W3C home > Mailing lists > Public > public-dxwg-wg@w3.org > July 2018

Re: Profile Guidance roundup

From: Antoine Isaac <aisaac@few.vu.nl>
Date: Tue, 17 Jul 2018 19:30:26 +0200
To: <public-dxwg-wg@w3.org>
Message-ID: <7c5e707e-e6b1-c25e-df32-415fe6cc0d4e@few.vu.nl>
Hi,

Thanks indeed, Karen. This is very useful!

@Rob: my model is an earlier attempt to caracterize what is in ProfileDesc, but it's not so early ;-)
It was at the last F2F. Looking at ProfileDesc, and trying to make sense of it - as you had done already some terminology choices, which I found hard to understand (probably because I missed the flexibility of being able to group notions with slashes ;-) )

Cheers,

Antoine

On 16/07/18 17:18, Karen Coyle wrote:
> Note about dct:Standard and more:
> 
> The DC terms definition of dct:Standard was NOT changed, precisely
> because it would have narrowed the definition and rendered some current
> data invalid.
> 
> But more significantly, DC is moving toward the use of defining domains
> and ranges of DC terms as being the semantic equivalent of schema.org's
> "domainIncludes" "rangeIncludes". This is in recognition of the reality
> that many uses of DC terms are working with legacy data that predates
> the use of IRIs. Domains and ranges thus become recommendations or best
> practices, but are not constraints.
> 
> On 7/15/18 5:21 PM, Rob Atkinson wrote:
>>
>> Thanks for doing this Karen - its very helpful.
>>
>> A couple of suggestions and notes: (i'll leave you to edit it unless you
>> ask me to)
>>
>> 1) I think Andrea's effort to list a few different constraint languages
>> https://docs.google.com/spreadsheets/d/1Zty4jTzhG0_1xoJlDOMq1XeHelIwVP2-STw6_-_ZxR4/edit#gid=0 is
>> significant - it pretty much evidences that there is no single
>> constraint language to rule them all. IMHO this needs to be part of the
>> guidance - to choose the most appropriate constraint language for your
>> community..
> 
> Yes, I'll add this as a section (or paragraph) on constraint languages.
> I assume (at the moment) that constraint LANGUAGEs are optional and
> potentially outside of the application profile itself, but represent a
> direction that APs should be moving toward, either as the AP language or
> as a separate representation of AP-defined constraints.
> 
> kc
> 
>>
>> 2) apropos the list of "data models" - Antoine's model is i think an
>> early attempt to render the profiledesc concepts being proposed -
>> (correct me if i'm wrong) - and it is in any way the same data model
>> that is expressed more formally in profiledesc.
>> Stephane's model has a different scope - it addresses the much broader
>> topic of schema mapping - if you split out the profile description part
>> it is trivially consistent with profiledesc, and reinforces its role and
>> design - a profile subclasses dcterms:Standard - though this is at risk
>> if DC tightens definitions of dcterms:Standard in a non-backward
>> compatible way.
>>
>> I havent located the reference (i think from you) to the proposed
>> redefinition of dct:Standard - if this is indeed a possibility then can
>> you drop in a link under the data models section - or perhaps we need to
>> reopen https://www.w3.org/2017/dxwg/track/actions/152
>>
>> Rob
>>
>>   
>>
>> On Thu, 12 Jul 2018 at 02:42 Karen Coyle <kcoyle@kcoyle.net
>> <mailto:kcoyle@kcoyle.net>> wrote:
>>
>>      As promised on the call yesterday, I've added a wiki page[1] based on my
>>      own list-making relating to the profile guidance document. It attempts
>>      to gather in one place various decisions and contributions so that it is
>>      easier to see where we are in that work. I started it because it felt to
>>      me that our work in this area is pretty scattered, although I think
>>      that's in a sense the nature of starting a new document that we haven't
>>      fully defined as yet. As I also said yesterday, this is far, far from
>>      complete, but I'll keep working on it, as should others.
>>
>>      I'd like to keep this as a list of pointers, so discussion should take
>>      place elsewhere, mainly github. We can link to the discussions from here
>>      so there is a "one-stop" place to find links.
>>
>>      I've added it to the main wiki page under Profiles for the moment.
>>
>>      Let me know if this looks useful.
>>
>>      kc
>>      [1] https://www.w3.org/2017/dxwg/wiki/ProfileRoundup
>>      --
>>      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 Tuesday, 17 July 2018 17:30:58 UTC

This archive was generated by hypermail 2.3.1 : Monday, 25 March 2019 10:33:24 UTC