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

Re: testcase for the option for collections

From: Dan Connolly <connolly@w3.org>
Date: 29 May 2002 14:16:04 -0500
To: Jos De_Roo <jos.deroo.jd@belgium.agfa.com>
Cc: w3c-rdfcore-wg <w3c-rdfcore-wg@w3.org>
Message-Id: <1022699765.21951.159.camel@dirk>

On Fri, 2002-05-24 at 11:17, Jos De_Roo wrote:
> Dan,
> 
> In the RDFCore telecon d.d. 2002-05-24 agenda item 11
> 
> [[[
> 11: daml:collection
> ACTION:2002-05-03#7  danc  summarise to the list the options for
>        collections, with the objections that've been raised
>        (or likelihood that "they'll fly")
> 
> 
> See:
>   http://www.w3.org/2000/03/rdf-tracking/#rdfms-seq-representation
>   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Apr/0334.html
> ]]]
> 
> I took the action to create a testcase for this issue.

This pretty much works for me (I have a different
preferences w.r.t. one of your points of attention below...).
i.e. I'd like to use this
to dispatch

ACTION 2002-05-24#2 (danc) to write up a proposal for incorporating
        daml:collection into RDF

as well.

Umm... syntax editors, does this tell you what you would need
in order to make the edits?

> The proposed testcase is that parsing
> 
> ---- test001.rdf
> <rdf:RDF
>     xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>     xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
>     xmlns:eg="http://example.org/eg#">
> 
>     <rdf:Description rdf:about="http://example.org/eg#eric">
>         <rdf:type rdf:parseType="Resource">
>             <eg:intersectionOf rdf:parseType="collection">
>                 <rdf:Description rdf:about="http://example.org/eg#Person"/>
>                 <rdf:Description rdf:about="http://example.org/eg#Male"/>
>             </eg:intersectionOf>
>         </rdf:type>
>     </rdf:Description>
> </rdf:RDF>
> ----
> 
> would generate following triples
> 
> ---- test001.nt
> <http://example.org/eg#eric>
> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> _:a0 .
> _:a0 <http://example.org/eg#intersectionOf> _:a1 .
> _:a1 <http://www.w3.org/1999/02/22-rdf-syntax-ns#type>
> <http://www.w3.org/2000/01/rdf-schema#List> .
> _:a1 <http://www.w3.org/2000/01/rdf-schema#first>
> <http://example.org/eg#Person> .
> _:a1 <http://www.w3.org/2000/01/rdf-schema#rest> _:a2 .
> _:a2 <http://www.w3.org/1999/02/22-rdf-syntax-ns#type>
> <http://www.w3.org/2000/01/rdf-schema#List> .
> _:a2 <http://www.w3.org/2000/01/rdf-schema#first>
> <http://example.org/eg#Male> .
> _:a2 <http://www.w3.org/2000/01/rdf-schema#rest>
> <http://www.w3.org/2000/01/rdf-schema#nil> .
> ----
> 
> Some points of attention are:
> 1/ using namespace xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"

That seems to introduce a dependency from RDF syntax to RDF schema.
I'd rather not do that.

A new namespace would be the cleanest option, but I think on
balance, if we're going to change the syntax, we might as
well use this namespace for first/rest/nil/List.

  http://www.w3.org/1999/02/22-rdf-syntax-ns#

I'm also uneasy about leaving the semantics of first/rest
'incomplete' at the RDF level; i.e. allowing layered
specs to constrain their extents further. I think I can live with
it, but I'd feel better if I had a more clear picture
of the rdfms-assertion and MIME type issues.

> 2/ using attribute rdf:parseType="collection"

works for me.

> 3/ just use rdf:Description for collected items

of course, you can use any typednode, right?

> 
> That's all.
> 
> --
> Jos
> 
-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Wednesday, 29 May 2002 15:16:24 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:48:16 EDT