W3C home > Mailing lists > Public > public-prov-wg@w3.org > May 2012

Re: PROV-ISSUE-369 (drop-startByActivity): Should we drop wasStartedByActivity? [prov-dm]

From: Paul Groth <p.t.groth@vu.nl>
Date: Wed, 2 May 2012 10:05:26 +0200
Message-ID: <CAJCyKRo1cUuDFuugz3JDPBr5TFLRxDKHCH3aEr6_b38aba9Ueg@mail.gmail.com>
To: Timothy Lebo <lebot@rpi.edu>
Cc: Provenance Working Group <public-prov-wg@w3.org>
I think Tim makes a good point. Just having communication should be
enough to model what people need (namely control flow) and this is
just an extra relation.

cheers
Paul

On Tue, May 1, 2012 at 11:51 PM, Timothy Lebo <lebot@rpi.edu> wrote:
> below.
>
> On Apr 30, 2012, at 6:57 AM, Provenance Working Group Issue Tracker wrote:
>
>> PROV-ISSUE-369 (drop-startByActivity): Should we drop wasStartedByActivity? [prov-dm]
>>
>> http://www.w3.org/2011/prov/track/issues/369
>>
>> Raised by: Luc Moreau
>> On product: prov-dm
>>
>>
>> There are regular comments about wasStartedByActivity.
>>
>> MacTed was saying why don't we also have wasStartedByAgent.
>>
>> Someone else  was saying, why don't we have wasEndedByActivity? ... and ByAgent.
>>
>> In the spirit of simplification, we could drop the concept.
>> Whenever we want to write
>>
>> wasStartedByActivity(a2,a2)
>>
>> we would instead have to write
>>
>> wasStartedBy(a2,e)
>> wasGeneratedBy(e,a1)
>>
>> for some entity.
>>
>>
>> Thoughts?
>
> While the diagram in PROV-O's starting points diagram [1] shows one essential relation among Agents (acted) and one essential relation among Entities (wasDerived),
> there are __two__ for Activities (wasInformed and wasStartedByActivity).
>
> I think this is reason to attempt simplification.
>
> Between wasStartedBy (an Entity -- i.e. a Start) and wasInformedBy (an Activity -- i.e. a Communication), don't we have enough ways to model an Activity starting an Activity?
>
> I still see wasStartedByActivity as a kind of Communication, and would rather replace it with just Communication (and not some specialization of Communication).
> The conflict is some notion that "triggers" are not "used", but I cannot see how this could be the case. (Perhaps I'm exposing my lack of OPM legacy).
>
> -Tim
>
>> Luc
>
>



-- 
--
Dr. Paul Groth (p.t.groth@vu.nl)
http://www.few.vu.nl/~pgroth/
Assistant Professor
Knowledge Representation & Reasoning Group
Artificial Intelligence Section
Department of Computer Science
VU University Amsterdam
Received on Wednesday, 2 May 2012 08:05:56 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 2 May 2012 08:05:57 GMT