Re: [QB] CR preparation

On 05/03/2013 10:48 AM, Dave Reynolds wrote:
> Richard, all,
>
> I've progressed the preparation for moving Data Cube to CR. Details 
> below. Please check since we need the WG to be happy that everything 
> in place for this (and for ORG) in time for Thursday's vote.
>
> 1. I've recorded all the editorial changes made in response to Last 
> Call comments in the document [1].
>
> 2. We ideally need to get back to each commenter to confirm how we 
> have dealt with their comments. We've done that, and have 
> acknowledgements, for all non-editorial comments but haven't yet done 
> so for the editorial changes. Richard are you OK for me to go ahead 
> and do that?
>
> 3. I've created a static render of the proposed CR version of document 
> at [2]. Pubrules seems happy. As usual this will not look right in 
> Chrome thanks to security restrictions. This will be part of the 
> documentation package we will be voting on so needs checking with the 
> stuff below.
>
> I've set the same arbitrary CR minimum period as for ORG. We need to 
> agree on what the right answer is for both of them.
>
> 4. I've filled in the our CR transition page [3]. The key aspect is 
> the CR exit criteria. As discussed last Thursday I'm proposing that 
> Data Cube conformance be tested using the integrity constraints 
> defined in the spec. Unlike ORG, I'm not suggesting splitting the 
> vocabulary into feature groups.
>
> I'm suggesting that the normalization algorithm and integrity checks, 
> that were At Risk at Last Call remain At Risk through CR. This 
> requires a bit of care since we need most of the integrity checks to 
> survive in order to use them for CR exit! If the phrasing on the CR 
> page isn't clear let me know. If it is clear but you don't like it 
> then definitely let me know :)
>
> 5. Linked from the CR page I've started a page of implementations [4]. 
> I've just filled in those I know directly about and the ones mentioned 
> by people in the Last Call comments. I know there are a lot more 
> around, especially via DERI so help on filling our more examples please.
>
> Sandro, chairs, what else is needed?

No, that seems fine.

>
> Cheers,
> Dave
>
> [1] 
> https://dvcs.w3.org/hg/gld/raw-file/default/data-cube/index.html#change-history
>
> [2] https://dvcs.w3.org/hg/gld/raw-file/default/data-cube/static-cr.html
>
> [3] http://www.w3.org/2011/gld/wiki/Data_Cube_CR_transition
>

Excellent work!    The Exit Criteria and At Risk descriptions seem spot on.

I fixed two trivial typos.

I'd suggest removing the enumeration of editorial changes, since it 
gives the impression of significant change.  It's probably better to 
just have the statement that none of them have been substantive and a 
link to the editorial ones in the document change-history.


> [4] http://www.w3.org/2011/gld/wiki/Data_Cube_Implementations
>

Very impressive list.   As I look at it, i wonder about adding 3rd party 
comments to the Description.    I can imagine the first one (hopefully 
this email HTML will be okay):

Dave Reynolds 	*Environment Agency, Bathing water quality.* Data Cubes 
are used to represent but current and history weekly and annual 
assessments of quality of water at bathing locations in England and 
Wales 	http://environment.data.gov.uk


being:

Dave Reynolds 	*Environment Agency, Bathing water quality.* Data Cubes 
are used to represent but current and history weekly and annual 
assessments of quality of water at bathing locations in England and Wales

*Comment *from Sandro Hawke (7 May 2013): I looked through this for 
about 20 minutes and it seemed sensible.  I ran the verification queries 
and the data passed.    (This is a lie; I didn't do any of this, I'm 
just showing what a comment might look like.)
	http://environment.data.gov.uk



The comments would just be stuff relevant to CR-exit and (in general) 
people trying to figure out whether these show the quality and value of 
the vocab.

        -- Sandro.