W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2003

Re: Closure Rule rdf2 generates infinite set of triples

From: pat hayes <phayes@ihmc.us>
Date: Wed, 11 Jun 2003 16:10:06 -0500
Message-Id: <p05210613bb0d4accd669@[10.0.100.24]>
To: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Cc: w3c-rdfcore-wg@w3.org

>Brian McBride wrote:
>
>>On Tue, 2003-06-10 at 14:59, pat hayes wrote:
>>
>>>>I noticed that closure rule rdf2 in
>>>>
>>>>  http://www.w3.org/2001/sw/RDFCore/TR/WD-rdf-mt-20030117/#rdf_entail
>>>>
>>>>will generate an infinite number of triples from any triple with an xml
>>>>literal as its object.
>>>>
>>>Is that really the case? I still have not got a clear picture of 
>>>what the XMLliteral value is supposed to be or how it relates to 
>>>the lexical form, but my understanding was that the lex2value 
>>>mapping was a pretty simple normalization process, so that the set 
>>>of lexical forms which map to the same 'value' would be reasonably 
>>>small.
>>>
>>
>>Jeremy is the expert.  I was thinking there may be places where I can
>>add unbounded amounts of whitespace, e.g. spaces between attribute
>>values.
>>
>>Brian
>>
>
>
>In the lexical space and the value space XMLLiterals are simply 
>(isomorhpic to?) strings.
>
>i.e. identity and equality are the same.
>
>However, in RDF/XML documents Brian is correct, you can have 
>arbitrarty whitespace in certain places - but it gets squashed 
>before you get to the lexical space.

The best thing in any case is to phrase the rule in terms of 
replacing XML literal strings by their canonical versions. This kills 
even the dead birds with one stone and is fine for checking closures. 
New editor's draft will have this modification.

Pat


-- 
---------------------------------------------------------------------
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 Wednesday, 11 June 2003 17:10:10 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:57:55 EDT