- From: Khalid Belhajjame <Khalid.Belhajjame@cs.man.ac.uk>
- Date: Thu, 03 May 2012 14:02:47 +0100
- To: Jun Zhao <jun.zhao@zoo.ox.ac.uk>
- CC: Stian Soiland-Reyes <soiland-reyes@cs.manchester.ac.uk>, public-prov-wg@w3.org
On 03/05/2012 12:42, Jun Zhao wrote:
>
> Stian,
>>> I am confused. Are you talking about wasStartedByActivity or
>>> wasInformedBy?
>>> I guess you are talking about wasStartedByActivity. And I think this is
>>> precisely why Tim is also pushing to keep only one of them in the DM.
>
> Good try!
>>
>> A wasInformedBy B says that B used X, wasGeneratedBy A. If we also
>> know entity X and how it was used and generated, wasInformedBy is just
>> restating that.
>
> Do you mean to say that
>
> A wasInformedBy B -->
>
> there is an X that wasGeneratedBy B, and used by A
That what I understood too. I suggest to make X explicit, as opposed to
leave it unspecified as the definition of wasInformedBy in the DM suggests.
>>
>> A wasStartedByActivity B says that B wasStartedBy X, which
>> wasGeneratedBy A. If we also know that entity X and how it started A
>> and was generated by B, then wasStartedByActivity is just restating
>> that.
>
> A wasStartedByActivity B ->
>
> A wasStartedBy X, which wasGeneratedBy B
>
> Although my understanding is that wasStartedByActivity was not meant
> as a shortcut, but for cases where X is truly unknown, for expressing
> control-flow relationship between activities. This is the 'hidden
> trigger' case that Tim said that he might have missed.
The example of workflow triggering a subworkflow uses this hidden
trigger. Here is an example that actually specify the entity (or
trigger) more explicit.
:letterRespection a prov:Activity .
:accidentNotification a prov:Entity .
:accidentNotification prov:wasGeneratedBy :letterReception .
:insurranceClaim a prov:Activity ;
prov:wasStartedBy :accidentNotification ;
prov:wasStartedByActivity :letterReception .
Notice that :insurranceClaim does not use :accidentNotification, and
therefore :insurranceClaim was not informed by :letterReception.
khalid
>
>
> cheers,
>
> Jun
>>
>>
>> So both are shortcuts, and should have value as such where we don't
>> know much about X, or where we add more data to that indirect
>> relationship.
>>
>>
>
>
>
Received on Thursday, 3 May 2012 13:03:17 UTC