which were not explicit
> (or as explicit as we'd like) in terms of RDF datatyping semantics,
> based on the definitions of those datatypes; so one could justify
> the equivalence of RDF plain literal values and the values of
> various arbitrary datatypes, such as xsd:string. But again,
> RDF is datatype framework agnostic and it is not for RDF to say.

It's not for RDF to say how its own syntax (plain literals)
works with XML Schema datatypes?

I disagree. Our charter includes...

"RDF Schema must use and build upon XML Schema datatypes to the fullest
extent that is practical and appropriate."
 -- http://www.w3.org/2001/sw/RDFCoreWGCharter

I find it practical and appropriate to use XML Schema datatypes
in this case.

> So, I agree that the membership of xsd:string values (probably) 
> coincides with the set of RDF plain literals. *However* the place to 
> state this is in a Note, such as that already proposed, which clarifies 
> issues regarding the use of XML Schema datatypes with RDF. *Not* in
> the RDF specs themselves.
>
> So my recommendation is to say nothing at all about any relationship
> with xsd:string and plain literals in the RDF specs (and this is not
> the same as leaving it an open issue -- as it will/should be addressed
> in a separate Note addressing such XML Schema related issues).
> 
> It's enough to be clear what plain literals are, so that one can
> be clear and precise about their possible relationship to particular
> datatypes such as xsd:string and others. The actual relationships are
> out of scope.
> 
> I expect/hope that Dan, and others, would be satisfied with that.
> 
> Yes?

No.

> Patrick
-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Monday, 16 December 2002 11:29:44 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:54:55 EDT