- From: Jos de Bruijn <jos.debruijn@gmail.com>
- Date: Thu, 04 Mar 2010 08:27:35 +0100
- To: Sandro Hawke <sandro@w3.org>
- CC: RIF <public-rif-wg@w3.org>
SWC doesn't map plain literals to anything. According to the RDF semantics, plain literals without language tags are always mapped to themselves, i.e., strings of unicode characters. Now, XSD specifies that xs:string literals are also mapped to themselves, and thus they correspond 1-to-1 to plain literals w/o language tags. Now, it happens to and be the case that the value space of rdf:PlainLiteral also includes all strings. Therefore, there is a one-to-one correspondence between RDF plain literals and xs:strings of the form "xyz", on the one hand, and rdf:PlainLiterals of the form "xyz@", on the other. Best, Jos On 2010-03-02 20:06, Sandro Hawke wrote: > During the telecon today we looked at > > http://www.w3.org/2005/rules/wiki/RDF_Combination_Constant_Equivalence_2 > > and it seems that while that example is supported by the SWC, SWC isn't > saying quite the right thing, here. It says plain literals without > language tags map to xs:string constants, but I think it would be better > to map to rdf:PlainLiteral constants. The difference is actually > invisible to any entailment test (I think? maybe it depends on the > entailment regime?), so in a sense RIF doesn't care, but for interchange > purposes is does matter. In particular, SPARQL, when not doing > entailment, will notice the difference. > > Do you remember why it's xs:string now? > > This isn't a huge problem, but if there's no compelling reason not to > change it, I think it's more correct to map to rdf:PlainLiterals. > > -- Sandro -- Jos de Bruijn Web: http://www.debruijn.net/ LinkedIn: http://it.linkedin.com/in/josdebruijn Skype: josdebruijn Google Talk: jos.debruijn@gmail.com Mobile phone: +43 660 313 5733
Received on Thursday, 4 March 2010 07:28:15 UTC