LC2 Responses/JC6

From OWL
Revision as of 08:00, 16 May 2009 by Michael Schneider (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Merged reply


To: jeremy@topquadrant.com
CC: public-owl-comments@w3.org
Subject: [LC response] To Jeremy Carroll

Dear Jeremy,

Thank you for your comment
     <http://lists.w3.org/Archives/Public/public-owl-comments/2009May/0017.html>
on the OWL 2 Web Ontology Language last call drafts.

The Working Group discussed this topic, in the context of ISSUE-81 <http://www.w3.org/2007/OWL/tracker/issues/81>, and decided not to use an RDF encoding of the proposed form for negative property assertions. Although such a mapping has the effect of reducing the additional vocabulary required for OWL 2 in RDF, the Working Group believes that this advantage is more than outweighed by several disadvantages: it obscures the fact that this is a new feature in OWL 2, impedes the ability to retain ontology structure in RDF, and would make it more difficult for both users and tools to specify and detect negative property assertions in RDF-encoded OWL 2 ontologies. The Working Group does not find any new information in your proposal that might justify reopening the issue.

Regarding the claim that this is an advanced feature that is unlikely to be interoperably supported, the Working Group sees no reason to believe that this is the case. On the contrary, the feature is already supported by FaCT++, HermiT and Pellet. In addition, Oracle is planning to support negative property assertions in a soon to be released version of OWL Prime (see <http://www.w3.org/2007/OWL/wiki/Implementations>), and the Working Group is also aware of another implementation of OWL 2 RL for which it has been reported that supporting negative property assertions was straightforward. Finally, the OWL 2 test suite includes relevant tests, and these are already being passed by multiple implementations (see <http://www.w3.org/2007/OWL/wiki/Test_Suite_Status>).

Therefore, the Working Group does not intend to make the change you propose.

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 and Michael Schneider
on behalf of the W3C OWL Working Group



CUT AND PASTE THE BODY OF THE MESSAGE (I.E. FROM "Dear" TO "Group") INTO THE BODY OF AN EMAIL MESSAGE. SET THE To:, CC:, AND Subject: LINES ACCORDINGLY.

PLEASE TRY TO REPLY IN A WAY THAT WILL ALLOW THREADING TO WORK APPROPRIATELY, I.E., SO THAT YOUR REPLY CONTINUES THE THREAD STARTED BY THE ORIGINAL COMMENT EMAIL



Draft by Michael Schneider


To: jeremy@topquadrant.com
CC: public-owl-comments@w3.org
Subject: [LC response] To Jeremy Carroll

Dear Jeremy,

Thank you for your comment
     <http://lists.w3.org/Archives/Public/public-owl-comments/2009May/0017.html>
on the OWL 2 Web Ontology Language last call drafts.

The Working Group asserts that the proposal does not provide new information. The Working Group has discussed this topic in the past, in the context of ISSUE-67, and has decided to not use an RDF encoding of the proposed form for negative property assertions.

Concerning the claimed merit of a reduced need for additional vocabulary, the Working Group believes that having additional vocabulary terms for the RDF encoding of negative property assertions is justified. Negative property assertions are a new language construct of OWL 2, and the Working Group has defined an RDF encoding for it that corresponds directly to the respective construct in the structural specification. The RDF encoding is also meant to be intention revealing. In contrast, the Working Group has the opinion that the proposed RDF encoding does not meet this criterion and will be unhelpful for both users and tools to specify and detect negative property assertions in RDF-encoded OWL 2 ontologies.

Concerning the claim that negative property assertions are an advanced feature, the Working Group believes that this is not the case. Negative property assertions can be expressed in terms of OWL 1 language features, as you show by your proposal, and so they have been supported for years by compliant OWL systems, although not in an explicit form. The working group is not aware of any evidence that interoperable support for negative property assertions will be unlikely. At the time of writing, several in-use implementations of OWL 2 are available that provide conformant support for negative property assertions, and there have not been any problems reported by now to the Working Group. In particular, the Working Group knows about an implementation of OWL 2 RL, for which it has been reported that supporting negative property assertions has not been a problem.

