W3C home > Mailing lists > Public > public-prov-wg@w3.org > December 2011

Re: PROV-ISSUE-200: Section 6.3 (PROV-DM as on Dec 5) [prov-dm]

From: Luc Moreau <L.Moreau@ecs.soton.ac.uk>
Date: Thu, 08 Dec 2011 11:04:19 +0000
Message-ID: <EMEW3|2c353cc8dffaf60e2363381c6c364c8anB7B4M08L.Moreau|ecs.soton.ac.uk|4EE099B3.5000001@ecs.soton.ac.uk>
To: public-prov-wg@w3.org
Hi Satya,

This is again *very frustrating*. This PROV-ISSUE-50, and your last
communication on this dates back from October 2nd!!!!

http://lists.w3.org/Archives/Public/public-prov-wg/2011Oct/0007.html

You have not responded to my message on Oct 3rd
http://lists.w3.org/Archives/Public/public-prov-wg/2011Oct/0009.html
and November 30th
http://lists.w3.org/Archives/Public/public-prov-wg/2011Nov/0419.html

Further comments below.

On 12/07/2011 02:22 AM, Provenance Working Group Issue Tracker wrote:
> PROV-ISSUE-200: Section 6.3 (PROV-DM as on Dec 5) [prov-dm]
>
> http://www.w3.org/2011/prov/track/issues/200
>
> Raised by: Satya Sahoo
> On product: prov-dm
>
> Hi,
> The following are my comments for Section 6.3 of the PROV-DM (as on Dec 5):
>
> Section 6.3
> 1. Given:
>
> "An information flow ordering record is a representation that an entity was generated by an activity, before it was used by another activity."
>
> How does the constraint:
>
> "Given two activity records denoted by a1 and a2, if the record wasInformedBy(a2,a1) holds, then the following temporal constraint holds: the start event of the activity record denoted by a1 precedes the end event of the activity record denoted by a2."
>
> make sense?
>
> Detailed comment: Let us consider the scenario: "a chemical reagent r1 was generated by activity a1 in September 2011" and "r1 was used by researcher in experiment activity a2 in December 2011". From a provenance perspective, we simply state r1 was generated before it was used - where is the relevance of activity ordering in above scenario (entity was generated before it was used)?
>    
you will note that the above constraint is a necessary condition and not 
a sufficient condition!

The point about information flow ordering is that the entity does not 
have to be explicitly mentioned.  It's an existential quantifier over an 
entity.

If you activity a1 had started after the end of a2, it would have been 
impossible to have this entity r1 generated by a1 and used by a2.

> Further, activity ordering is important in provenance from a very different perspective - "analyzing provenance of bank transactions to justify penalty fee for customer c1 - the $100.00 deposit activity da1 took place before or after $80.00 withdrawal activity wa1 happened in account with starting balance of $5.00". So, if da1 happened before wa1 there should not be any penalty fee, otherwise customer has to pay fee for withdrawing more money than was available in the account.
>    

This example is handled by having various account entities for the 
various balances.
I don't think we need an explicit activity ordering here.

> Clearly, the informedBy property does not address the requirement of activity ordering for provenance. In addition, the current definition of informedBy for representing whether entity was generated before it was used does not need activity information - it can be asserted either in terms of event ordering or temporal ordering. I believe we should remove wasInformedBy or move it out of activity ordering section.
>
>    
What do you mean by *clearly*?
How can we assert this by event/temporal ordering ?
1. The relations precedes/follows do not belong to the data model (they 
are used in interpretation)
2. The interpretation over wasInformedBy is a necessary condition, not a 
sufficient condition.
            There needs to be an entity used and generated between these 
activities, though we dont have to assert it.


> 2. "The relationship wasInformedBy is not transitive. Indeed, consider the records wasInformedBy(a2,a1) and wasInformedBy(a3,a2), the record wasInformedBy(a3,a1), may not necessarily hold, as illustrated by the following event line."
>
> Comment: It is not clear from the description and the figure, why wasInformedBy is not transitive? It is difficult to interpret the figure without additional description.
>    

Tim mentioned this. THis will be dealt with.
> 3. "Given two activity records identified by a1 and a2, the record wasStartedBy(a2,a1) holds if and only if there exist an entity record identified by e and some attributes gAttr and sAttr, such that wasGeneratedBy(e,a1,gAttr) and wasStartedBy(a2,e,sAttr) hold."
>
> Comment: The above definition and related example for wasStartedBy are not clear at all. What is meant by the statement that "spawn-request" was generated by a1 and "spawn-request" is in a wasStartedBy relation with a2, hence a1 and a2 also have a wasStartedBy relation? Is "spawn-request" supposed to represent control message exchanged between a1 and a2 or something else?
>    

There were problems in the text. I hopefully fixed them. Can you check?

Luc
> Thanks.
>
> Best,
> Satya
>
>
>
>    

-- 
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
Received on Thursday, 8 December 2011 11:05:04 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:51:04 UTC