Re: Language Tags Requirement

On 4/25/15 2:59 PM, Jose Emilio Labra Gayo wrote:
> On Sat, Apr 25, 2015 at 12:45 AM, Holger Knublauch 
> <holger@topquadrant.com <mailto:holger@topquadrant.com>> wrote:
>
>     Sorry, Jose, for possibly appearing pedantic, but this is not how
>     I understood the vote that took place yesterday. I thought the
>     vote was about splitting the requirement into two, and then vote
>     on the spawned off requirement separately. As defined right now, I
>     have no idea what this requirement is supposed to do, and I would
>     prefer a better definition.
>
>     Does it mean that the high-level vocabulary shall include
>     something like
>
>     ex:MyGermanAddressShape
>         sh:property [
>             sh:predicate ex:strasse ;
>             sh:lang "de"
>         ] .
>
>     and then all values of ex:strasse must carry the language tag "de"?
>
>
> Yes, I would expect something like that.
>
>
>     If so, is this a common enough use case that needs to go into the
>     Core vocab?
>
>     Let's keep in mind that the Core is expected to be supported by a
>     wide array of tools. I see no technical problems supporting this,
>     yet I wonder where to draw the line, or whether such things should
>     rather go into some extension library of Core. For example the
>     requirement that there should be no two values with the same
>     language tag appears to me much more common.
>
>
> The requirement does not talk about core vocab because there is no 
> such a thing at this moment. I prefer to talk about SHACL high level 
> language and I prefer that high level language to be the more 
> expressive that we can so the constraints expressed with it can be 
> compatible between implementations.
>
> The separation in two requirements was suggested by Richard saying 
> that the former could be considered a simple constraint while the 
> latter could be considered a complex constraint. I think if we have 
> approved the whole requirement, I see no reason why we should not 
> approve the requirement divided in two parts.

One reason why approving the complex requirement does not automatically 
also imply approval of the high-level language requirement is that we 
risk ending up with a bloated high-level language that increases the 
implementation and training overhead - the core is expected to be 
supported by every platform. Another argument is that we are duplicating 
things that are already solved with languages like SPARQL that provide 
many more ways of combining the primitive language elements. This 
doesn't sound like the right strategy to me, especially given that 
SPARQL already provides the compatibility that you mention above.

>
> AFAIK, at this moment we are not talking about how to implement those 
> requirements.

I am really puzzled by all this. It seems like once we decided that we 
leave the question open whether something becomes high-level language or 
complex feature, all previous votes on these requirements have become 
quite meaningless. Must have been a very big misunderstanding.

Holger


>
> Best regards, Jose Labra
>
>
>     Thanks for clarification.
>     Holger
>
>
>     On 4/25/15 4:02 AM, Arnaud Le Hors wrote:
>>     Hi Jose,
>>     You don't need to create another issue for this. I expect the
>>     editors to pick it up and include it in their draft as part of
>>     their normal activities.
>>     Thanks.
>>     --
>>     Arnaud  Le Hors - Senior Technical Staff Member, Open Web
>>     Technologies - IBM Software Group
>>
>>
>>     Jose Emilio Labra Gayo <jelabra@gmail.com>
>>     <mailto:jelabra@gmail.com> wrote on 04/23/2015 10:26:46 PM:
>>
>>     > From: Jose Emilio Labra Gayo <jelabra@gmail.com>
>>     <mailto:jelabra@gmail.com>
>>     > To: RDF Data Shapes Working Group
>>     <public-data-shapes-wg@w3.org> <mailto:public-data-shapes-wg@w3.org>
>>     > Date: 04/23/2015 10:28 PM
>>     > Subject: Language Tags Requirement
>>     >
>>     > As per my action from yesterday's WG telecon, I have created a new
>>     > requirement in the Wiki called Language Tags:
>>     >
>>     > https://www.w3.org/2014/data-shapes/wiki/Requirements#Language_Tags
>>
>>     > As I understand that the requirement has been approved, Should it
>>     > also be added to the FPWD or should I add another Issue to ask for
>>     > its inclusion?
>>     >
>>     > --
>>     > -- Jose Labra 
>
>
>
>
> -- 
> -- Jose Labra
>