Re: proposed response to Jeremy's comment on owl:imports and graph names and issue 38

I read the use case that Jeremy provided as being intimately tied up with OWL 
constructs, notably owl:imports.  RDF itself has no analogous constructs, so I 
don't see a way of divorcing this use case from OWL, thus taking it out of the 
charter of the working group.

peter

On 10/08/2013 09:00 PM, Pat Hayes wrote:
> As I understand him, he is providing a (claimed) genuine use case which requires IRI names in datasets to denote their associated graphs. The fact that this uses OWL terminology seems irrelevant to me. Clearly the issue at which this is aimed is an RDF concern, and indeed arises directly from our charter. Given the decision we took last week, however, we could respond citing that decision.
>
> Pat
>
> On Oct 8, 2013, at 7:33 AM, Peter F. Patel-Schneider wrote:
>
>> If the use of owl:imports and owl:Ontology is not central to Jeremy's concerns, then let him put forward an example illustrating his concerns that does not use OWL vocabulary.
>>
>> That is the essence of the reply, couched in W3C WG-speak.
>>
>> peter
>>
>> On 10/08/2013 12:01 AM, Pat Hayes wrote:
>>> Well, as you know, because I have said this in earlier emails on this topic, I do not think this is an appropriate response, and that is is close to being deliberately disingenuous. It is obvious that the use of OWL is not central to the point that Jeremy is making here; but in any case, the issue concerns the relationship between an IRI used as a graph label in an RDF Dataset, and the same IRI used to refer inside RDF, and this matter is outside the scope of the OWL WG.
>>>
>>> Pat
>>>
>>> On Oct 7, 2013, at 10:37 AM, Peter F. Patel-Schneider wrote:
>>>
>>>> See below for a proposed response.
>>>>
>>>> peter
>>>>
>>>>
>>>> On 10/06/2013 11:22 AM, Guus Schreiber wrote:
>>>>> On 12-09-13 05:27, Peter Patel-Schneider wrote:
>>>>>> Because Jeremy's comment uses the OWL vocabulary, and particularly
>>>>>> because it uses owl:imports, the RDF Working Group should not be even
>>>>>> considering making any changes to RDF in response to the comment.  It is
>>>>>> the business of some future W3C working group on OWL to determine
>>>>>> whether owl:imports can be reasonably extended to RDF datasets, and
>>>>>> definitely not the business of the RDF working group.
>>>>>>
>>>>>> If Jeremy wants to provide some "common practice" where there is
>>>>>> inter-graph inference going on in RDF datasets that does not involve
>>>>>> vocabulary that is none of the RDF Working Group's business, then let
>>>>>> him bring that forward in a continuation of this comment (which we
>>>>>> should then consider as if it was brought forward during the LC period).
>>>>> I agree with Peter. I suggest to respond in this fashion.
>>>>> Guus
>>>>>
>>>> Hi Jeremy:
>>>>
>>>> This is an official response to your message about owl:imports and graph
>>>> names and issue 38.
>>>>
>>>> The practice that you illustrate concerns the OWL vocabulary for describing
>>>> and combining ontologies.  These facilities form a core portion of the W3C
>>>> OWL Web Ontology Language and are thus outside the scope of the RDF Working
>>>> Group.  The working group will thus not be addressing this issue. You may
>>>> wish to officially raise this issue against OWL, to be considered the next
>>>> time that OWL is updated.
>>>>
>>>> If you feel that there is a related issue that within the scope of the RDF
>>>> Working Group, feel free to raise it.
>>>>
>>>> Yours sincerely,
>>>> Peter F. Patel-Schneider
>>>> for the W3C RDF Working Group
>>>>
>>>>
>>>> From: Jeremy J Carroll <jjc@syapse.com>
>>>> Date: Thu, 11 Jul 2013 12:15:56 -0700
>>>> To: "public-rdf-comments@w3.org Comments" <public-rdf-comments@w3.org>
>>>>
>>>>> This is a formal comment on RDF Concepts 1.1
>>>>>
>>>>> I am concerned that the resolution of issue 38 leaves a disconnect.
>>>>>
>>>>> In particular, I think it is common practice to have datasets
>>>>>
>>>>> <g1>  {
>>>>>      <g1> rdf:type owl:Ontology
>>>>> }
>>>>> <g2> {
>>>>>      <g2> rdf:type owl:Ontology ;
>>>>>            owl:imports <g1> .
>>>>> }
>>>>>
>>>>> and this practice is somewhat undermined by the resolution of issue-38 which
>>>>> leaves a disconnect (^sd:name sd:graph) between the name and the graph.
>>>>>
>>>>> Jeremy J Carroll
>>>>> Principal Architect
>>>>> Syapse, Inc.
>>>>
>>> ------------------------------------------------------------
>>> IHMC                                     (850)434 8903 home
>>> 40 South Alcaniz St.            (850)202 4416   office
>>> Pensacola                            (850)202 4440   fax
>>> FL 32502                              (850)291 0667   mobile (preferred)
>>> phayes@ihmc.us       http://www.ihmc.us/users/phayes
>>>
>>>
>>>
>>>
>>>
>>>
>>
>>
> ------------------------------------------------------------
> IHMC                                     (850)434 8903 home
> 40 South Alcaniz St.            (850)202 4416   office
> Pensacola                            (850)202 4440   fax
> FL 32502                              (850)291 0667   mobile (preferred)
> phayes@ihmc.us       http://www.ihmc.us/users/phayes
>
>
>
>
>
>

Received on Wednesday, 9 October 2013 11:57:25 UTC