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

Re: PROV-ISSUE-368 (no-responsibility-in-derivation): No responsibility in derivation [prov-dm]

From: Timothy Lebo <lebot@rpi.edu>
Date: Mon, 30 Apr 2012 15:28:10 -0400
Message-Id: <FF38E9E2-9589-4D17-A3AA-CBD6AAFC6E95@rpi.edu>
To: Provenance Working Group <public-prov-wg@w3.org>
(below)

On Apr 30, 2012, at 6:47 AM, Paul Groth wrote:

> I agree with this suggestion. While we have had a long debate about
> quoting. I think there's an argument for simplicity here and if people
> need more complex forms of quoting then they need to do this as an
> extension.
> 
> cheers
> Paul
> 
> On Mon, Apr 30, 2012 at 12:41 PM, Provenance Working Group Issue
> Tracker <sysbot+tracker@w3.org> wrote:
>> PROV-ISSUE-368 (no-responsibility-in-derivation): No responsibility in derivation [prov-dm]
>> 
>> http://www.w3.org/2011/prov/track/issues/368
>> 
>> Raised by: Luc Moreau
>> On product: prov-dm
>> 
>> 
>> We are having ongoing discussion about responsibility in derivations (see ISSUE-357 and ISSUE-352).
>> 
>> In the spirit of simplification, I would like to suggest that agents should not be mentioned in derivation relations.
>> 
>> Instead of
>>  wasRevisionOf(id,e2,e1,ag,attrs)
>> we should write
>>  wasRevisionOf(id,e2,e1,attrs)
>>  and  wasAttributedTo(e2,ag)


I am in favor of this decoupling.
It reduces the need for PROV-O properties that will not be commonly used, while still being representable with existing, more fundamental constructs (wasAttributedTo and it's qualified form Attribution with Roles)

Making this change would allow us to eliminate the outlying PROV-O properties:
* prov:hadQuoter
* prov:hadQuoted
* prov:wasAttributedBy

-Tim


>> 
>> 
>> Instead of
>>  wasQuotedFrom(id,e2,e1,ag2,ag1,attrs)
>> we should write:
>>  wasQuotedFrom(id,e2,e1,attrs)
>>  and  wasAttributedTo(e1,ag1)
>>  and  wasAttributedTo(e2,ag2)
>> 
>> 
>> 
>> We are not losing in expressivity, I believe, instead, we decouple components 2 and 3 in the data model.
>> 
>> Furthermore, if we allow optional arguments in derivations,
>> wasDerivedFrom(id, e2, e1, a, g2, u1, attrs)
>> they should also be allowed in quotation/original source/revision, to
>> make these proper subrelations.
>> 
>> Cheers,
>> Luc
Received on Monday, 30 April 2012 19:28:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 30 April 2012 19:28:40 GMT