- From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
- Date: Sun, 27 Jan 2002 18:24:00 -0000
- To: "Dan Connolly" <connolly@w3.org>
- Cc: "RDF core WG" <w3c-rdfcore-wg@w3.org>
> > Dan:
> > > Brace yourself for mind-bogglinly deep
> > > formal argument:
> > >
> > > premise:
> > > <http://www.w3.org/> dc:title "W3C".
> > > conclusion:
> > > <http://www.w3.org/> dc:title "W3C".
> > >
> > > That's it.*
OK. Definitely a bug with TDL model theory.
I can describe the approach I will work on, but only in English at this
stage. Mathematics to follow later.
Pat's model theory (both the current draft and the editor's draft) on
entailment amount to:
An RDF Graph A entails another RDF Graph B if, given a set of URIRefs from
which the uriref labels are taken, and any interpretation of these uriref
into some universe, then:
whenever there is
some mapping of the blank nodes of A into that universe such that the
triples of A are satifisfied,
then
there is also a mapping of the blank nodes of B into that universe such
that the triples of B are satisfied.
In TDL what we should have is:
An RDF Graph A entails another RDF Graph B if, given a set of URIRefs from
which the uriref labels are taken, and any interpretation of these uriref
into some universe, then:
whenever there is
some mapping of the blank nodes and literal nodes of A into that
universe such that the triples of A are satifisfied (including all
datatyping triples, understood as in the TDL document)
then
there is also a mapping of the blank nodes and literal nodes of B into
that universe such that the triples of B are satisfied.
Peter suggests (if I have understood him correctly) that what I in fact
wrote had dropped the existential quantification over the blank nodes as
well!
Sorry for the mistake.
I haven't done model theory in the last fifteen years!
Jeremy
Received on Sunday, 27 January 2002 13:24:02 UTC