Re: [Asdf] Roman Danyliw's No Objection on charter-ietf-asdf-00-03: (with COMMENT)

[Niklas Widell <[e-mail]>]

On 2020-10-07, 21:02, "ASDF on behalf of Roman Danyliw via Datatracker" <[e-mail] on behalf of [e-mail]> wrote:

    ----------------------------------------------------------------------
    COMMENT:
    ----------------------------------------------------------------------

    To revisit my two items of feedback from the initial charter review:

    ** What is "Thing Interaction and Data Modelling"?
    ** What is currently outstanding to make SDF “IETF quality”?  I worry that the
    text currently reads as “start with SDF and make it better”.

    Here is text that could address those:

    OLD
    The objective of the ASDF working group is to develop SDF to an IETF-quality
    specification for Thing Interaction and Data Modelling, working with experts
    from OneDM and its contributing organizations.  On the way to that
    specification, further functionality requirements will be addressed that emerge
    in the usage of SDF for model harmonization.

    The ASDF WG will work closely with the CBOR WG, home of the CDDL specification.

    NEW

    The objective of the ASDF working group is to develop a standards-track SDF
    specification for thing interaction and data modelling. In the course of
    developing this specification, further functional requirements that emerge from
    the model harmonization will be addressed.

    The ASDF WG will work with the experts from OneDM and its contributing
    organization.  In the IETF, it will work closely with the CBOR WG, home of the
    CDDL specification.

I think these changes look good

PR here: https://github.com/one-data-model/ietf108/pull/10/files
(not merged to master yet if Carsten or Michael has comments, and not in DT)

Minor diffs: 
s/course/process/
s/organization/organizations/ - OneDM is the liaison activity of the participating organizations (OCF, OMA SpecWorks etc), so it is right to use plural here


    Explanation of the edit proposed edit:

    -- s/an IETF-quality specification/standards-track specification/ which speaks
    more concretely the quality bar expected

    -- s/Thing Interaction and Data Modelling/thing interaction and data modeling/
    because none of these are proper nouns.  I got confused in the initial charter
    review because the capitalization made it read like this was a formal name for
    something (proper noun)

Agree on Interaction and Data Modelling, on the fence with "Thing/thing" since an IoT Thing is a common concept. But no big deal, so it is "thing" in proposed -04.

    -- s/working with experts from OneDM and its contributing organizations//
    because this paragraph should describe what the WG is doing.  The reference to
    OneDM and contribution organization should be noted with the rest of partners
    (in the next paragraph).

Good point, moved

    -- s/On the way to that specification, further functionality requirements will
    be addressed that emerge in the usage of SDF for model harmonization/ In the
    course of developing this specification, further functional requirements that
    emerge from the model harmonization will be addressed/ to make clear the source
    of the functional requirements and why they can’t be articulated now.  If the
    basis of how any deconfliction will be done can be noted, this would be helpful
    (e.g., minimize disruptive change across the existing models in the ecosystem
    now)

Good point wrt deconfliction (interesting word, had to check exact meaning), the idea is to follow normal IETF process (rough consensus, objections handled etc.). From OneDM perspective, the toolchains for model translation have been successfully updated to reflect the development of SDF so far, so as Carsten noted in the BoF, so no special treatment ought to be needed.


    -- 
    ASDF mailing list
    [e-mail]
    https://www.ietf.org/mailman/listinfo/asdf