W3C home > Mailing lists > Public > public-owl-comments@w3.org > March 2009

[LC response] To Ivan Herman- LC comment on OWL 2 QL

From: Uli Sattler <sattler@cs.man.ac.uk>
Date: Wed, 25 Mar 2009 18:27:01 +0000
Message-Id: <C2337925-842F-4E5A-9FFC-7C10C9E5E72C@cs.man.ac.uk>
To: Ivan Herman <ivan@w3.org>
Cc: public-owl-comments@w3.org
Dear Ivan,

Thank you for your comment
      <http://lists.w3.org/Archives/Public/public-owl-wg/2009Feb/0007.html 
 >
on the OWL 2 Web Ontology Language last call drafts.

Your comment is closely related to the one by Maurizio (see http://lists.w3.org/Archives/Public/public-owl-comments/2009Jan/0014.html) 
  and Misha (see http://lists.w3.org/Archives/Public/public-owl-comments/2009Jan/0019.html) 
.

The working group has decided to implement the editorial changes and  
will correct the typos, thanks for pointing them out. We have also  
decided to add, to OWL 2 QL, reflexive, irreflexive, & asymmetric  
property axioms.

Finally, we have decided *not* to add sameAs or functional properties  
to OWL 2 QL:

- [sameAs] OWL 2 QL was designed so that any query Q against an  
ontology whose data (i.e., information about individuals, the classes  
they are instances of and how they are related via properties) is  
stored in a relational database DB, can be answered by rewriting the  
query Q into an SQL query Q1 and then answering Q1 against DB using a  
standard RDBMs. This property is known to be lost in the presence of  
sameAs. For the LOD community, we have added a small paragraph  
explaining that, if one wants to handle ontologies that are OWL 2 QL  
plus sameAs, then one can use a preprocessing step that materializes  
the sameAs relation and uses it in query answering procedures to  
simulate individual equality reasoning (see http://www.w3.org/2007/OWL/wiki/Profiles#Feature_Overview_2) 
.

- functional properties can only be added under the so-called unique  
name assumption, i.e., different names denote different indiviudals.  
This assumption, together with functional properties, will lead to  
inconsistencies if an individual has 2 successors w.r.t. a functional  
property -- a semantics that is suitable for some applications and  
unsuitable for others. Hence we have chosen to not include functional  
properties to OWL 2QL.

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,
Uli Sattler
on behalf of the W3C OWL Working Group
Received on Wednesday, 25 March 2009 18:27:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 March 2009 18:27:27 GMT