Re: N-Triples literals

>>>"Karsten-A. Otto" said:
> Hello,
> 
> I noticed that the productions for literals in the N-Triples specification
> (RDF Test Cases, http://www.w3.org/TR/rdf-testcases/) has been changed to
> allow xmlString and langstring. This should have a capital S by the way.

Ah yeah, the link label text is wrong.  Thanks for the catch

> I applaud the addition of a language tag in the langString production,
> which in my opinion greatly increases the expressiveness of N-Triples.

I considered it crucial.  RDF M&S hinted at this but we are now
making it clearer where language appears in the RDF graph -
part of the literal.  We do need to improve this and will be making
further clarifications of the RDF graph.

> However, I do not quite understand what xmlString is supposed to be for.
> Is this for parseType=Literal cases? Please add some explanations for
> these new literal variants.

Yes, that is exactly what it is for - XML literal content, generated
by that form.  See the tests for rdfms-xmllang and
rdfms-literal-is-xml-structure for the examples of where it is used.
See them below: http://www.w3.org/2000/10/rdf-tests/rdfcore/

This stuff is/will be folded into the RDF primer and syntax docs in
due course.  See
http://www.w3.org/TR/rdf-syntax-grammar/#parseTypeLiteralPropertyElt
and follow the threads.

You can track where I've got to in the draft syntax wd at
http://ilrt.org/discovery/2001/07/rdf-syntax-grammar/

Dave