RE: Schema spec

OK, I understand about the ns processing.  Thanks for the pointer to Ralphs
comments.  

I came across this when I used SiRPAC to read the schema RDF as given in the
spec.  The spec provides a serialization of an RDF model for RDF Schema
which produces the wrong model when fed into the RDF sevlet on the W3C site.

Note that I made two changes to the original: one was the inclusion of the
name space, the other was changing 'ID' to 'about' which I think is
consistent with what Ralph is saying.

Brian McBride
HPLabs


>-----Original Message-----
>From: Dan Brickley [mailto:danbri@w3.org]
>Sent: 21 April 2000 17:21
>To: McBride, Brian
>Cc: RDF Interest Group
>Subject: Re: Schema spec
>
>
>
>
>On Fri, 21 Apr 2000, McBride, Brian wrote:
>
>> Appendix A of the schema spec gives an RDF serialization of 
>schema which
>> begins:
>> 
>> <rdf:RDF
>>    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>>    xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#">
>> 
>> <rdfs:Class rdf:ID="Resource">
>>   <rdfs:label xml:lang="en">Resource</rdfs:label>
>>   <rdfs:label xml:lang="fr">Ressource</rdfs:label>
>>   <rdfs:comment>The most general class</rdfs:comment>
>> </rdfs:Class>
>> 
>> Should the 4th line read:
>> 
>> <rdfs:Class rdf:about="rdfs:Resource">
>
>
>I think this is really an RDF syntax question. RDF syntax doesn't
>currently allow xml-namespace qualified names inside attributes. At the
>time, there was a widespread desire to do so but an awareness that such
>abbreviations would be invisible to other namespace aware processors,
>which might set about re-writing namespace prefixes on element and
>attribute names, breaking the linkage between 'rdfs:' and the 
>full URI if
>we used qualified names in attribute content. Now we (nearly) have XML
>Schema, it could be easier to express such usage in a fashion that
>machines can be made aware of.
>
>The intention is that the construct we use represents an RDF model
>equivalent to that which parsers would produce if encountering:
>
>	<rdfs:Class 
>rdf:about="http://www.w3.org/2000/01/rdf-schema#Resource">
>
>if the context was that the 'base URI' for the syntax parsing was
>http://www.w3.org/2000/01/rdf-schema#
>
>
>Ralph Swick has just posted some very detailed comments on 
>this in reply
>to a submission to the rdf-comments list; it might be good to move
>discussion of that to the RDF IG forum:
>
>http://lists.w3.org/Archives/Public/www-rdf-comments/2000AprJun
/0014.html

Dan

Received on Friday, 21 April 2000 13:24:47 UTC