W3C home > Mailing lists > Public > public-owl-wg@w3.org > April 2008

RE: First crack at profiles in primer

From: Michael Schneider <schneid@fzi.de>
Date: Sun, 13 Apr 2008 18:20:30 +0200
Message-ID: <0EF30CAA69519C4CB91D01481AEA06A08BD601@judith.fzi.de>
To: "Ivan Herman" <ivan@w3.org>
Cc: "OWL Working Group WG" <public-owl-wg@w3.org>
Ivan Herman wrote:

>----- OWL-R -----
>
>It is *really* not the way of shooting you:-), but I do not understand 
>what you mean by "all the OWL Fullish features of RDFS";
>but you say 
>that yourself that it is not helpful... If you mean 
>reflection, well, at 
>the moment, OWL-R does not have that a.f.a.i.k. (that was the 
>essence of 
>the issue I had on the axiomatic triples)... 

It's easy to say in the profiles document something like:

  "This ruleset for OWL-R Full extends 
  the ruleset given in the RDFS spec"

The OWL 1 Full spec also doesn't restate all the RDFS semantic conditions, but refers to the RDFS spec instead. And the pD* paper also only enumerates the pD* specific rules in the section where pD* is introduced. So this approach has precedence.

>The only thing I see that 
>may be possible in OWL-R (well, OWL-R-Full, actually!) is the usage of 
>the same symbol for a data and object property. I may miss something...

Let's try an example, and start from the following RDF data:

  :alice :isMarriedWith :bob ;
         foaf:knows :charly ;
         foaf:knows "Peter"^^xsd:string ;
         dc:author blog:alicesWonderland .

This alone is already OWL-R Full valid data, and OWL-R will even produce a few basic entailments, for example:

  :alice a rdfs:Resource . 		# should also be 'owl:Thing'
  :isMarriedWith a rdf:Property .	# should also be 'owl:ObjectProperty'

But, of course, this is not yet very interesting information. So let's add a bit of semantics (just what comes to mind first, no deep analysis of the given data, let alone rigorous analysis of the data's application domain):

  :isMarriedWith rdf:subPropertyOf foaf:knows .
  :isMarriedWith a owl:SymmetricProperty . 

Now, OWL-R Full will provide us some more interesting entailments, namely:

  :bob :isMarriedWith :alice .
  :alice foaf:knows :bob .
  :bob foaf:knows :alice .

This is certainly what one expects here.

More interesting might be what *not* happens in this situation. There is no syntax error, although the property 'foaf:knows' is used with both an URI and a data literal as its object. We get the expected result, anyway, since datavalues are normal resources in OWL Full, which means that 'foaf:knows' is inferred to be an object property. 

For most of the stuff in the data there is no declaration of the category it belongs to (Individual, Class, Object/Data Property). But this doesn't matter, since in OWL Full /everything/ is a resource, and everything in predicate position is an object property. So we only have to declare, if we want to be more specific than this "default" typing information. For example that ':isMarriedWith' is actually a /symmetric/ property. 

We will, of course, only want to do such declaring for those parts of the data, for which we want to have reasoning results at all. In our example, we aren't /at the moment/ much interested in reasoning about the dc:authorShip of :alice, so we do not need to treat this data semantically now, but can add semantics later on demand.

In this way, we can now start to incrementally add arbitrary data, perhaps from different sources. Or arbitrarily extend our ontology, or apply ontologies that we find on the SemWeb, which cover parts (but not necessary all) of our data (perhaps a FOAF ontology, or a DC ontology). OWL-R Full will gracefully accept this all, and do inferencing on this data.

Bottom line: OWL-R allows me to work with RDF data and OWL in a pretty agile and ad-hoc way. This is of course not a distinguishing aspect in a purly /technical/ sense. But it is IMHO a /pragmatical/ advantage for certain development scenarios.

Cheers,
Michael

--
Dipl.-Inform. Michael Schneider
FZI Forschungszentrum Informatik Karlsruhe
Abtl. Information Process Engineering (IPE)
Tel  : +49-721-9654-726
Fax  : +49-721-9654-727
Email: Michael.Schneider@fzi.de
Web  : http://www.fzi.de/ipe/eng/mitarbeiter.php?id=555

FZI Forschungszentrum Informatik an der Universität Karlsruhe
Haid-und-Neu-Str. 10-14, D-76131 Karlsruhe
Tel.: +49-721-9654-0, Fax: +49-721-9654-959
Stiftung des bürgerlichen Rechts
Az: 14-0563.1 Regierungspräsidium Karlsruhe
Vorstand: Rüdiger Dillmann, Michael Flor, Jivka Ovtcharova, Rudi Studer
Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus


Received on Sunday, 13 April 2008 16:21:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 13 April 2008 16:21:12 GMT