RE: PROV-SEM staged, ready for review

Hello James,

Here's my review of PROV-SEM.

Yes, the document seem clear and the semantics correct. I see no blocking issues, but some more minor ones below. The two issues mentioned seem fully addressed to my (shallow) knowledge of them.

Abstract: "PROv" should be "PROV"

Section 1.1: "Although it is a work in progress" (3rd paragraph) and "(as yet)" (fourth paragraph). Is this appropriate for the final release of the note?

Section 2.2: "a set Values... we sometimes use the set P(Value)" (Values vs Value)

Section 2.3: "Times SUBSET Val" should be "Times SUBSET Values"

Section 2.4: "a null symbol "-". Placeholder symbols - " -> They seem to be the same symbol except that null has quotes round it (though it doesn't when used later). Can this be clarified?

Section 3.1: "Each thing has an associates set of Events" -> "associated set of Events"

Section 3.1, Component 1, point 3: "from things to sets of events." does not seem to explain the relation. Shouldn't it be "from things to the sets of events associated with those things."?

Section 3.2.4.1: "An Event is an influence whose events is a single time instant" -> I just couldn't interpret this statement. Are there one or several events (it is ungrammatical)? I assume when you say "whose events" you mean the events associated to the (influence) object, but if so, I think this could be rephrased to be clearer. It seems a circular definition, as "events" refers to elements of the set "Event", right? I think I would need some rephrasing to help understand what was intended.

Section 3.2.4.1: "relates an activity to an entity (which could be an agent)" -> should this be "(either of which could be an agent)"?

Sections 4.4.*: Many of the rules use a mixture of "e" and "ent", "e1" and "ent1", "e2" and "ent2", and in Semantics 30, "ent2.gen.act.use.ent1" should be "e2.g.a.u.e1" I think.

Section 4.4.12: "The events of ent1 is contained" (ungrammatical) 

Section 6.2.2: "the values of attributes of e are those immediately declared for e along with those of any specialization". Why? Isn't this the wrong way round? An entity's specialization has attribute values that the entity does not have (across its lifetime). I may have misunderstood the intent here.

Section 6.2.3: "amond" -> "among".

thanks,
Simon

Dr Simon Miles
Senior Lecturer, Department of Informatics
Kings College London, WC2R 2LS, UK
+44 (0)20 7848 1166

Modelling the Provenance of Data in Autonomous Systems:
http://eprints.dcs.kcl.ac.uk/1264/

________________________________________
From: Luc Moreau [l.moreau@ecs.soton.ac.uk]
Sent: 08 April 2013 21:35
To: public-prov-wg@w3.org
Subject: Re: PROV-SEM staged, ready for review

Hi James,

This is a very good and very clear document.
The outstanding issues can be closed for me.
I have a few minor questions/suggestions/typos below.

Cheers,
Luc


- abstract: PROv -> PROV
- sotd paragraph: add the paragraph about feedback and errata.

- section 1: paragraph 1: definition of provenance: cite prov-dm.

- section 1.1: generation/invalidation  -> generation, invalidation

- section 1.1: make 'not' bold
    not a definitive specification of the ONLY semantics
    ->
    NOT a definitive specification of the ONLY semantics


- I like the mapping to events.

-  remark in 3.2.4.6: assocated -> associated

- remark in 3.2.4.6: I didn't understand the last paragraph of the
remark. why is it there is no way to express multi step derivation
path?

- component 15 axioms: I was surprised to see a reference to things among
the axioms (see axiom 5).  But I guess, this is OK.


- section 3.3, remark: well spotted that we miss a constraint for
   invalidation.
   Did you identify it when proving properties?

- section 4.3.2. How is \rho(st) defined?  what it 'st' is a constant
   or an identifier?

- section 6.2: for the reader, can you make explicit why this is "WEAK
completness".

- just before 6.2.1: why can you set each time variable to some dummy value?

   Is it because once, you have established an instance is valid, then
   you have already unified all possible time variable to ground
   values. Then, instantiating any time variable can no longer break
   any constraints.

- 6.2.3:

   amond -> among




On 05/04/2013 17:38, James Cheney wrote:
> Hi,
>
> PROV-SEM is now ready for review here:
>
> https://dvcs.w3.org/hg/prov/raw-file/default/semantics/releases/NOTE-prov-sem-20130430/Overview.html
>
> As before, because it renders math using MathML, different browsers do better/worse jobs with it.  I get the best results with Safari (Mac OS X) and Firefox does OK, while Chrome does not do very well.  Accordingly, I've put a PDF built using Safari here:
>
> https://dvcs.w3.org/hg/prov/raw-file/default/semantics/releases/NOTE-prov-sem-20130430/prov-sem.pdf
>
> Several people volunteered to review by next week's teleconference, when (I believe) we will vote on all remaining NOTEs.
>
>
> Please address the following review questions:
>
> 1. Is the purpose of the document clear and consistent with the working group's consensus about the semantics? If not, can you suggest clarifications or improvements?
>
> 2. Are there minor issues that can be corrected easily prior to final release?
>
> 3. Are there blocking issues that must be addressed prior to final release?
>
> 4. ISSUE-579 requested that we incorporate an axiomatization using a more standard logic formalism e.g. first-order logic.  The current draft attempts to address this.  Can this issue be closed?
>
> 5. ISSUE-635 requested that we address the issues of soundness and completeness in the semantics.  This is currently attempted, by generalizing the semantics (which unfortunately also decreases the connection to intuitive notions of time.)  As a result, we have a soundness and weak completeness result stating that any valid PROV instance has a model and vice versa.  Can this issue be closed?
>
> --James
>
>
>

--
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 Tuesday, 9 April 2013 09:54:42 UTC