W3C home > Mailing lists > Public > public-prov-wg@w3.org > October 2011

Re: PROV-ISSUE-90 (namespace-in-ontology): Namespace used in ontology [Formal Model]

From: Graham Klyne <GK@ninebynine.org>
Date: Thu, 20 Oct 2011 14:30:24 +0100
Message-ID: <4EA02270.2060009@ninebynine.org>
To: public-prov-wg@w3.org
Not necessarily *need* - the namespace could contain names not used in the Data 
Model.  But you might find it neater to do that.  I think there's a trade-off 
here between reducing the number of namespaces used in the RDF and each 
namespace having a clear(er) scope.

#g
--

On 20/10/2011 14:10, Paul Groth wrote:
> I think it's as you suggest. We may need extra namespace for anything that's not
> in the datamodel.
>
> cheers
> Paul
>
>
> Stian Soiland-Reyes wrote:
>> On Thu, Oct 20, 2011 at 10:13, Paul Groth<p.t.groth@vu.nl> wrote:
>>> I agree absolutely here. I would like to see one URL for the major concepts
>>> in the data model. It's weird to have two "official" urls.
>>>
>>> I wonder how we can do this?
>>> Can we not just have /ns/prov ?
>>
>> I would prefer that as well - but what about other potential
>> serialisations like "pure XML" which have been mentioned? Or the
>> implication of some of the PROV-O constraints like domain and range
>> vs. what is stated in PROV-DM?
>>
>>
>> If needed we can have ns/prov-o for "additional" concepts which are
>> not in PROV-DM, like EntityInRole - but then that would mean two
>> prefixes in the RDF, say prov: and provo:
>>
>>
>> Perhaps the OPM guys could help by enlightening us on how you did this
>> with OPM-V (vocabulary) vs. OPM-O (ontology). I believe that OPM-O
>> reuses the OPM-V concepts where it can.. right?
>
>
>
Received on Thursday, 20 October 2011 13:47:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 13:06:46 GMT