Re: Outcome of profile definition discussion

Hi everyone,

Honestly I don't see this definition is 'obviously' distinct from another, because it's fairly complex ;-)
Joke apart, I see the points being made, and the attempt to remain flexible while close enough to the use cases.
So I can live with it for now.

But I feel that we may have to be ready to change it when we present it to a wider audience that's not very familiar with our work. Two examples at different levels:
- "subclasses" has a quite specific meaning in RDF and other languages (UML too...). Perhaps 'specializations' would be more neutral.
- "including the identification [...]" gives a focus to the definiton but doesn't formally excludes the things we don't want into (MIME types, programming languages...). Someone may still argue that it's possible to use it to include these things.

Best,

Antoine

On 07/02/18 05:21, Rob Atkinson wrote:
> Note:
> 
> This definition is obviously distinct from that of an encoding, hence negotiation over profiles is distinct from negotiation over MIME-types.  There should be no confusion.
> 
> This covers requirements that the definition support general profiles data may conform to, DCAT profiles (DCAT resources being data) and content-negotiation using profile identifiers. It is expect that profile negotiation will require identifiers to be IRIs, and DCAT guidance should recommend this - and possibly DCAT can enforce this using an owl:ObjectProperty to bind profiles to things.
> 
> DCAT profiles may be a further refinement of this more general definition, for example requiring that constraints are expressed specifically against RDF properties, and that IRIs identifying such profiles must resolve to a particular form of resource - such as instances of a RDF model describing the profile and any documents (PDF, SHACL) that describe the set of constraints.
> 
> 
> 
> 
> On Wed, 7 Feb 2018 at 13:23 Karen Coyle <kcoyle@kcoyle.net <mailto:kcoyle@kcoyle.net>> wrote:
> 
>     All,
> 
>     After considerable discussion, both in email and during the meeting, the
>     group approved the following definition for profiles:
> 
>     "A named set of constraints on one or more identified base
>     specifications, including the identification of any implementing
>     subclasses of datatypes, semantic interpretations, vocabularies, options
>     and parameters of those base specifications necessary to accomplish a
>     particular function. "[1]
> 
>     In reading over the email thread it occurs to me that we did not discuss
>     the term "vocabularies" - we may wish to revisit that.
> 
>     The requirement that the definition be contrasted to mime types was seen
>     as being specific to content negotiation and therefore would not be part
>     of a single definition of profile that would be appropriate for both the
>     Guidance deliverable and the Content Negotiation deliverable. This could
>     be added to the Content Negotiation document in its definition section,
>     if desired.
> 
>     This decision does not preclude further discussion if needed, but
>     hopefully provides a stake in the ground for our work.
> 
>     [1] https://www.w3.org/2018/02/06-dxwg-minutes
>     --
>     Karen Coyle
>     kcoyle@kcoyle.net <mailto:kcoyle@kcoyle.net> http://kcoyle.net
>     m: 1-510-435-8234 (Signal)
>     skype: kcoylenet/+1-510-984-3600 <tel:+1%20510-984-3600>
>

Received on Wednesday, 7 February 2018 08:53:30 UTC