Re: description-logic/consistency605

Charles:
> Note the two instances of /2002/07/owl#FunctionalProperty - is that correct? 
> I don't think so.
> I would be surprised if anyones parser would pass this.

 </rdfs:subPropertyOf>
  <rdf:type rdf:resource="/2002/07/owl#FunctionalProperty"/>
 </owl:ObjectProperty>
 <owl:ObjectProperty rdf:about="http://oiled.man.example.net/test#rxa"/>
 <owl:ObjectProperty rdf:about="http://oiled.man.example.net/test#rx">
  <rdf:type rdf:resource="/2002/07/owl#FunctionalProperty"/>
 </owl:ObjectProperty>

> Can someone in the know check this out?


Technically these are fine.

The form /2002/07/owl#FunctionalProperty is a relative URI which resolves 
against the base URI of 
http://www.w3.org/2002/03owlt/description-logic/consistent605

as
http://www.w3.org/2002/07/owl#FunctionalProperty

I agree these forms are surpising, we could modify 
http://www.w3.org/TR/owl-test/#style
appendix B stylistic preferences, 
perhaps by adding a sentence to section B.2 on xml:base e.g.

[[
Relative URIs used in the tests should not begin with "/" or ".".
]]

If we agreed that then the change to description-logic-605 and and other 
affected tests would be editorial.

Anyone else have an opinion.

Peter and Ian both spoke recently strongly opposing making some other tests 
easier. I wonder if they would oppose this change too. It would make this 
test easier, but perhaps in an area where the WG does not want to include 
unnecessary difficulties.

(I am neutral - HP software deals with this fine, but I don't see it as 
critical to not simplify this test)

Jeremy

Received on Saturday, 13 September 2003 06:25:10 UTC