Re: prov-dm: draft for the new 'hadPlan' relation

Hi Stian,

I am in agreement with what you said.  That's a concern we also had.
But it is a natural consequence of defining an agent as any entity being
associated with an activity.

In fact, you will see that this problem is not specific to the new relation
hadPlan.
Look at the second example in the section
http://dvcs.w3.org/hg/prov/raw-file/default/model/ProvenanceModel.html#record-OrderingOfActivities

There, a request 'starts' the activity, and hence is also an agent.

We may want to revise the notion of agent, and have it defined standalone,
independently of wasAssociatedWith.  Then, wasAssciatedWith could be
associated with entities and agents.

I think we had a hard fought battle on agents at F2F1. I will not change 
the text until we
have a new vote on agents.

Luc

On 12/14/2011 02:38 PM, Stian Soiland-Reyes wrote:
> On Wed, Dec 14, 2011 at 12:37, Luc Moreau<L.Moreau@ecs.soton.ac.uk>  wrote:
>    
>> Following last week's agreement about introducing a notion of Plan and a
>> hadPlan relation,
>> find a first draft of the section describing this relation.
>>
>> http://dvcs.w3.org/hg/prov/raw-file/default/model/ProvenanceModel.html#record-planLink
>>      
> Generally looks good. Captures most of the concerns previously raised,
> such that it might not go according to plan, or the plan could change.
>
>
> However I don't quite understand why a plan is always also an agent,
> except due to restrictions on wasAssociatedWith(), which could be
> revised.
>
>
> An agent is to me something active, which might make its own
> decisions.  (I don't mind that a plan CAN be an agent)
>
>
> So I would expect there could be an agent who is *using* the plan (ie.
> informed by it) and who is *controlling* (or even performing) the
> activity. That is probably usually the case.
>
> If I have built an IKEA book case according to the assembly
> instructions, then the entity, agent and plan should be quite obvious
> and distinct.
>
>
> Your examples:
>
>    
>> a plan can be a workflow for a scientific experiment
>>      
> plan: workflow
> agent: workflow engine
> activity: workflow execution
>
>    
>> a recipe for a cooking activity
>>      
> plan: recipe
> agent: chef
> activity: cooking
>
>
>    
>> a list of instructions for a micro-processor execution
>>      
> plan: instructions
> agent: micro processor
> activity: micro processor execution
>
> (the last can be an example of where agent and activity can also be the same)
>
>
> I however struggle to think of an example where the plan is also an agent.
>
>
>    

-- 
Professor Luc Moreau
Electronics and Computer Science   tel:   +44 23 8059 4487
University of Southampton          fax:   +44 23 8059 2865
Southampton SO17 1BJ               email: l.moreau@ecs.soton.ac.uk
United Kingdom                     http://www.ecs.soton.ac.uk/~lavm