PROV-ISSUE-495: ivan's feedback on prov-dm LC [prov-dm]

PROV-ISSUE-495: ivan's feedback on prov-dm LC [prov-dm]

http://www.w3.org/2011/prov/track/issues/495

Raised by: Luc Moreau
On product: prov-dm



Original email: http://lists.w3.org/Archives/Public/public-prov-wg/2012Aug/0001.html

Copied below for convenience:




"A Prov document a day keeps the, hm, doctor? away" 

Got myself through the prov-dm document today. Congrats, by the way:-) Only a few comments below, and none of them are show-stopper in terms of process...

- General editorial issue: the document does not make any difference between normative and non-normative sections. At the moment, all sections are then normative. Is this intentional? After all, the examples (like 4.1), but maybe all sections 2-3-4 can be considered as  non-normative... 

The differentiation does make sense. If, God Forbid!, there is a discrepancy between, say, an example and the definition than the normative section counts. If both of them are normative then, well, we have a spec inconsistency...

- it seems that the " ◊ " character is used for references (or anchors) or some sort. It is a bit strange for an average spec reader like me, although it may be a question of taste. Can we try to find something else (colour, etc)

- (I may be wrong in my understanding here…) My impression is that, in some cases, some relations are actually shorthands that could be 'opened up' by an application and filled in by further application details. For example, derivation is a very high level relationship between two entities; an application may decide to 'open up' that relationships, describe how an entity was derived by others by virtue of describing the various activities and agents that are responsible for the 'WasDerivedFrom' relationship.

If what I say is true, I think this is an important feature that should be emphasized in the introduction section. Applications are free to decide which level of granularity they want describe, and the current prov-dm gives them the way to do that. Which is great, but I think is worth emphasizing in the text; for the time being it is, sort of, hidden between the lines.

(There may be other, similar situations to derivation)

- This is actually the same comment as I had yesterday for the primer: in 4.1 the character '-' appear in the examples. It is clear from the description that it replaces a positional argument; it is probably a good idea to put a note somewhere in the text that that particular positional argument is time (which of course becomes clear later, but readers are linear...). Or simply add time as part of the example?
 
- Let us be consistent in the in the examples in section 5 with respect to spaces (or not) after a ','. Compare example 17 and example 18, for example...

- It is well defined that, for example, start is part of component #1 and not a core term. However, when one gets to the formal definition of start (5.1.6) then this is not really clear any more, the reader may forget (after all, there are many terms...). Maybe it would be possible to separate the core and the non-core terms within 5.1 (using subsections)… (this is of course valid for the other sections, too)

- Informative references: the dates are wrong for all Prov documents (all of them are set to 2011…) 

THIS IS NOT A COMMENT JUST A THOUGHT!: once the prov spec is final, I wonder whether it would be possible to add to recspec an explicit generation of provenance into the W3C document using prov and using RDFa. That would be cool; eating our own dog food, that sort of things...

Cheers

Ivan

----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
FOAF: http://www.ivan-herman.net/foaf.rdf

Received on Monday, 10 September 2012 08:26:16 UTC