comments on OWL 2 QL

Dear Working Group,

We have two comments on OWL 2 QL, document 7:

1. In section 3.1 it's written that "the following constructs are not 
supported in OWL 2 QL:

....

- reflexive properties
- irreflexive properties
- asymmetric properties
- equality (sameIndividual)

In fact, as shown in [1] below, one can (easily) reduce (in LogSpace) 
query answering in OWL 2 QL extended with the above mentioned constructs 
to query answering  in OWL 2 QL as defined in the document (that is, 
without those constructs). Thus, the data complexity remains in 
LogSpace, and so the constructs CAN BE INCLUDED in OWL 2 QL without 
changing complexity.

Furthermore, if we add

- transitive properties

then query answering becomes data complete in NLogSpace.

2. In Section 5, Table 10, it's written that taxonomic complexity and 
combined complexity of OWL 2 QL is in PTIME. In fact, it 
NLogSpace-complete, even with all the constructs mentioned above.

[1] A. Artale, D. Calvanese, R. Kontchakov, and M. Zakharyaschev, The 
DL-Lite family and relatives. Available at 
http://www.dcs.bbk.ac.uk/~michael/DL-LiteFamily.pdf

Regards,

Roman Kontchakov & Michael Zakharyaschev

-- 
Michael Zakharyaschev
Professor of Computer Science
School of Computer Science and Information Systems
Birkbeck College
Malet Street
London WC1E 7HX
UK
Tel: +44-20-7631 6716
Fax: +44-20-7631 6727
http://www.dcs.bbk.ac.uk/~michael



The W3C OWL Working Group has just published a set of eleven documents
for public review.  These documents cover "OWL 2", which is now fairly
stable.  This is a good time to review this work and send us comments
(at public-owl-comments@w3.org).  Comments received by 23 January will
be answered and taken into account in the next round of publications.
The group expects to issue the OWL 2 Candidatate Recommendations, with
a call for implementations, shortly after that comment deadline.

The documents are:

    1. Structural Specification and Functional-Style Syntax
       http://www.w3.org/TR/2008/WD-owl2-syntax-20081202/
    2. Direct Semantics
       http://www.w3.org/TR/2008/WD-owl2-semantics-20081202/
    3. RDF-Based Semantics
       http://www.w3.org/TR/2008/WD-owl2-rdf-based-semantics-20081202/
    4. Conformance and Test Cases
       http://www.w3.org/TR/2008/WD-owl2-test-20081202/
    5. Mapping to RDF Graphs
       http://www.w3.org/TR/2008/WD-owl2-mapping-to-rdf-20081202/
    6. XML Serialization
       http://www.w3.org/TR/2008/WD-owl2-xml-serialization-20081202/
    7. Profiles
       http://www.w3.org/TR/2008/WD-owl2-profiles-20081202/
    8. Quick Reference Guide
       http://www.w3.org/TR/2008/WD-owl2-quick-reference-20081202/
    9. New Features and Rationale
       http://www.w3.org/TR/2008/WD-owl2-new-features-20081202/
   10. Manchester Syntax
       http://www.w3.org/TR/2008/WD-owl2-manchester-syntax-20081202/
   11. rdf:text: A Datatype for Internationalized Text
       http://www.w3.org/TR/2008/WD-rdf-text-20081202/

Of these eleven documents:

      * The first seven are the core technical specification for OWL 2.
        Six of these now at "Last Call".  This means the Working Group
        believes the design is done and that the only changes going
        forward will be editorial, along with certain items highlighted
        as "at risk" in the documents.  The other one, "RDF-Based
        Semantics" (which specifies OWL 2 Full) is expected to proceed
        to Last Call shortly.

      * The "Quick Reference Guide" and "New Features and Rationale"  
will
        be a good place to start for many OWL users, these documents  
will
        be a good place to start.  (A Primer is also being developed,  
and
        a earlier draft was published, but it has not yet been  
updated to
        account for the current state of the language.)

      * "Manchester Syntax" specifies an alternative syntax for OWL
        which some users prefer and some tools implement, but which is
        not required.

      * "rdf:text" specifies an XML datatype (developed jointly with
        the RIF Working Group) which provides a way to deal with RDF
        language-tagged literal strings without making them be a
        special case.

Again, please send comments to public-owl-comments@w3.org by 23 January.
Discussion on either of these lists is fine, but might not be seen by
the Working Group.

Ian Horrocks
Chair, OWL Working Group

Received on Friday, 23 January 2009 10:53:22 UTC