RE: addressing requirements around daml:collection (rdfms-seq-representation)

Can't we split responsibility here.

RDF Core provides:
- rdf:parseType="collection"
- rdf:List, rdf:first, rdf:rest, rdf:nil

WOWG 
- provides functionality (i.e. uniqueness)


Consider


<rdf:RDF>
  <rdf:Description rdf:about="#foo" rdf:parseType="collection">
  </rdf:Description>  
  <rdf:Description rdf:about="#foo" rdf:parseType="collection">
    <rdf:Description rdf:about="#bar"/>
  </rdf:Description>
</rdf:RDF>

#foo is a non-standard collection with lengths of both 0 and 1.

This could be legal RDF and ill-formed at the ontology level.

Jeremy
  
 

> -----Original Message-----
> From: w3c-rdfcore-wg-request@w3.org
> [mailto:w3c-rdfcore-wg-request@w3.org]On Behalf Of Dan Connolly
> Sent: 24 April 2002 04:39
> To: Pat Hayes
> Cc: w3c-rdfcore-wg@w3.org
> Subject: Re: addressing requirements around daml:collection
> (rdfms-seq-representation)
> 
> 
> On Tue, 2002-04-23 at 17:41, Pat Hayes wrote:
> [...]
> > >By 'ala daml:collection' I meant: including
> > >the fact that first/rest are functional.
> > >
> > 
> > Oh, I see.  But then that is a much bigger change/extension to RDF 
> > than it seems, since now RDF has a way to encode functional 
> > properties.
> 
> It's a big change, yes. I'm not sure I'm comfortable with it.
> 
> The longhand-OK option is acceptable to me, but I dunno
> if it'll fly in the WebOnt WG and community.
> 
> The other options, i.e. techniques for enhancing
> the <rdf:li> style collections to express closed
> lists (a) have to be at least as powerful as first/rest
> in order to get the job done (I think?), and (b) look uglier.
> 
> > That goes beyond just adding a different kind of 
> > container. I would like to explore what the other implications of 
> > that might be.
> 
> -- 
> Dan Connolly, W3C http://www.w3.org/People/Connolly/
> 
> 
>

Received on Wednesday, 24 April 2002 04:44:01 UTC