Re: prov-dm, prov-n, prov-constraints preliminary staging

Hi

Thanks. Feedback below.

On 23/11/12 17:06, Graham Klyne wrote:
> On 21/11/2012 12:58, Luc Moreau wrote:>
> > Hi all,
> >
> > Given that our CR-track documents should be frozen by Monday, it would
> > be useful to have a few pairs of eyes checking the preliminary staged
> > versions of the specs.
>
> I'm skimming PROV-DM:
>
> Intro: do we drop the "editors latest draft" link in the document 
> header for an actual published document?

Maybe Ivan may want to comment on this?

>
> Section 1.4 (namespaces/prefixes):  is this really non-normative? I'd 
> have thought the information here is needed to properly interpret some 
> of the normative material.

The choice of prefix is indeed non normative.

>
> Should section 4.2 have a "non-normative" annotation?  4.3? (section 
> 4.1 does).  On reflection, I'd suggest dropping the annotation from 
> 4.1 as there's already one for section 4.

Good catch. Don't know why. It looks like a respec.js bug. Only section 
4 has class informative.

>
> Section 5.6: if I'm reading the diagram notation correctly, this 
> indicates that EmptyCollection is a subtype of Collection.  (That 
> would imply it can have HadMember relations with other Entities.) 
> Normally, I would expect it to be an *instance* of Collection.  In the 
> usage here, I'd expect it to be a related type, but not a subtype.  I 
> didn't follow the discussions of containers, so I don't know what is 
> the motivation for including an EmptyCollection type.

Correct reading.
EmptyCollection shouldn't be an instance. I can see my wine cellar today 
and an empty list as both empty collections (no wine left in my cellar 
today :-(

Constraint 56 deals with membership and empty collections 
http://dvcs.w3.org/hg/prov/raw-file/default/model/prov-constraints.html#membership-empty-collection_text


>
> Section 7: why bullet points?  I'd suggest dropping these and just 
> having 3 paragraphs.

Done

>
> Appendix A.  In it's current form, the cross reference links to PROV-N 
> are not working.  Same for PROV-O.  This may be an artifact of the 
> staging, but I think this should be checked carefully as the document 
> moves to publication.
>

Yes

> Appendix B:  Should this (change-log) remain for the CR publication?
>

W3C requirement

> References: it seems strange to me that we have a normative reference 
> to RFC3987, but none to RFC3986.
>

why?


> PROV-OVERVIEW link broken (probably a staging artifact)

In the process of being written


Luc

> #g
> -- 
>
>
> On 21/11/2012 12:58, Luc Moreau wrote:
>>
>> Hi all,
>>
>> Given that our CR-track documents should be frozen by Monday, it would
>> be useful to have a few pairs of eyes checking the preliminary staged
>> versions of the specs.
>>
>> http://dvcs.w3.org/hg/prov/raw-file/default/model/releases/CR-prov-dm-20121211/Overview.html 
>>
>>
>> http://dvcs.w3.org/hg/prov/raw-file/default/model/releases/CR-prov-n-20121211/Overview.html 
>>
>>
>> http://dvcs.w3.org/hg/prov/raw-file/default/model/releases/CR-prov-constraints-20121211/Overview.html 
>>
>>
>>
>> Known issues:
>> - text about CR exit criteria, implementation report, etc needs to be 
>> added to
>> intro.
>> - documents use dated URL where they will be published, so dangling 
>> links.
>>
>> Can you let us know if you can spare a few minutes glancing at the 
>> documents,
>> providing
>> any feedback on potential bugs and/or typos.
>>
>> Luc
>>
>> PS. I am not aware that prov-o is ready yet
>>
>>

-- 
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 Friday, 23 November 2012 20:47:26 UTC