Wednesday, March 21, 2007

Success with OWL1.0

So, after the aforementioned
disappointment
, I now have a fairly ugly, but successful workaround to QCRs operating well in my Jena-Pellet set up. (Suggested by the supplier of this troublesome ontology). Basically, when you want to say "X has exactly 2 eyeballs" , this workaround has you say "X hasEyeballs exactly 2", forcing you to add an additional property for each range-related cardinality restriction you want to add. This is fine when you don't have very many potential properties, but does seem a bit awkward. I'm trying to think of a case where this wouldn't work, but am struggling - perhaps some one could suggest one?

Will try again with QCRs and new Pellet, but, give that OWL1.1 may or may not become a standard (thanks Richard), it seems wise to know and use the OWL1.0 constructs as long as it is possible.

0 comments: