W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > March 2003

Re: LC Issue xmlsch-08 proposed response

From: Dave Beckett <dave.beckett@bristol.ac.uk>
Date: Fri, 21 Mar 2003 14:08:04 +0000
To: w3c-rdfcore-wg@w3.org
Message-ID: <9518.1048255684@hoth.ilrt.bris.ac.uk>

>>>Dave Beckett said:

<snip/>

On this one, the examples I gave and the analysis were slightly
wrong, another few reasons to delete it from the propsoed response.

> 7. xsd as a magic namespace
> 
> Related, it was also considered that the simple XSD datatypes (only)
> are so useful that they should be automatically recognised in
> rdf/xml, such that they could appear in documents using the term
> xsd:string etc. alone.
> 
> In RDF this would be with the namespace prefix
>   xmlns:xsd="http://www.w3.org/2001/XMLSchema#"
> so that it constructed the URIs correctly using the RDF 
>   uri=concatenate(namespace-name,local-name)
> method.
> 
> In that way you might have been able to do:
> 
>   <ex:Node xsd:string="foo" ... />
> 
> but not in the element form since it would be like;
>   <ex:prop xsd:string="??what goes here???">foo</ex:prop>

This examples are wrong for the current datatype solution and was
being discused when we were proposing to use (the URI of) xsd:string
as an rdf property. In that case the two examples would have been:

   <ex:Node xsd:string="foo" ... />

   <xsd:string>foo</xsd:string>

but isn't so relevant to the current state.  I don't want to even
hint of re-opening this in a response :)

> 
> This important set of types was considered building in but the
> flexibility of using any URI won.

Dave
Received on Friday, 21 March 2003 09:08:43 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:56:15 EDT