proposed response to LC comment 43 sections 2,5,6

I am proposing the following response to LC Comment 43 sections 2, 5,
and 6.

In the absence of disagreement I suggest that the RDF mapping document
be modified as suggested in the response, the modifications be accepted
as editorial, and the response sent out.

peter



Dear Zhe,

Thank you for your message
  http://lists.w3.org/Archives/Public/public-owl-wg/2009Jan/0083.html
on the OWL 2 Web Ontology Language last call drafts.

Your message contains multiple sections, affecting more than one
document, and will thus generate multiple replies.  This response
is for sections 2, 5, and 6, which affect the mapping from the
functional syntax to RDF graphs.


************************
5. In the RDF mapping document, is it possible to keep OWL 2 vocabulary
   a bit smaller by replacing owl:minQualifiedCardinality with the
   existing owl:minCardinality?  Same idea applies to
   owl:qualifiedCardinality, owl:maxQualifiedCardinality.  After all,
   owl:onClass is there to differentiate the qualified vs.
   non-qualified case.
************************

The problem here has to do with monotonicity of the RDF semantics.
Consider a qualified min cardinality translation, i.e., something like
MinCardinality(2 ex:p ex:C), which translates into

	_:x rdf:type owl:Restriction 
	_:x owl:minQualifiedCardinality "2"^^xsd:nonNegativeInteger
	_:x owl:onProperty ex:p
	_:x owl:onClass ex:C

If this suggestion was made the translation would instead be

	_:x rdf:type owl:Restriction 
	_:x owl:minCardinality "2"^^xsd:nonNegativeInteger
	_:x owl:onProperty ex:p
	_:x owl:onClass ex:C

However, this contains the three-triple translation of MinCardinality(2
ex:p), and The RDF semantic will pick this up, and augment the meaning
of the above four triples with the meaning for MinCardinality(2 ex:p).

For minimum cardinality things are not so bad, because MinCardinality(2
ex:p ex:C) implies MinCardinality(2 ex:p).  However for Cardinality and
MaxCardinality this is not the case, and an incorrect meaning will be
determined.

This kind of problem has been known ever since the original Web Ontology
Working Group.  The RDF mapping document does not contain all the
rationale for the various choices in the mapping, so no change is
envisioned in response to this part of your comment.


************************
2. very minor typo
    RDF mapping document has a typo in Section 2.2. s/auhtor/author/.

6. In Section 2.2 of RDF mapping document, are we missing a translation?
   It is unclear how the second example in 2.2 is translated into
   triples.  The AnnotationAssertion in Table 1 has three parameters and
   that example has only two parameters for AnnotationAssertion.
************************

The second example in Section 2.2 is

	AnnotationAssertion( a:Peter
	    Annotation(
	       Annotation( a:author a:Seth_MacFarlane )
	       rdfs:label "Peter Griffin"
	    )
	) 

This is not syntactically correct.  The example was not correctly
changed from a previous syntax for annotation assertions.  The correct
example is

	AnnotationAssertion( 
	       Annotation( a:author a:Seth_MacFarlane )
	       rdfs:label a:Peter "Peter Griffin"
	) 

namely a singly-annotated annotation assertion.

Thank you for pointing out this error.  You also point out the
mis-typing of a:author in the example.

The document has been changed to fix these editorial mistakes.  The diffs
can be found at ..................................


Please acknowledge receipt of this email to
<mailto:public-owl-comments@w3.org> (replying to this email should
suffice). In your acknowledgment please let us know whether or not you
are satisfied with the working group's response to your comment. 

Regards,
Peter F. Patel-Schneider
on behalf of the W3C OWL Working Group

Received on Tuesday, 10 February 2009 19:11:36 UTC