W3C home > Mailing lists > Public > www-webont-wg@w3.org > January 2003

Re: XSLT: datatype problems (*)

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Wed, 22 Jan 2003 19:19:14 +0000
Message-ID: <3E2EEEB2.1020102@hpl.hp.com>
To: Jerome Euzenat <Jerome.Euzenat@inrialpes.fr>
CC: Mike Dean <mdean@bbn.com>, www-webont-wg@w3.org



Jerome Euzenat wrote:

> 
> A follow-up on my previous mail. Can I write:
> 
> <owl:oneOf>
>  <rdf:List>
>     <rdf:first datatype="xsd:integer">1</rdf:first>
>     <rdf:rest>
>        <rdf:List>
>           <rdf:first datatype="xsd:integer">2</rdf:first>
>           <rdf:rest>
>             <rdf:List>
>               <rdf:first datatype="xsd:integer">3</rdf:first>
>               <rdf:rest>
>                 <rdf:List>
>                   <rdf:first datatype="xsd:integer">4</rdf:first>
>                 </rdf:List>
>               </rdf:rest>
>              </rdf:List>
>            </rdf:rest>
>          </rdf:List>
>        </rdf:rest>
>      </rdf:List>
>   </owl:oneOf>
> 
> for my initial question:
> 
> <owl:oneOf rdf:parseType="Collection">
>   1 2 3 4
> </owl:oneOf>
> 
> ??


The first is OK the second is not.
See Peter's msg:
http://lists.w3.org/Archives/Public/www-rdf-comments/2003JanMar/0027.html

Note that this has become a postponed issue for RDF (IIRC).

Note also that this is up for discussion tomorrow under the AS&S, wunder 
the topic of my agenda request msg item 2 dataRange. [1]

I am half proposing that dataRange be removed from AS&S on the grounds that 
it wasn't in DAML+OIL and has not been formally added to the language. (and 
is also not reflected in the other docs)
I tend to agree with Mike that dataRange is a disguised form of datatyping 
and should be left to the XML Schema group (even if we are not wholly happy 
with what they do).
There is an XML Schema requirements doc out [2] more on that next.

 > This would have some consequence for OWL implementers in RDF that they
 > should be able to understand it in order to do valid inference such that:
 >
 > IF
 >     PropertyConstraint( C, P, oneOf( "bla" ) )
 >     PropertyConstraint( C, P, MinCardinality( 1 ) )
 >     O rdf:type C
 > THEN
 >     O P "bla".


uncompelling, quite a lot of knowledge of XSD is already required to use 
OWL effectively.
Consider
 > IF
 >     Range(  P, xsd:boolean )
 >     PropertyConstraint( C, P, MinCardinality( 2 ) )
 >     O rdf:type C
 > THEN
 >     O P "true"^^xsd:boolean.

I better add that to the test cases.

Jeremy






[1]
http://lists.w3.org/Archives/Public/www-webont-wg/2003Jan/0364.html
[2]
http://www.w3.org/TR/2003/WD-xmlschema-11-req-20030121/
Received on Wednesday, 22 January 2003 14:19:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:57 GMT