RE: Annotation properties with range and other rdf:types

At 2:38 PM -0800 12/4/03, Holger Knublauch wrote:
>Folks,
>
>I still have not read why there are problems with the
>logic which prevent annotation properties from having
>a range in OWL DL, and I would appreciate if someone
>could at least give me a hint on why that's impossible.
>To me it looks rather like a syntactic problem.
>
>Anyway, given that the specification is not going to be
>changed soon, I will have to build a compromise solution
>for Protege:  At least the interface that I am now using
>to access Racer would not allow me to pass in an OWL Full
>ontology, so I need to find a workable solution for our
>users.  This solution basically will mean that most ontologies
>will be in OWL Full at edit and save time.  When the user
>presses the classify button, I will temporarily remove any
>AnnotationProperties or other OWL Full constructs on the
>fly before I send them to the classifier.
>
>Does anyone know of an open-source Java API that would
>take an OWL Full ontology as input and create an OWL DL
>ontology as output?
>
>Thanks
>Holger

Holger,
  This mailing list is really aimed at specific discussion of the OWL 
langauge design and documents, and your message asking about 
annotation was quite appropriate.  However, followup discussion on 
the reasons why DL is restricted and about other systems would make 
more sense on www-rdf-logic@w3.org.  This mailing list has a very 
limited readership, that one has a much larger one, and I suspect 
you'd be more likely to get answers there.   That's not to say askign 
here is wrong, just that you'd have a better shot at more answers 
there...
  -Jim H

-- 
Professor James Hendler			  http://www.cs.umd.edu/users/hendler
Director, Semantic Web and Agent Technologies	  301-405-2696
Maryland Information and Network Dynamics Lab.	  301-405-6707 (Fax)
Univ of Maryland, College Park, MD 20742	  240-277-3388 (Cell)

Received on Thursday, 4 December 2003 20:48:15 UTC