Fwd: Re: simplifying attribution

Hi Tim,

I take the liberty to forward this email to the mailing list, as a 
record for tracker,
regarding ISSUE-258.

The outcome is that we are not going to merge wasAttributedTo and 
wasAssociatedWith as per discussed
below.  However, all responsibility relations have been grouped in a 
same component, in the new presentation.

Hence, I am closing ISSUE-258.

Cheers,
Luc



-------- Original Message --------
Subject: 	Re: simplifying attribution
Date: 	Wed, 14 Mar 2012 16:52:44 -0400
From: 	Timothy Lebo <lebot@rpi.edu>
To: 	Paul Groth <p.t.groth@vu.nl>
CC: 	Stephan Zednik <zednis@rpi.edu>, Luc Moreau 
<L.Moreau@ecs.soton.ac.uk>, Paolo Missier <paolo.missier@ncl.ac.uk>



Luc,

I think based on the feedback from Stephan and Paul, I should scratch this off my list of concerns.

I spoke with Jim McCusker today, and he convinced me that old:wasAttributedTo is a property chain on prov:wasGeneratedBy and prov:wasAssociatedWith,
so they would clearly not be the same and gives a reasonable distinction.

Thanks for considering the consolidation.

Best,
Tim


On Mar 15, 2012, at 1:07 AM, Paul Groth wrote:

>  Stephan says it well.
>
>  I think we are using Attribution in the sense of "ascribe a work or remark to"
>
>  Association is a more broad term, it probably has more of the feeling of participation. That is you needed to be there for the activity to occur as it did. But the activity wasn't ascribed directly to me.
>
>  I think we could add add activity as a domain for wasAttributedTo but we should keep the definition as and just leave wasAssociatedWith alone. Would that work?
>
>  cheers
>  Paul
>
>  Stephan Zednik wrote:
>>  --Stephan
>>
>>  On Mar 14, 2012, at 8:53 AM, Timothy Lebo<lebot@rpi.edu>   wrote:
>>
>>>  Hi, guys.
>>>
>>>  On Mar 14, 2012, at 5:11 PM, Paul Groth wrote:
>>>
>>>>  Hmm... I thought wasAssociatedWith had a different role than wasAttributedTo.
>>>>
>>>>  Being associated with an activity is different than an entity/activity being attributed to someone?
>>>  What do you see different between the two?
>>
>>  I my initial reaction would be that there is a distinction between association (participation?  I've always felt the term association was a little broad for the semantics we are implying here) in an activity and attribution.
>>
>>>>  Am i missing something?
>>>  I don't see the difference, and the inspiration for consolidating is to factor out the "type" of subject that we are tying to an Agent.
>>
>>  Do you think it would make sense to merge the binary (non-qualified) relations?
>>
>>  It would give the binary property a non-simple domain in OWL.
>>
>>>  A paper was attributed to an Author.
>>>  The typing activity was attributed to the Author.
>>
>>  I think there is a slight difference in what we are saying here.  Similar, yes, but the meaning of the statements is different.
>>
>>  I worry that the meaning of the combined attribution property will change depending on the subject type.
>>
>>>  The author has some responsibility in both the results and the activity that led to the results.
>>
>>  I agree, but I am not sure yet that it makes sense to collapse responsibility for activity and entity into a single relation.
>>
>>  --Stephan
>>
>>>  Regards,
>>>  Tim
>>>
>>>
>>>
>>>>  Paul
>>>>
>>>>  Luc Moreau wrote:
>>>>>  I should have included Paul in this thread.  This is now done.
>>>>>
>>>>>  Professor Luc Moreau
>>>>>  Electronics and Computer Science
>>>>>  University of Southampton
>>>>>  Southampton SO17 1BJ
>>>>>  United Kingdom
>>>>>
>>>>>  On 14 Mar 2012, at 07:46, "Luc Moreau"<l.moreau@ecs.soton.ac.uk>    wrote:
>>>>>
>>>>>>  A further question, can
>>>>>>  we see the plan in new:wasAttributedTo(entity,plan)
>>>>>>  as a license?
>>>>>>
>>>>>>  Thanks,
>>>>>>  Luc
>>>>>>
>>>>>>  On 14/03/2012 07:33, Luc Moreau wrote:
>>>>>>>  Hi Paolo, Tim and Stephan,
>>>>>>>
>>>>>>>  Tim asked whether we could simplify wasAttributedTo and wasAssociatedWith and have
>>>>>>>  a single relation, which could take the form
>>>>>>>
>>>>>>>   new:wasAttributedTo(entityOrActivity,agent,plan,attributes)
>>>>>>>
>>>>>>>  So new:wasAttributedTo is proposed as a replacement for old:wasAttributedTo
>>>>>>>  and old:wasAssociatedWith.
>>>>>>>
>>>>>>>  Benefit: one less relation, smaller model
>>>>>>>
>>>>>>>  Question:
>>>>>>>
>>>>>>>  - does the term attribution make sense for activities:
>>>>>>>      new:wasAttributedTo(activity,agent)
>>>>>>>
>>>>>>>  - does it make sense to involve a plan in the absence of an activity
>>>>>>>  In particular, could we also write
>>>>>>>       new:wasAttributedTo(entity,plan)
>>>>>>>
>>>>>>>
>>>>>>>  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
>>>>
>>>>
>
>  -- 
>  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
>
>