RE: proposed responses to public comments (deadline: Wednesday 09/26)

Hi,

ISSUEs 447 and 500 are closely related.

I would be glad to help work through the example, but unfortunately I have several deadlines this week and will not have much time until they pass.

Thanks,
Bob



________________________________
From: public-prov-wg-request@listhub.w3.org [mailto:public-prov-wg-request@listhub.w3.org] On Behalf Of Luc Moreau
Sent: Tuesday, September 25, 2012 6:49 AM
To: public-prov-wg@w3.org
Subject: Re: proposed responses to public comments (deadline: Wednesday 09/26)

Hi Robert,

Thanks for your feedback.  Just to help us track issues properly, I think that you are
responding to ISSUE-447, for which the group has produced its response [1]
but has not formally communicated it to you.

May I suggest the following: could we try to write your example in detail,
and use dcterms:hadPart to express the subactivity relation you are suggestion?
Are you able to list the entities and activities of your example?

Regards,
Luc

[1] http://www.w3.org/2011/prov/wiki/ResponsesToPublicComments#ISSUE-447_.28subactivity.29

On 24/09/2012 18:57, Freimuth, Robert, Ph.D. wrote:

Hi all,

As the submitter of ISSUE-500, I think it would be helpful to provide a specific use case to aid the discussion.  At its core, ISSUE-500 is:

"If activities were hierarchical (see comments for 2.1, wasPartOf), it would be possible to specify when a "parent" activity started to use an entity by defining the "usage" period as a separate child activity."

Note that the response on the wiki focuses only on the condition within the statement.  The impact of the second portion of the suggestion is not addressed.

Suggested use case for this issue:

The release of the final version of a white paper may include the following steps:
1. Editing the penultimate version based on feedback
2. Reviewing the updated document for accuracy
3. Approving the final version for release
4. Releasing the document on a public web site

Each of these steps could involve different actors/agents/responsible parties.  As a result, each one might be described using PROV.  However, it is also likely that these steps would be referenced as a single activity ("publication of the white paper") by someone higher up in the organization.  That more abstract description may also have PROV statements listing who is responsible for the oversight of the entire process (even though others may be responsible for the individual steps).

Without a way to indicate that each of the individual steps is part of a larger, more complex activity, it would be difficult to integrate the PROV statements to provide a complete and semantically accurate view of the process.  The ability to integrate statements will be critical when multiple parties create PROV statements about the same entities or activities, from different points of view.

For example, the four steps above collectively "use" the penultimate version of the document as well as the final version.  However, only the first sub-activity (editing) uses the penultimate version.  The first activity also generates the final version, which is then used in steps 2-4.

If I were to create PROV statements for only the parent activity ("publication of the white paper"), I would expect the activity to "use" the final version of the doc.  This does not, however, capture the full provenance described above.  To do that, I would also have to state (according to my understanding of the current spec) that the parent activity "uses" the penultimate version of the document and "generates" the final version of the document.  Clearly, there is a significant loss of semantics with this approach, and the problem will grow as the difference in level of abstraction between the statements increases.

If the suggestion above were supported, however, it would be possible to indicate that "publication of the white paper" used the penultimate version of the doc as part of one subactivity, generated the final version as part of another subactivity, and used the final document in other subactivities.

In my opinion, PROV must be able to support use cases like this so that it can be used to describe the provenance of complex activities.  Without this capability, it would be difficult to use PROV in complex organizations and processes.

As per the comments in ISSUE-447, I understand that the WG is intentionally avoiding modeling workflows.  However, I believe capturing the relationship between activities is just as important as capturing the relationship between entities.  Furthermore, I believe it is possible to support subactivities (e.g., subactivity wasPartOf parentactivity) without creeping too far into the workflow space.

Thanks,
Bob




-----Original Message-----
From: public-prov-wg-request@listhub.w3.org<mailto:public-prov-wg-request@listhub.w3.org>
[mailto:public-prov-wg-request@listhub.w3.org] On Behalf Of Luc Moreau
Sent: Monday, September 24, 2012 1:22 AM
To: public-prov-wg@w3.org<mailto:public-prov-wg@w3.org>
Subject: proposed responses to public comments (deadline:
Wednesday 09/26)


Dear all,

At http://www.w3.org/2011/prov/wiki/ResponsesToPublicComments,
find our proposed responses to public comments:
- ISSUE-492
- ISSUE-500
- ISSUE-505
- ISSUE-508

These will become the group responses, unless we hear objections by
Wednesday 09/26.

Best regards,
Luc

PS. To help tracker, please include only the relevant issue
number when
responding.

--
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<mailto:l..moreau@ecs.soton.ac.uk>
United Kingdom                     http://www.ecs.soton.ac.uk/~lavm








--
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<mailto:l..moreau@ecs.soton.ac.uk>
United Kingdom                     http://www.ecs.soton.ac.uk/~lavm