On 22/05/13 13:25, jo pol wrote:
Hello
Now that the fix for the nullpointer execption is released (thanks) I'm
arriving at the next hurdle:
I expected the owl reasoner to complain about "_:p <
http://purl.org/vocab/bio/0.1/mother> _:p" as the property is a subclass of
http://www.w3.org/2002/07/owl#differentFrom. But it does not. My code is
available on
https://code.google.com/p/gedcom2sem/source/browse/trunk/src/test/java/gedcom2sem/semweb/ValidationTest.java?r=302
Am I missing something?
Reasoner incompleteness.
By extending the OWL machinery itself you are into OWL/full territory
here. The jena rule reasoner is indeed an OWL/full reasoner but as noted
many times is not complete. In particular the layered design is biting
in this case.
There are work arounds. For example you could materialize the RDFS
closure of your ontology first. Thus finding all the places that you
have extended OWL itself. Then run the OWL reasoner over that closure.
Dave
