- From: David Singer <singer@apple.com>
- Date: Tue, 07 Apr 2015 10:22:30 -0700
- To: Nigel Megitt <nigel.megitt@bbc.co.uk>
- Cc: Timed Text Working Group <public-tt@w3.org>
> On Apr 7, 2015, at 4:09 , Nigel Megitt <nigel.megitt@bbc.co.uk> wrote: > > On 02/04/2015 17:58, "David Singer" <singer@apple.com> wrote: > >> I am so sorry, I totally spaced this morning. I meant to call in! >> >> Once we are sure we have both (a) the comments or lack of, from the >> reviewers and (b) a disposition for each comment, I don’t see a problem >> with me or someone writing up a proposed summary for review by the WG and >> presentation to the director. >> >> At the moment, I am more focused on making sure we get the comments. >> >> I meant to encourage everyone to look at the bug tracker for the comments >> they care about, and subscribe to it if they want notifications, also. I >> assume we don’t need the WG mailing list automatically notified on every >> change. >> >> I am aware of the different modus operandi here between the VTT community >> and the TTML community, and I want to make sure we respect both: please >> let me know if you think information is not flowing, and we’ll see what >> we can do to address it. >> >> Yes, a CR transition is formally the WG’s to handle, so the comments, the >> discussion, the proposed resolution, all need to be visible here. >> >> As a joint chair I expect it’s my job to do any bridging here. Let me >> know what you need. The good news is that the data is all there if you >> know where to look. > > Looks to me as though the flow of information, broken down into detailed > steps is like this: > > 1) CG prepares spec editors prepare. They can discuss with the CG if they like, but the editors are members of this WG. > 2) CG sends spec snapshot to WG Editors send. > 3) WG publishes spec snapshot on TR > 4) WG requests review > 5) reviewers respond to WG > 6) WG chair (David) makes sure that all comments are visible to both WG > and CG > > <for each comment:> > 7) members of CG and members of WG (and members of both) discuss the > response, including any proposed spec changes. we’re not proud; we’ll take helpful comments from wherever. ‘Contributions’ in the formal sense, from members of the WG, but I think we’re past that point. > 8) WG prepares draft response to commenter > 9) WG chair (David) makes sure that draft response if visible to both WG > and CG and indeed the commenters and (i think) the public (as if the public cares, sob). > 10) After a suitable review period, WG responds to commenter with response > 11) Commenter has a fixed period to signal disposition to the WG response > </for each comment> > > 12) WG chair (David) or staff contact (Thierry) collates commenter's > dispositions in some suitable document, makes visible to both WG and CG. > 13) WG chair includes report of all dispositions in request to transition > to next stage of maturity. > > > Which I think answers the question about what bridging is needed. > > We're up to step 4 or 5 for some reviewers and step 7 for (at least) i18n. Yes, this is all normal, apart from making sure that the CG is kept aware of matters. > > > >> >> >> >> >>> On Apr 2, 2015, at 8:14 , Nigel Megitt <nigel.megitt@bbc.co.uk> wrote: >>> >>> WebVTT comments >>> >>> tmichel: A couple of months ago I talked to the staff contact >>> of the interaction domain for CSS, >>> ... HTML and i18n, and requested them to ask their WGs to >>> review the WebVTT FPWD. About a >>> ... month later we have received a good set of feedback >>> comments from i18n. We're still missing >>> ... the other three. A few days ago I pinged again Bert and >>> Mike. For CSS, Bert has added the >>> ... review request to the next CSS agenda. For HTML Paul Cotton >>> has followed up on Dave's message. >>> ... So we should be set on those two. For accessibility, I know >>> plh has pinged Judy again a few days >>> ... ago so I hope to get feedback from them but I can't say >>> when. I'd like to get the feedback so we >>> ... can move to CR when ready. >>> ... Another issue is how we handle the comments. Currently >>> they're archived in the CG. So at some >>> ... point we will have to respond to the commenter and get >>> agreement from them so how are we >>> ... going to track that? >>> >>> atai: Because the comments are on the CG bug tracker and the >>> responses are there, should this be >>> ... delegated to one place so that everyone can see them? I >>> don't think it makes sense to have a >>> ... separate tracker for that. >>> >>> tmichel: Okay. At some point we have to