- From: Jos De_Roo <jos.deroo@agfa.com>
- Date: Fri, 14 Feb 2003 22:37:03 +0100
- To: Andy_Seaborne@hplb.hpl.hp.com
- Cc: "'Libby Miller'" <Libby.Miller@bristol.ac.uk>, "'www-rdf-rules@w3.org'" <www-rdf-rules@w3.org>, www-rdf-rules-request@w3.org
> > I would suggest sometime in the week of 24th February, maybe Thursday
> > 27th at 17:00 GMT.
>
> Great idea - I will try and make it.
great idea (but I'm then having webont telecon
but I will follow you offline...)
> > The main issue for me is how to express the query itself.
>
> Agreed. I would like to see the expressivity as poer the QL98 paper -
> arbitrary graph patterns - including ones that do not serialize
> conveniently.
sounds good ;-)
> > My inclination is to try and express queries as graphs with parts
> > missing, that is, to interpret an RDF graph as a query, as Jos de Roo
> > suggested [5].
>
> > I think this would give us one big win, which is that we could ignore
> > the syntactic differences between several very similar query languages
> > for RDF (although this will by no means encompass all RDF query
> > languages). Plus of course, we get the query parser for free.
>
> An alternative would be to express the query as cwm rules of the form:
>
> { query pattern } => { output format }
>
> (or some such) then there is a implementation of the query processor as
well
> :-)
querying given graph A with (graph B => graph C)
is proving that graph A => (graph B => graph C)
or proving that graph A and graph B => graph C
so we can just run the simple query C against a
(transient) merge of graph A and graph B
> In Jos's message:
> > ... it seems to us that a query is simply
> > a set of triples where bnodes play the role of variables
> > we then basically can apply a resolution process
>
> OK if blank arcs are allowed (also, with the _:a syntax to relate bNodes
to
> get arbitrary graph patterns).
right, you had that idea before
I just did a small test that
:a :b :c.
:d :c :f.
:g :c :i.
queried with
_:x :b _:u.
_:y _:u _:z.
indeed gives
:a :b :c.
:d :c :f.
:a :b :c.
:g :c :i.
with our actual entailment test code (euler)
# Proof found for file:/temp/testC.n3 in 3 steps
but to work it out in general we would need
some goal reordering mechanism...
-- ,
Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/
Received on Friday, 14 February 2003 16:38:53 UTC