Re: OWL Test Results page, built from RDF

From: Sandro Hawke <sandro@w3.org>
Subject: Re: OWL Test Results page, built from RDF 
Date: Mon, 22 Sep 2003 13:52:18 -0400

> > > > I am now getting a KeyError from the script when run on 
> > > > 
> > > > http://www-db.research.bell-labs.com/user/pfps/owl/RESULTS/rdf
> > > 
> > > I wasn't properly handling the case where the tested system has no
> > > name declared.  That should be fixed now.  (There are called anon_1,
> > > etc)
> > > 
> > > But I think the real problem is that your data says the system is the
> > > string "#owlp" instead of something with the URI "#owlp".  That is,
> > > you have 
> > >   <rtest:PassingRun rtest:system="#owlp">
> > >   ...
> > > instead of
> > >   <rtest:PassingRun>
> > >     <rtest:system rdf:resource="#owlp"/>
> > >   ...
> > 
> > I will change this, but shouldn't it work as is?
> 
> You shouldn't have gotten the KeyError, and I've changed the software
> so I don't think you will, but in the first version your system will
> be called "anon_1" and not have a description.
> 
> > > Also, it looks like the data is claiming passes of some tests when it
> > > should be claiming passes of rtest:syntacticLevelTestFrom those
> > > tests.   Something more like:
> > > 
> > >   <rtest:PassingRun>
> > >     <rtest:test rdf:parseType="Resource">
> > >       <rtest:syntacticLevelTestFrom rdf:resource="http://www.w3.org/2002/03
> > owlt/I5.2/Manifest002#test"/>
> > >     </rtest:test>
> > >     <rtest:system rdf:resource="#owlp"/>
> > >   </rtest:PassingRun>
> > 
> > I don't see syntacticLevelTestFrom in this ontology.  Shouldn't it be added?
> 
> We talked about this at the end of
> 
>    http://lists.w3.org/Archives/Public/www-webont-wg/2003Sep/0143.html
> 
> and you said it might also apply to RDF tests, ... and I don't really
> know anything about RDF tests, so ...  I haven't done anything yet.
> So my lame-o plan is for us to use an undocument
> rtest:syntacticLevelTestFrom predicate until someone gets motivated to
> figure out something better.
> 
>       -- sandro

I do not believe that there is any technical reason not to include
information about rtest:syntacticLevelTestFrom in the OWL test ontology.  I
believe that this is a much better solution than one using an undocumented
property.

peter

Received on Monday, 22 September 2003 13:56:33 UTC