demonstrate wide review >>> to the Director, so are you saying >>> ... just look into the archive? >>> >>> atai: So it has to be proved that there has been a review and >>> that the comments have been >>> ... incorporated into the document? >>> >>> tmichel: We should have responded to the comments and got >>> acceptance for whatever action we >>> ... have taken. >>> >>> atai: Does this need to be a formal list of raised issues and >>> how they've been communicated and dealt with? >>> >>> tmichel: Yes, at some point we should have a list of the >>> comments and at least some wording from >>> ... i18n to say that they're satisfied with each comment. That >>> would be good enough. >>> >>> atai: Can this be done in the CG, to ask them to follow a >>> specific procedure to satisfy the process? >>> >>> tmichel: It's not a problem to have it in the CG. We just need >>> to show the evidence to the Director. >>> >>> atai: My only question is how we handle the same spec in >>> parallel in the CG and the WG. Is there a >>> ... communication with the CG about the complete process? >>> >>> tmichel: I think the CG chair is Dave Singer and there's no >>> staff contact. It doesn't matter who >>> ... does the job of reporting. What we need is at least a basic >>> document describing the dispositions. >>> >>> nigel: For IMSC 1 we used the dispositions tracker, but I guess >>> that's not the only way. >>> >>> tmichel: Agreed. We need at least something. >>> >>> atai: Is it possible just to have a spreadsheet referencing the >>> bugs? >>> >>> tmichel: Yes I think that's good enough. >>> >>> atai: And then link to the bug tracker? >>> >>> tmichel: Yes that's good enough. You do need the final >>> agreement from i18n. >>> >>> atai: So we check back with them? >>> >>> tmichel: Yes >>> >>> pal: We just went through that with IMSC 1 - why not go through >>> the same process? >>> >>> tmichel: We certainly can. >>> >>> atai: I see the problem being additional overhead. Of course it >>> worked fine. From my perspective >>> ... we need to check that it's okay for the CG. That's my only >>> concern. >>> >>> pal: But there's only a single consensus group for this work, >>> and it's this group. The other group >>> ... could be a place for the exchange of ideas, but the formal >>> place where the work happens is in >>> ... this group if I'm not mistaken. >>> >>> atai: That's right, but the groups work differently. >>> >>> pal: But as far as W3C is concerned there's only one group and >>> its this one. >>> >>> glenn: Agreed. Managing the comms is the challenge. There have >>> been a lot of cross-postings >>> ... which is useful for allowing everyone to have an input, but >>> I think the formal decision process >>> ... has to happen here. >>> >>> atai: Can we agree on this without Dave Singer being here? >>> ... What would be good would be to have a close cooperation >>> with the CG. >>> >>> nigel: There are a number of people interested in progressing >>> WebVTT who are in both groups >>> ... so I'd encourage those people to take the lead on this. >>> >>> Loretta: I think we'd have a hard time persuading the CG to do >>> the paperwork but any changes to >>> ... the document do need to be agreed by the CG. >>> >>> pal: I think by choosing to put WebVTT on the Rec track that >>> demonstrated a commitment to follow >>> ... the W3C process so that needs to happen. >>> >>> nigel: Can I propose that someone who is a member of the WG >>> takes ownership of the comments >>> ... and documents them, and liaises with the CG to agree all >>> actions and responses in both groups? >>> ... Can anyone volunteer for that? >>> >>> Loretta: Can we pick up on this next week? >>> >>> tmichel: Are you proposing a 'two tick' system where both the >>> CG and the WG approve every >>> ... response to each comment? As well as the original commenter >>> e.g. i18n. >>> >>> nigel: We could do that - I'm not actually bothered about >>> whether the trail of approval between >>> ... CG and WG is on the dispositions document or if its managed >>> separately. But what I do care about >>> ... is that the WG doesn't propose a response that is not >>> acceptable to the CG. >>> >>> tmichel: This is fairly new so we don't have much prior example >>> to work from. >>> >>> nigel: It's a good time to be creative with ideas! >>> ... If you do have any ideas please bring them to the table >>> next Friday. >>> >>> >>> >> >> David Singer >> Manager, Software Standards, Apple Inc. >> > David Singer Manager, Software Standards, Apple Inc.
Received on Tuesday, 7 April 2015 17:23:01 UTC