Re: ISSUE-76 (DLP): REPORTED: DLP

My understanding of the story of DLP is similar to Carsten's. The goal 
of the work was to identify the ``intersection'' between logic 
programming and OWL. Of course, the meaning of ``intersection'' has to 
be taken carefully since, for instance, Logic Programming and OWL adopt 
different semantics.

As it happens with other fragments, like DL-Lite, there are many flavors 
of DLP. Even more, if one calls DLP any Horn-description logic, then 
HORN-SHIQ could also be seen as a variant of DLP. I must confess that 
the selection of the particular flavor of DLP included in the document 
was rather arbitrary and was taken from a set of papers about DLP 
published by the Karlsruhe people. I wouldn't be opposed to removing DLP 
and keeping Horn-SHIQ, since I am also not aware of non-toy ontology 
that belongs to the version of DLP in the document, but not to 
Horn-SHIQ. I expect that the people in Karlsruhe may have something to 
say about this issue (Markus?)

Bernardo

OWL Working Group Issue Tracker wrote:
> ISSUE-76 (DLP): REPORTED: DLP
>
> http://www.w3.org/2007/OWL/tracker/issues/
>
> Raised by: Bijan Parsia
> On product: 
>
> (On behalf of Carsten Lutz.)
>
> I would like to raise the question whether we really want DLP in the
> document. In my understanding, the history of DLP is as follows:
>
> - the origin was an academic exercise: to understand what one gets
>   when taking the common part of logic programming and OWL
>
> - the answer was (in my very personal opinion; never mind): nothing
>   very useful
>
> - ontologies written in DLP have never shown up (I am happy to
>   stand corrected)
>
> - it was superceeded by Horn-SHIQ of which it is fragment (right?),
>   and which is also in the document (where it is not really visible
>   that DLP is a fragment of Horn-SHIQ).
>
> I can see that the connection between logic programming and OWL is
> important, for a number of reasons. Still, I feel that DLP is an odd
> fragment and that we would do better to drop it. When discussing
> Horn-SHIQ, we could still mention that the intersection of logic
> programming and OWL is contained in it.
>
> If anybody wants to make a case for DLP, go ahead. I only want to
> avoid that we include fragments that nobody really supports.
>
>
>
>
>

Received on Wednesday, 28 November 2007 18:34:00 UTC