Following your proposal would require significant, and in some cases non-obvious changes to several of the documents, including the Mapping to RDF Graphs, the RDF-Based Semantics and the Profiles specifications. Given the late stage of the standardization process for these recommendation-track documents, the Working Group is not willing to perform significant changes without a pressing need, which the Working Group believes is not given in this case.

Therefore, the Working Group will not make the change you propose.

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,
Michael Schneider
on behalf of the W3C OWL Working Group



CUT AND PASTE THE BODY OF THE MESSAGE (I.E. FROM "Dear" TO "Group") INTO THE BODY OF AN EMAIL MESSAGE. SET THE To:, CC:, AND Subject: LINES ACCORDINGLY.

PLEASE TRY TO REPLY IN A WAY THAT WILL ALLOW THREADING TO WORK APPROPRIATELY, I.E., SO THAT YOUR REPLY CONTINUES THE THREAD STARTED BY THE ORIGINAL COMMENT EMAIL



Draft by Peter F. Patel-Schneider


To: jeremy@topquadrant.com
CC: public-owl-comments@w3.org
Subject: [LC response] To Jeremy Carroll

Dear Jeremy,

Thank you for your comment
     <http://lists.w3.org/Archives/Public/public-owl-comments/2009May/0017.html>
on the OWL 2 Web Ontology Language last call drafts.

The Working Group acknowledges that the mapping you suggest has the effect of reducing the additional vocabulary required for OWL 2 in RDF, in line with one of the goals of the Working Group.

Nonetheless, the Working Group does not view this as a desirable change. The Working Group feels that the mapping you suggest actually obscures the fact that this is a new feature in OWL 2, and thus would not have the effect that you appear to desire. The mapping that you suggest would also impede the ability to retain ontology structure in RDF, which is another goal of the Working Group.

The Working Group sees no reason to believe that this feature is unlikely to be interoperability supported (sic). In fact it is already supported by FaCT++, HermiT and Pellet, and Oracle is planning to support negative property assertion in a soon to be released version of OWL Prime (see <http://www.w3.org/2007/OWL/wiki/Implementations>). Moreover, the OWL 2 test suite includes relevant tests, and these are already being passed by multiple implementations (see <http://www.w3.org/2007/OWL/wiki/Test_Suite_Status>).

The Working Group will therefore not make the change that you suggest to the RDF mapping of negative property asserions.

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



CUT AND PASTE THE BODY OF THE MESSAGE (I.E. FROM "Dear" TO "Group") INTO THE BODY OF AN EMAIL MESSAGE. SET THE To:, CC:, AND Subject: LINES ACCORDINGLY.

PLEASE TRY TO REPLY IN A WAY THAT WILL ALLOW THREADING TO WORK APPROPRIATELY, I.E., SO THAT YOUR REPLY CONTINUES THE THREAD STARTED BY THE ORIGINAL COMMENT EMAIL



We suggest replacing the two mapping rules:

NegativeDataPropertyAssertion( DPE a lt ) 	
=>
_:x rdf:type owl:NegativePropertyAssertion .
_:x owl:sourceIndividual T(a) .
_:x owl:assertionProperty T(DPE) .
_:x owl:targetValue T(lt) .

And

NegativeObjectPropertyAssertion( OPE a1 a2 ) 	 
=>
_:x rdf:type owl:NegativePropertyAssertion .
_:x owl:sourceIndividual T(a1) .
_:x owl:assertionProperty T(OPE) .
_:x owl:targetIndividual T(a2) .

We suggest the following alternative to the second one (the first is similar):

NegativeObjectPropertyAssertion( OPE a1 a2 ) 	 
=>
T(a1) rdf:type _:c .
_:c rdf:type owl:Class .
_:c owl:complementOf _:r .
_:r rdf:type owl:Restriction .
_:r owl:onProperty T(OPE) .
_:x owl:hasValue T(a2) .


The merits are: a) reduces need for additional vocabulary b) makes it clear that this is an advanced feature that is unlikely to be interoperability supported

Jeremy Carroll, AC Rep, TopQuadrant, Inc.