W3C home > Mailing lists > Public > public-owl-wg@w3.org > November 2007

RE: ISSUE-54 (Entities as Triples): Translation to Triples missing for entities in RDF mapping

From: Boris Motik <boris.motik@comlab.ox.ac.uk>
Date: Wed, 21 Nov 2007 14:57:04 -0000
To: "'Rinke Hoekstra'" <hoekstra@uva.nl>
Cc: "'OWL Working Group WG'" <public-owl-wg@w3.org>
Message-ID: <000201c82c4e$c766bf50$2711a8c0@wolf>

Hello,

OK, now I understand what this comment is about.

Still, I am not sure that we need this translation in the table. The production for "entity" from the FS document is used only in
EntityAnnotation and Declaration. In fact, the reason why we have the "entity" production is in order to avoid having
OWLClassAnnotation, ObjectPropertyAnnotation, etc.

Now Table 2 in the document specifying the translation from OWL to RDF never really translates the "entity" production. Instead, it
contains separate rows for each of

EntityAnnotation(OWLClass(cID) ...)
EntityAnnotation(ObjectProperty(opID) ...)

and so on. For each of these axioms, the operator T is then applied to cID, opID, and so on (now after I've fixed the earlier bug);
hence, we indeed get the correct typing triple in the resulting RDF graph.

A similar approach is adopted in declarations. Thus, we never really get to translate the "entity" production, so this is why it is
missing from the table.

Considering all this confusion, it might be better to modify the FS document and, instead of the "entity" production, to introduce
OWLClassAnnotation, ObjectPropertyAnnotation, and so on. It seems to me that this would make the spec more consistent.

Regards,

	Boris


> -----Original Message-----
> From: Rinke Hoekstra [mailto:hoekstra@uva.nl]
> Sent: 21 November 2007 14:42
> To: Boris Motik
> Cc: 'OWL Working Group WG'
> Subject: Re: ISSUE-54 (Entities as Triples): Translation to Triples missing for entities in RDF
> mapping
> 
> Hi Boris,
> 
> I think your solution solves a different issue than the one Peter and
> I raised. The FS document specifies the language entities:
> 
> entity := datatype | owlClass | objectProperty | dataProperty |
> individual
> datatype := 'Datatype' '(' datatypeURI ')'
> owlClass := 'OWLClass' '(' owlClassURI ')'
> objectProperty := 'ObjectProperty' '(' objectPropertyURI ')'
> dataProperty := 'DataProperty' '(' dataPropertyURI ')'
> individual := 'Individual' '(' individualURI ')'
> 
> But the RDF mapping only specifies a translation for their URI's, and
> not for the entities themselves.
> 
> As (I think) Peter pointed out, the mapping of the entities themselves
> is probably the same as the mapping of their URI's. If this is the
> case, then these trivial mappings should be added to the table as well.
> 
> Does this make sense?
> 
> Best,
> 
> 	Rinke
> 
> 
> On 21 nov 2007, at 15:25, Boris Motik wrote:
> 
> >
> > Hello,
> >
> > I've just looked the RDF mapping, and now I'm not sure whether I
> > understand exactly what this comment is about.
> >
> > At the top of Table 2, one can find a translation of various types
> > of URIs into RDF, and this translation does add the appropriate
> > typing triple to the RDF graph.
> >
> > In the last version of the RDF mapping, there was an error in the
> > translation of annotations. For example,
> >
> > EntityAnnotation(Datatype(dID)
> >    Annotation(apID1 ct1) ... Annotation(apIDn ctn)
> > )
> >
> > was translated into triples
> >
> > dID T(apIDi) T(cti)     for 1 <= i <= n.
> >
> >
> > The error was in the fact that I wrote dID instead of T(dID). I've
> > just changed the translation, and the above translation is now
> > changed into
> >
> > T(dID) T(apIDi) T(cti)     for 1 <= i <= n.
> >
> > Now, T(dID) would actually add the appropriate typing triple/
> >
> >
> > Have I now correctly addressed this comment, i.e., is there
> > something else that I've overlooked and that should be changed here?
> > If
> > I've correctly interpreted this problem and everyone is happy with
> > the solution, I'd like to close this issue.
> >
> > Regards,
> >
> > 	Boris
> >
> >
> >> -----Original Message-----
> >> From: public-owl-wg-request@w3.org [mailto:public-owl-wg-request@w3.org
> >> ] On Behalf Of OWL Working
> >> Group Issue Tracker
> >> Sent: 07 November 2007 14:51
> >> To: public-owl-wg@w3.org
> >> Subject: ISSUE-54 (Entities as Triples): Translation to Triples
> >> missing for entities in RDF mapping
> >>
> >>
> >>
> >> ISSUE-54 (Entities as Triples): Translation to Triples missing for
> >> entities in RDF mapping
> >>
> >> http://www.w3.org/2007/OWL/tracker/issues/
> >>
> >> Raised by: Rinke Hoekstra
> >> On product:
> >>
> >> The RDF mapping document does not provide a mapping for entities
> >> (e.g. OWLClass(X), Datatype(Y) etc.)
> >> in Table 2. These mappings should be added.
> >>
> >>
> >>
> >>
> >
> >
> 
> -----------------------------------------------
> Drs. Rinke Hoekstra
> 
> Email: hoekstra@uva.nl    Skype:  rinkehoekstra
> Phone: +31-20-5253499     Fax:   +31-20-5253495
> Web:   http://www.leibnizcenter.org/users/rinke
> 
> Leibniz Center for Law,          Faculty of Law
> University of Amsterdam,            PO Box 1030
> 1000 BA  Amsterdam,             The Netherlands
> -----------------------------------------------
> 
> 
Received on Wednesday, 21 November 2007 14:58:06 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:13:27 GMT