W3C home > Mailing lists > Public > public-esw-thes@w3.org > October 2006

RE: Protege and SKOS Core TODO

From: George Anadiotis <George.Anadiotis@cwi.nl>
Date: Thu, 19 Oct 2006 10:26:25 +0200 (CEST)
Message-ID: <49643.192.16.196.191.1161246385.squirrel@webmail.cwi.nl>
To: public-esw-thes@w3.org

Hello everyone

I am trying to develop an OWL ontology that is based on SKOS, so i run
into the known issue of importing SKOS in Protege (in my case, v.3.2b and
OWL plugin v2.2 on Fedora Core Linux). I found that issue also mentioned
in :

1.http://lists.w3.org/Archives/Public/public-esw-thes/2005Aug/0033  
2.http://www.w3.org/2004/02/skos/core/proposals#owlImport-7
3.http://article.gmane.org/gmane.comp.misc.ontology.protege.general/12066/match=skos

The conditions in my case are identical to 3., except for the Protege
version. The proposed solution seems to  be to 'create a new resource
<http://www.w3.org/2004/02/skos/core-owl-dl> which contains a pure OWL DL
description of the SKOS Core Vocabulary'.

However, the OWL DL version i found in
http://isegserv.itd.rl.ac.uk/cvs-public/~checkout~/skos/drafts/owlImport-7/core-owl-dl.rdf
, although working perfectly with Protege 3.1.1, unfortunately does not
work in Protege 3.2! It gives similar symptoms to the initial issue -
inability to display properties and the following exception:

WARNING: java.lang.ClassCastException:
edu.stanford.smi.protegex.owl.model.impl.DefaultRDFProperty
        at
edu.stanford.smi.protegex.owl.ui.properties.OWLObjectPropertySubpropertyRoot.getTopLevelObjectProperties(Unknown
Source)

After encountering this, i decided to check the OWL DL version for OWL
Consistency before posting to the protege-discussion list as well. I used
both the Mindswap (http://www.mindswap.org/2003/pellet/demo.shtml) and the
Wonderweb (http://phoebus.cs.man.ac.uk:9999/OWL/Validator) validators for
this purpose.

As it turns out, the proposed OWL DL version of SKOS is not OWL DL
compliant after all: both validators complain about Untyped Classes -
rdf:Resource, rdf:List and foaf:Document. Wonderweb also complains about
redifining rdf:List (although i admit i missed that).

So, it would seem like the answer to the question posed in
http://www.w3.org/2004/02/skos/core/proposals#owlImport-7 - 'Problem with
use of rdfs:Class also? Need classes declared as type owl:Class?' is
'yes'?

But, if you add owl:Class definitions for the above classes (as hinted by
the Mindswap validator), then Mindswap gives the thumbs, while Wonderweb
still complains about redifining rdf:List - rightly so, in this case.

The only other reference i could find on the list for this issue was:
4. http://lists.w3.org/Archives/Public/public-esw-thes/2006May/0005.html

However, i think there is no mentioning of these issues there. Any ideas?

Regards

--
George Anadiotis
PhD Researcher
CWI (Centre for Mathematics and Computer Science),
Kruislaan 413, 1098 SJ Amsterdam, The Netherlands
e-mail: george.anadiotis@cwi.nl
Tel: +31 (0)20 - 592 4080
Fax: +31 (0)20 - 592 4312
Web: http://www.cwi.nl/ins2/
Received on Thursday, 19 October 2006 08:26:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:38:55 GMT