W3C home > Mailing lists > Public > www-rdf-comments@w3.org > July to September 2003

Re: RDF test case on datatypes

From: pat hayes <phayes@ihmc.us>
Date: Thu, 28 Aug 2003 09:24:28 -0700
Message-Id: <p06001a07bb73d9bcc634@[10.0.1.4]>
To: Jan Grant <Jan.Grant@bristol.ac.uk>, Olivier Corby <Olivier.Corby@sophia.inria.fr>
Cc: www-rdf-comments@w3.org

>On Thu, 28 Aug 2003, Olivier Corby wrote:
>
>>
>>  In the RDF Test Cases
>>  http://www.w3.org/2001/sw/RDFCore/TR/WD-rdf-testcases-20030117/
>>  the test case RDF + RDFS + DT(xsd:decimal, xsd:string)
>>  states that the following triples lead to a FALSE conclusion :
>>
>>  ex:foo ex:bar "25"^^xsd:decimal
>>
>>  ex:bar rdfs:range xsd:string
>>
>>
>>
>>  Suppose the following triples, where there is now a resource ex:gee of
>>  rdf:type xsd:decimal, do they also lead to a FALSE conclusion ?
>>
>>
>>  ex:foo ex:bar ex:gee
>>
>>  ex:gee rdf:type xsd:decimal
>>
>>  ex:bar rdfs:range xsd:string
>
>That's an interesting case, and for an accurate answer I'll defer to the
>Semantics document editor... Pat?

Thanks, Jan; Hi, Olivier.

The answer is NO, but a full explanation requires some subtlety.  The 
actual conclusion here is that ex:gee has types both xsd:string and 
xsd:decimal. Now, that is impossible, of course, given the XSD spec, 
and an application which was sufficiently XSD-savvy might be able to 
recognize the implicit contradiction: but in fact it is not formally 
a contradiction in RDF itself, since RDF entailments about rdf:type 
are not required to be sensitive to information from external sources.

In the case of a typed literal, however, the datatyping conditions 
may themselves access such external information to check typed 
literals. If the interpretation is XSD-sensitive in this way then the 
implicit contradiction will be made explicit by the datatype check. 
Hence the test case: note that it refers explicitly to the datatype 
mapping which includes the datatypes. As the semantics document 
states, these are the only contradictions recognized by RDF: without 
datatype checks, any RDF graph is consistent.

Hope this helps.

Pat

>--
>jan grant, ILRT, University of Bristol. http://www.ilrt.bris.ac.uk/
>Tel +44(0)117 9287088 Fax +44 (0)117 9287112 http://ioctl.org/jan/
>(ECHOY GRUNTING) (EERIE WHISPERS) aren't subtitles great?


-- 
---------------------------------------------------------------------
IHMC	(850)434 8903 or (650)494 3973   home
40 South Alcaniz St.	(850)202 4416   office
Pensacola			(850)202 4440   fax
FL 32501			(850)291 0667    cell
phayes@ihmc.us       http://www.ihmc.us/users/phayes
Received on Thursday, 28 August 2003 12:24:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 21 September 2012 14:16:32 GMT