- From: Sandro Hawke <sandro@w3.org>
- Date: Fri, 24 May 2002 11:38:05 -0400
- To: "Thomas B. Passin" <tpassin@comcast.net>
- cc: www-rdf-interest@w3.org
> [Seaborne, Andy]
> >
> > I think of the minimal complete subgraph case is a sideways transformation
> > of one RDF graph into another. RDF -> RDF. [An complete solution to this
> > would also involve reformatting the RDF, creating new statements, not just
> > selecting them.]
> >
>
> GSLT, the analog of XSLT. Sounds like it would be good.
Sounds like a rule/implication to me, as in cwm [1] or PTL [2].
The interesting technical question in such structures is whether you
allow new nodes to be "created" in the new graph (allowing an
existential variable in the consequent). My guess is that doing so
gives you Turing completeness. There are also challenges in encoding
this kind of rule in RDF (if you want to do that) without running
afoul of RDF entailment; that's what I'm working on in PTL.
(I haven't announced PTL yet, because I want to clean it up some more,
but it seemed too relevant here to not mention.)
-- sandro
[1] http://www.w3.org/2000/10/swap/doc/cwm.html
[2] http://www.w3.org/2002/05/positive-triples/
Received on Friday, 24 May 2002 11:38:53 UTC