FW: FO Council for Timed Text WG charter

For information ahead of tomorrow's call, here is the response from the FO Council for the Timed Text WG charter:

On 23/11/2022, 17:42, "rhiaro" <amy@rhiaro.co.uk> wrote:

    Hi all,

    Here is the response from the council:

    The council has considered the formal objections raised against the
    proposed Timed Text WG charter.

    **In summary: we propose an update to the charter wording which aims to
    resolve the objections while keeping our understanding of the intent of
    the WG.**

    The council acknowledged that the WG felt it was inappropriate to use
    FOs to impose additional specification success requirements beyond what
    is stated in the Process. However, the council had consensus that the
    Process actually does not contain any hard requirements for success
    criteria, but rather expresses guidance on the kinds of things to
    consider when making a judgement about sufficient implementation
    experience for a particular specification or feature. Thus, we see the
    Process as a baseline upon which WG charters build to clarify their
    implementation expectations in their particular context. In particular,
    since charters often go beyond the guidance of the Process, it is within
    scope of an AC review for AC reps to express an objection if they
    believe that for a particular charter there should be tighter criteria.

    We also understand that the Timed Text WG did not intend to weaken the
    success criteria when updating the charter wording. However, it seemed
    to the council that the current phrasing of 3.1 is ambiguous. We would
    like to see if the group agrees with our interpretation of the text,
    which would address the objections. If it does, clarifying the text
    accordingly would resolve this situation by consensus. If not, the
    council would like to ask a follow up question.

    The current text of 3.1
    (https://w3c.github.io/charter-timed-text/#advancement) says (with added
    emphasis):

     > Within the requirements of the Process **each new normative feature**
    will be considered as ready to advance to a higher maturity level if its
    implementability has been demonstrated.
     >
     > When considering suitability to advance **any feature** beyond
    Candidate Recommendation, at least two independent factors of
    verification MUST be demonstrated, which may come from any of:
     >
     > * Presentation implementation
     > * Content
     > * Validating implementation

    Here is our understanding of how to interpret this. While verification
    can generally come from any of the three factors listed above, they need
    to be appropriate for the feature being tested.  Here are some examples:
    the spec can contain normative requirements on validators. For those
    requirements, 2 validating implementations would be acceptable factors
    of verification. The spec can contain normative requirements on authors,
    and for those requirements, two independent sources of content would be
    acceptable factors of verification. The spec also contains requirements
    on presentations, and for those, 2 presentation implementations would be
    required. A validating implementation and a presenting implementation
    could be considered as 2 factors of verification for parsing
    requirements (for example), but would not be sufficient for normative
    requirements pertaining only to presentation. Under this interpretation,
    the spec would not be able to advance beyond Candidate Recommendation if
    there existed, for example, only one validating implementation and only
    one presentation implementation of the entire specification, because
    there wouldn't be two factors of verification **for each feature**.

    If this is the group's intent, we suggest clarifying that. In the
    council's understanding, this would resolve the objection without the
    council needing to rule on it.

    A proposed rephrasing to achieve that would be:

     >Within the requirements of the Process each new normative feature will
    be considered as ready to advance to a higher maturity level if its
    implementability has been demonstrated.
     >
     > When considering suitability to advance any feature beyond Candidate
    Recommendation, at least two independent factors of verification
    <ins>for each normative requirement</ins> MUST be demonstrated, which
    may come<ins>, as relevant for that requirement,</ins> from any of:
     >
     > * Presentation implementation
     > * Content <ins>producing implementation</ins>
     > * Validating implementation

    A different rewording, with similar intent, would probably lift the
    objectors' concerns as well.

    If this is not the group's intent, and the group does mean that a
    specification should be able to exit CR if there existed, for example,
    only one content producing implementation and one validating
    implementation for each feature, then the council has a follow-up
    question. Given the common practice across different WGs of requiring
    two independent implementations, in the view of the Timed Text WG, what
    particularity of the technology or ecosystem makes is so that adequate
    implementation experience could be judged sufficient despite the lack of
    a second implementation of some normative requirements?

    I'll be happy to discuss more and answer any questions on your call
    tomorrow. See/hear you then.

    Cheers,

    Amy

    On 23.11.2022. 12:28, Nigel Megitt wrote:
    > Hi Amy,
    >
    > Thank you for this, I'm keen to see the proposals, and our call on Thursday is planned to go ahead, even though some folk may not attend due to Thanksgiving in the USA. For expediency, I think it would be useful for you to join, and I'll rejig the agenda so we put this topic first. Even if not everyone can attend, at least folk who don't attend can read the minutes.
    >
    > I'm slightly surprised that the way you're phrasing this is that you hope the updates will resolve the FO by consensus: the reason this got to an FO Council was because previous attempts, over many months, failed, largely due to the unresponsiveness of the objector. I also noticed that the objector was included in the list of people on the council, which is troubling from a governance perspective, especially since there were objections to their presence. It would be interesting to know how much input they had in the discussion and in the decision.
    >
    > I was expecting the FO Council to determine the end result, and that would be it, but if there remains some flexibility, it would be really helpful to understand the range of options.
    >
    > Anyway, I will hold off any further judgement until I've seen the proposal!
    >
    > Kind regards,
    >
    > Nigel
    >
    >
    >
    > On 22/11/2022, 21:04, "rhiaro" <amy@rhiaro.co.uk> wrote:
    >
    >      Hi Nigel and Gary,
    >
    >      Members of the AB and TAG met this week as a council to discuss the FOs
    >      to the Timed Text WG charter. I chaired the meeting, so I'm following up.
    >
    >      I'm still waiting for final sign-off on the written outcome, but the
    >      short version is that we have drafted some updates to the charter text
    >      that we hope would resolve Apple's FO by consensus, if the WG is
    >      agreeable to the suggestion.
    >
    >      I'm emailing you before the text is ready, because I noticed that your
    >      next meeting is scheduled for this Thursday, the 24th (assuming it
    >      hasn't been cancelled due to US holidays). I'd be happy to drop into
    >      your meeting to explain the outcome from the council meeting, and answer
    >      any questions you might have, if you think that would be useful. I'm
    >      expecting to have sent you the suggested charter text and some
    >      additional summary from the council discussion before then either way.
    >      Do let me know if that would be useful.
    >
    >      I can only apologise on behalf of the council members for how long this
    >      has taken to schedule and carry out.
    >
    >      Take care,
    >
    >      Amy (rhiaro)
    >
    >

Received on Wednesday, 23 November 2022 18:13:20 UTC