RDF Reification for annotations and negative property assertions

[related to ISSUE-67 ISSUE-81 ISSUE-104]


In today's telco, we are going to discuss the usage of RDF reification
vocabulary for axiom annotations (ISSUE-67) and negative property assertions
(ISSUE-81). We will also discuss the issue of currently not having a
"disallowed vocabulary" (ISSUE-104). There is a connection between these two
A point, which I had in mind these days was that introducing RDF reification
into the OWL RDF syntax would lead to backwards compatibility problems. The
OWL 1 spec did actually explicitly /allow/ the use of reification vocabulary
in OWL DL ontologies in RDF graph form, see [1]. 

I was speculating that, if reification would have been used in the RDF
syntax of OWL 1, the reification vocabulary would have been disallowed, too.
So my concern was that using RDF reification in the OWL 2 RDF syntax would
cancel out custom usage of RDF reification. And this would then be a
backwards compatibility issue, since one can create a legal OWL 1 DL
ontology in RDF graph form, which makes use of RDF reification vocabulary,
and this ontology would then be syntactically invalid in OWL 2 DL. 

But now, after my mail [2] earlier this day, I think that this is not really
a problem. Like in the case of rdf:List, it should be easy for the reverse
RDF mapping to clearly recognize usage of reification for axiom annotations
and specifying negative property assertions. Everything else will then be
regarded as custom usage.

The only drawback would be if some existing OWL 1 ontology really uses the
URIs "owl:Axiom" and "owl:NegativePropertyAssertion" in combination with RDF
reification vocabulary. But, hey, this is really only a theoretical
backwards compatibility issue. 


[1] <http://www.w3.org/TR/owl-semantics/mapping.html#4.2>
[2] <http://lists.w3.org/Archives/Public/public-owl-wg/2008May/0188.html>

Received on Wednesday, 28 May 2008 12:58:55 UTC