Re: raised in comment: owl:class still needed? Does this effect Test LC?

Peter:
> Well, in a certain sense none of owl:Class, owl:DatatypeProperty,
> owl:ObjectProperty, and probably quite a few other bits of OWL vocabulary
> are not *needed*.  However, it is *desirable* to have them around.

My reading of this thread is that there are no test cases that demonstrate the 
distinction except for ones in OWL Full which we could call either way (i.e. 
they demonstrate that owl:Class and rdfs:Class are not the same thing not why 
they need to be different).

Now we have to drill down on "*desirable*".

Peter does not give an account of this important word.

I would argue that:
  owl:DatatypeProperty is desirable because it is an important class of 
properties (from the DL perspective) that is not defined in RDFS.
  While owl:ObjectProperty could be identifed with rdf:Property without loss 
as far as I can see the current symmetry between owl:DatatypeProeprty and 
owl:ObjectProperty is *desirable* for user understanding, and having 
owl:DatatypeProperty and rdf:Property instead would be more confusing.
 Personally I argued against some of the other OWL vocabulary, 
owl:AnnoatationProperty being an example; but suggest that it is only those 
two (owl:Class and owl:ObjectProperty) where there is a strong overlap 
between OWL concepts and predefined RDF concepts. We did not get a comment 
asking us to look at owl:ObjectProperty and I believed I heard many members 
of the group say that it was *desirable* not to have the distinction between 
owl:Class and rdfs:Class.

  Hence I see it as down to those who oppose making a change here to 
articulate either:

1. Why this distinction is *necessary*
or
2. Why the straw poll judgement that the distinction is not *desirable* was 
incorrect.

Successfully completing (1) should convince the group; working on aesthetics 
(2) inevitably takes longer.

Jeremy

Received on Monday, 26 May 2003 05:48:28 UTC