Re: charter nominally extended + old drafts

4.5.3 R11.8 looks to me as identical to 4.5.6 R12.3. I suggest removing one of them.

I am not sure about 4.5.4 R12.1 and 4.5.2 R9. They are both based on the same use case, but are stated slightly differently from each other. 

The problem is that I do not see an obvious relationship between the use case and these requirements. The use case basically says that the validation does not have to rely on all nodes in the graph being connected to each other. And, indeed, targets address this use case. But both of the requirement statements are about something else. They imply that there should be a target that says “validate every resource in the graph against this constraint”. Such thing doesn’t exist - other than you enumerate all resources explicitly in the target. However, the requirement explicitly says "without referring to a specific set of resources or class". Further, the use case doesn’t support the requirement and there is no other use case that supports the requirement. I suggest removing one of the requirements. And re-wording the remaining requirement to address the use case.

For 4.5.1 R8, I would add something that indicates that when a shape is also a class, then complex shapes can be created by associating property constraints directly with classes. Subclass shape would then have all the constraints of the parent class shape plus its own.

Otherwise, looks good.

> On Jun 16, 2017, at 9:45 AM, Sandro Hawke <sandro@w3.org> wrote:
> 
> Great. Can people please take a look and say if they think it's good to publish as a note? Thanks.
> 
> - Sandro
> 
> On June 16, 2017 2:47:52 AM EDT, Simon Steyskal <simon.steyskal@wu.ac.at> wrote:
> fyi: I've just finished updating the UCR (cf. commit [1])
> 
> br simon
> 
> [1] 
> https://github.com/w3c/data-shapes/commit/11609d69a14545a3cd50e89619c4d81b648faad2
> 
> ---
> DDipl.-Ing. Simon Steyskal
> Institute for Information Business, WU Vienna
> 
> www: http://www.steyskal.info/  twitter: @simonsteys
> 
> Am 2017-06-12 20:53, schrieb Simon Steyskal:
>  Hi!
>  
>  I'll give the UCR a final read tmrw and report back as soon as I'm
>  finished.
>  
>  br simon
>  Hi!
>  I'll give the UCR a final read tmrw and report back as soon as I'm 
>  finished.
>  br simon
>  -------- Original message --------
>  From: Sandro Hawke <sandro@w3.org>
>  Date: 6/12/17 20:38 (GMT+01:00)
>  To: public-data-shapes-wg@w3.org
>  Subject: charter nominally extended + old drafts
>  
>  For people who didn't see the AC announcement last week when SHACL was
>  
>  published, the WG was also extended by two months, so it can help
>  address any issues that might arise during the AC review of SHACL.
>  I
>  don't expect we'll need to meet, and we should not take up any new
>  work.
>  
>  I just noticed, though, that there are two old Working Drafts:
>  
>  https://www.w3.org/TR/shacl-ucr/
>  and
>  https://www.w3.org/TR/shacl-abstract-syntax/
>  
>  These should be republished as WG NOTEs.   If there's no useful
>  consensus text in them, as I can imagine might be the case with absyn,
>  
>  the NOTE can just be a status section explaining why the draft was
>  abandoned.
>  
>  Thoughts on these?   Any last minute cleanup to UCR?   What should we
>  do
>  about AbSyn?
>  
>         -- Sandro
>  -------- Original message --------From: Sandro Hawke <sandro@w3.org>
>  Date: 6/12/17  20:38  (GMT+01:00) To: public-data-shapes-wg@w3.org
>  Subject: charter nominally extended + old drafts
>  For people who didn't see the AC announcement last week when SHACL was
>  published, the WG was also extended by two months, so it can help
>  address any issues that might arise during the AC review of SHACL.    I
>  don't expect we'll need to meet, and we should not take up any new 
>  work.
>  
>  I just noticed, though, that there are two old Working Drafts:
>  
>  https://www.w3.org/TR/shacl-ucr/
>  and
>  https://www.w3.org/TR/shacl-abstract-syntax/
>  
>  These should be republished as WG NOTEs.   If there's no useful
>  consensus text in them, as I can imagine might be the case with absyn,
>  the NOTE can just be a status section explaining why the draft was
>  abandoned.
>  
>  Thoughts on these?   Any last minute cleanup to UCR?   What should we 
>  do
>  about AbSyn?
>  
>         -- Sandro

Received on Saturday, 17 June 2017 19:52:11 UTC