- From: Dan Connolly <connolly@w3.org>
- Date: Tue, 31 Jul 2001 14:58:48 -0500
- To: jos.deroo.jd@belgium.agfa.com
- CC: geoff@sover.net, www-rdf-calendar@w3.org
jos.deroo.jd@belgium.agfa.com wrote:
>
[...]
> that's indeed a problem with empty descriptions a la []
> so I hacked your poolGame-kb.n3 line 610 into
> } log:implies {<?who> = [ pa:name ""
> (just to work around around for the moment)
OK... I can work around that issue too...
But I have run into an issue that looks more substantial: Euler.java
doesn't seem to grok variables in the property part of a triple,
which seems critical for dealing with rdfs:subPropertyOf.
I reduced it to a small test case:
------
http://www.w3.org/2000/10/swap/test/subprop.n3
# testing rdfs rules...
@prefix : <subprop#>.
@prefix u: <http://www.w3.org/2000/01/rdf-schema#> .
:husband u:subPropertyOf :spouse.
:Mary :husband :Dan.
------
------
http://www.w3.org/2001/05ve/rdfs.n3
# generalization...
{ :s [ u:subPropertyOf :p ] :o } log:implies { :s :p :o }.
------
---------
http://www.w3.org/2000/10/swap/test/subprop-query.n3
@prefix log: <http://www.w3.org/2000/10/swap/log#> .
@prefix : <subprop#>.
@prefix v: <subprop?>.
{ v:x :spouse v:y } log:forSome v:x, v:y.
---------
but Euler doesn't find a proof:
$ CLASSPATH=/home/connolly/src/euler java Euler
../../../../2001/05ve/rdfs.n3 subprop.n3 subprop-query.n3
# Generated with Euler 27.015 on Tue Jul 31 14:57:35 CDT 2001
# for query
file:/home/connolly/w3ccvs/WWW/2000/10/swap/test/subprop-query.n3
# given
[file:/home/connolly/w3ccvs/WWW/2000/10/swap/test/../../../../2001/05ve/rdfs.n3,
file:/home/connolly/w3ccvs/WWW/2000/10/swap/test/subprop.n3]
# No proof found for
file:/home/connolly/w3ccvs/WWW/2000/10/swap/test/subprop-query.n3 in 1
steps (999 steps/sec)
--
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Tuesday, 31 July 2001 15:58:49 UTC