RE: deciding on rdf:PlainLiteral this week



> -----Original Message-----
> From: Peter F.Patel-Schneider [mailto:pfps@research.bell-labs.com]
> Sent: 1 June 2009 17:09
> To: Seaborne, Andy
> Cc: lee@thefigtrees.net; sandro@w3.org; team-sparql-chairs@w3.org; team-
> rif-chairs@w3.org; team-owl-chairs@w3.org; public-rdf-text@w3.org
> Subject: Re: deciding on rdf:PlainLiteral this week
> 
> From: "Seaborne, Andy" <andy.seaborne@hp.com>
> Subject: RE: deciding on rdf:PlainLiteral this week
> Date: Mon, 1 Jun 2009 09:05:03 -0500
> 
> > I am looking for explicit text that covers the use of SPARQL extended
> > matching [3] but the draft does not include any text that refers to,
> > only to SPARQL syntax and SPARQL results.
> 
> Umm.  What more do you want?

The text you proposed to cover bindings was enough.  It has gone and I don't see by text that covers it now.

What text in the draft do you see as already covering this? because I don't see any now.

> 
> > [1] option 2 would cover it.  Section 4 is a tolerable place to put it
> > tough not ideal.
> >
> > The response in [2] did mention  "queries, bindings, and results"
> > which is not very precise but at least covers the right area.
> > However, I now can't find any mention of that text in the current
> > .../PlainLiteral draft.
> >
> > Section 4 says
> > "in syntaxes for RDF graphs and for SPARQL"
> > "do not occur in syntaxes for RDF graphs, nor in syntaxes for SPARQL."
> >
> > This is SPARQL syntax (the query string), not the matching (the output
> > of the matching step).  Because rdf:PlainLiteral covers a non-datatype
> > aspect, it need to be clear that bindings from BGP extended matching
> > MUST be in existing forms.
> 
> What forms?  In a *very* strong sense, rdf:PlainLiteral typed literals
> are already in existing forms.

As this is being placed under the section "Syntax for rdf:PlainLiteral Literals", and while this is bending that title a bit, I was suggesting that the interaction with literal accessors was viewed as syntax.  But the doc talks about "results" and query syntax and that does not cover the algebra and FILTERs.  "results" applied to the overall results, not the interaction of layers.

> 
> > This also does not cover it:
> > "in existing syntaxes for RDF graphs and SPARQL results."
> >
> > We are not talking just about SPARQL XML results format (which should
> > be covered) but about how the extended matching fits into existing
> > implementations and how bindings flow from one BGP matching to another
> > in the same query, possibly where the BGP matching are under different
> > entailment regimes.  Applying to the extended matching would
> > automatically include SPARQL XML results although it is good to call
> > those out anyway, as the draft does (may be written by a non-SPARQL
> > engine).
> 
> Are you saying that you want to change the internals of SPARQL?

Exactly the opposite.  SPARQL defines an extension framework - the definers of the extensions are then responsible for the exact details of each extension.

 Andy

> 
> >  Andy
> >
> >
> > [1] http://lists.w3.org/Archives/Public/public-rdf-

> text/2009AprJun/0229.html
> > [2] http://lists.w3.org/Archives/Public/public-rdf-

> text/2009AprJun/0230.html
> > [3] http://www.w3.org/TR/rdf-sparql-query/#sparqlBGPExtend

> 
> peter

Received on Monday, 1 June 2009 16:20:42 UTC