W3C home > Mailing lists > Public > public-dxwg-wg@w3.org > July 2019

Re: Narrowing down profile definition choices

From: Karen Coyle <kcoyle@kcoyle.net>
Date: Mon, 1 Jul 2019 15:07:57 -0700
To: Rob Atkinson <robatkinson101@gmail.com>
Cc: Dataset Exchange Working Group <public-dxwg-wg@w3.org>
Message-ID: <59cd6833-a6d0-04a9-d2e7-800c140686d9@kcoyle.net>
Anything with "may" is not a strict constraint. We have a lot of
requirements in section that read "may" and "should" but not "must".
Remember, and this gets confusing, we are talking about the definition
of profiles in general - any one profile could be quite specific, but a
statement like "Profiles may provide rules governing value validity" is
not a strict constraint on profiles. They may provide those rules, but
if they don't, they don't.

In checking, in the requirements document, section 6.11, there are no
instances of MUST, and the requirements are generally worded as "can"
"may", etc. In the profiles guidance document (which admittedly is
incomplete) we have only one instance of MUST: discoverability of the

"5.1 Profile metadata [RPFMD]

Profiles must be discoverable through a machine-readable metadata that
describes what is offered and how to invoke the offered profiles."

There is a note about requiring a URI for a profile as MUST, but it is
from a github discussion, not a requirement:

(Antoine:) From
Administrative metadata:

    A profile MUST have URI (or an IRI?) identifying it. The URI SHOULD
be an http URI
    creator/maintainer (+ contact info)


On 7/1/19 2:46 PM, Rob Atkinson wrote:
> I think it may be even more useful to try to capture what seem to be
> missing requirements and propose a Use Case that explains what you are
> getting at with notions of suggestions that are not strict constraints. 
> I have no objections to expanding the scope of the definition to meet
> such needs if there is a consesus to accept such a use case, but i dont
> think we can ignore effort and process so far and develop a definition
> not supported by agreed use cases.
> E.g.  i can see a value in a new property prof:followsRecommendations to
> use in conjunction with dc:conformsTo  and a property 
> prof:makes Recommendations rdfs:range dct:Standard
> And even prof:constrainsProperty x
> And prof:suggestsUseOf x
> Etc.
> As long as we follow a procesd of grounding in agreed use cases.
> On Tue, 2 Jul 2019, 05:53 Karen Coyle <kcoyle@kcoyle.net
> <mailto:kcoyle@kcoyle.net>> wrote:
>     All,
>     In order to get some traction on profile definition, I'd like to suggest
>     that we do a very loose straw poll just to see how the group is leaning.
>     There is a lot of discussion but only a few actual proposals, so I
>     suggest taking a straw vote on these three proposals, which seem to
>     cover the main directions:
>     1) Antoine's version  of the current definition using "may" [1]
>     2) Tom's version but with the looser list of qualities suggested by
>     Karen [2]
>     3) Richard's version emphasizing conformance [3]
>     We can up/down vote these on github with the emoji's. We could limit
>     each person to one up vote, but I think we should ask people to vote on
>     each one, either up or down, as a way to testing the waters. This won't
>     be a deciding vote but a view of how people are leaning today, and some
>     folks may be happy enough with more than one option.
>     How does that sound?
>     kc
>     [1] https://github.com/w3c/dxwg/issues/963#issuecomment-506436886
>     [2] https://github.com/w3c/dxwg/issues/963#issuecomment-507129887
>     [3] https://github.com/w3c/dxwg/issues/963#issuecomment-506438552
>     -- 
>     Karen Coyle
>     kcoyle@kcoyle.net <mailto:kcoyle@kcoyle.net> http://kcoyle.net
>     skype: kcoylenet

Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
skype: kcoylenet
Received on Monday, 1 July 2019 22:08:23 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:42:18 UTC