W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > October 2001

Re: 2001-10-12#2 - review rdfs-domain-and-range TCs

From: <jos.deroo.jd@belgium.agfa.com>
Date: Fri, 19 Oct 2001 10:11:22 +0100
To: barstow@w3.org
Cc: w3c-rdfcore-wg@w3.org
Message-Id: <OF90753959.064622B0-ON41256AEA.0031561B@bayer-ag.com>

[some more bits of explanation]





jos.deroo.jd@belgium.agfa.com@w3.org on 2001-10-17 11:41:10 AM

Sent by:  w3c-rdfcore-wg-request@w3.org


To:   barstow@w3.org
cc:   w3c-rdfcore-wg@w3.org
Subject:  Re: 2001-10-12#2 - review rdfs-domain-and-range TCs



>> WRT 2001-10-12#2 - review the test cases in:
>>
>>  http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfs-domain-and-range/
>>
>> for issue:
>>
>>  http://www.w3.org/2000/03/rdf-tracking/#rdfs-domain-and-range
>>
>> 1. test001.rdf - since RDFS:
>>
>>   http://www.w3.org/TR/2000/CR-rdf-schema-20000327/
>>
>>  already allows a property to have more than one domain, it's not
>>  clear if this TC is designed to verify that or is intended to reflect
>>  a WG decision.  In either case the purpose of the test case should
>>  be included in the file.  See attached test001.rdf
>
>fine, I've updated that one, however without the 2 extra triples
>to "define some Classes"
>(see point 3.)
>
>> 2. test002.rdf - looks OK.  I added a Description and defined
>>  some Classes for the values of the range properties.  See attached
>>  test002.rdf
>
>idem
>
>> 3. I don't understand TC's test003.rdf and test004.rdf are supposed
>>  to test so I don't approve them.
>
>Art, the idea was that
>  http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfs-domain-and-range/test001.nt
>and
>  http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfs-domain-and-range/test002.nt
>and
>  http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfs-domain-and-range/test003.nt
>rdfs-entails
>  http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfs-domain-and-range/test004.nt
>
>and some evidence for that can be found at
>  http://lists.w3.org/Archives/Public/www-archive/2001Oct/att-0016/01-etc.n3

the issue resolution is
-> Multiple domain and range constraints are permissable
   and will have conjunctive semantics.

so suppose we have
  ex:baz1 ex:bar ex:baz2 .        #test003
and
  ex:bar rdfs:domain ex:Domain1 . #test001
and
  ex:bar rdfs:domain ex:Domain2 . #test001

then we can rdfs_entail
  ex:baz1  rdf:type ex:Domain1 .  #test004
and
  ex:baz1  rdf:type ex:Domain2 .  #test004

now, it is that last 'and' that expresses
those 'conjunctive semantics' i.e.
ex:baz1 in in the intersection of ex:Domain1 and ex:Domain2

the same story for rdfs:range

>and as you can see there is no need to "define some Classes" because
>that information can be RDFS entailed
>
>> 4. axioms.n3 - I propose that NO N3 files get approved at this point.
>
>OK :-|

--
Jos

PS the $Id$ in the .rdf files doesn't seem to be evaluated (in the .nt it's ok)
Received on Friday, 19 October 2001 04:11:38 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:41:06 EDT