Re: [LC response] To Marko Luther

Dear Mr. Horrocks, Ms. Sattler, Working Group,

I acknowledge that communication protocols are not in the scope of the  
OWL Working Group according to Section 1 (Scope) of the OWL Charter  
[1]. However, without an axiom-level transport protocol OWL feels to  
me like HTML without HTTP. I am convinced that standardizing such a  
protocol could "..easing the adoption of OWL 1.1 features by OWL users  
and other members of the Semantic Web community.." (cf. Section 2.1 of  
the OWL Charter [1]) and support advanced infrastructures like stream  
reasoning [2] and distributed ontology editing [3]. To me it seems  
that the Manchester Syntax is covered by the Charter in a similar way.

The need for an implementation-neutral communication protocol that  
goes along OWL is reflected in the number of recent postings about the  
outdated DIG protocol found at the Pellet, Protege and Topbraid  
mailing lists [4-7]. The initial implementation of OWLlink as part of  
RacerPro 1.9.3 (soon to be released) demonstrates not only that it is  
implementable, but also that for communication intensive applications  
OWLlink/retraction exchanging OWL/XML axioms via HTTP may outperform  
even in-memory connections [8].

I would like to suggest to take the OWLlink specification available at  
<http:\\www.owllink.org> and lift it to a OWL 2 Working Note. OWLlink  
is defined in terms of a structural specification expressed in UML  
(initially contributed by Boris Motik), two bindings (HTTP/XML and  
HTTP/S-Expression) and a set of extensions (Retraction, Told,  
OntologyBasedDataAccess, EpistemicGroundedConjunctiveQueries). As  
OWLlink is extensible, it is open and ready for additions such as a  
SPARQL/OWL extension.

If needed, it should be possible for me to join the OWL Working Group  
(as my company is a W3C member) to ensure its finalization before  
10/2009.

Best regards,
Marko Luther

PS: Strange, that I couldn't find any discussion on my OWLlink (or my  
LC) in the public accessible WG minutes despite the very positive  
reaction of Alan Ruttenberg on our OWLlink presentation at OWLED'08 [9].

[1] http://www.w3.org/2007/06/OWLCharter.html
[2] http://streamreasoning.org/
[3] Timothy Redmond et al.: Managing Change: An Ontology Version  
Control System, In Proc. of OWLED 2008, Karlsruhe, 2008.
<http://www.webont.org/owled/2008/papers/owled2008eu_submission_33.pdf>
[4] <http://lists.owldl.com/pipermail/pellet-users/2009-January/003233.html 
 >
[5] <https://mailman.stanford.edu/pipermail/protege-owl/2009-February/009644.html 
 >
[6] <http://lists.owldl.com/pipermail/pellet-users/2009-February/003334.html 
 >
[7] <http://groups.google.com/group/topbraid-composer-users/browse_thread/thread/9c117caa5467c09f 
 >
[8] M. Luther et al. "Who the Heck is the Father of Bob?" to appear in  
Proc. of ESWC'09
[9] Thorsten Liebig et al.: OWLlink: DIG for OWL 2, In Proc. of OWLED  
2008, Karlsruhe, 2008.
<http://www.webont.org/owled/2008/papers/owled2008eu_submission_26.pdf>


On 16.03.2009, at 18:42, Ian Horrocks wrote:

> Dear Marco,
>
> Thank you for your comment
>     <http://lists.w3.org/Archives/Public/public-owl-comments/2009Jan/0048.html 
> >
> on the OWL 2 Web Ontology Language last call drafts.
>
> We acknowledge the importance of implementations and tool support  
> and, indeed, implementations supporting OWL 2 will be a condition  
> for the standardization of OWL 2. It is, however, not in the scope  
> of this working group to standardize communications protocols [1].
>
> On the other hand, we are creating a collection of test cases [2]  
> and would welcome help in the generation and testing of these cases.  
> So, if you think that OWLlink would be a suitable tool for testing  
> our test cases, then it would be great if you could coordinate with  
> the working group, for example Markus Kroetzsch and Mike Smith.
>
> [1] http://www.w3.org/2007/06/OWLCharter.html
>
> [2] http://km.aifb.uni-karlsruhe.de/projects/owltests/index.php/OWL_2_Test_Cases
>
> 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
> p.p. Ian Horrocks
> on behalf of the W3C OWL Working Group
>

--
Dr. Marko Luther

DoCoMo Communications Laboratories Europe GmbH
Landsberger Strasse 312, 80687 Munich, Germany
Geschäftsführer: Dr. Toru Otsu, Dr. Narumi Umeda, Kazushige Yoshida
Amtsgericht München, HRB 132967