14:59:57 RRSAgent has joined #tt 15:00:01 logging to https://www.w3.org/2025/05/22-tt-irc 15:00:01 RRSAgent, make logs Public 15:00:02 Meeting: Timed Text Working Group Teleconference 15:00:14 agenda: https://github.com/w3c/ttwg/issues/307 15:00:21 Previous meeting: https://www.w3.org/2025/05/08-tt-minutes.html 15:00:28 Present: Nigel 15:00:31 scribe: nigel 15:01:42 Present+ Matt 15:01:56 Present+ Chris_Needham 15:04:57 Present+ Cyril 15:05:25 Topic: This meeting 15:05:50 cpn has joined #tt 15:06:16 scribe+ cpn 15:06:20 Nigel: Today: note publications, DAPT, IMSC 1.3, TPAC planning, and Charter review 15:06:37 ... Anything else? 15:07:06 Matt: User groups for DAPT, people making systems that use it 15:07:39 Topic: Republication of png-hdr-pq WG Note 15:07:50 Nigel: Note published at https://www.w3.org/TR/2025/NOTE-png-hdr-pq-20250518/ and auto-publication on PR merge now enabled. 15:08:21 Nigel: Any comments or questions? 15:08:25 (none) 15:08:27 Topic: Apply streamlined publication to all of Note track documents 15:08:53 Nigel: This has been done, for png-hdr-pq and DAPT Recs, but AFAICT not on DAPT profile registry 15:09:18 ... so it's in progress 15:09:31 Topic: DAPT 15:09:59 Subtopic: Transition DAPT requirements WG draft Note as formal Note 15:10:06 Nigel: dapt-reqs published as WG Note on 12th May at https://www.w3.org/TR/2025/NOTE-dapt-reqs-20250512/ 15:10:31 Nigel: There are no open issues on dapt-reqs at the moment 15:10:41 s/DAPT Recs/dapt-reqs/ 15:11:17 Nigel: Actually, there are two PRs. One is for publishing via GitHub Action, which has been done 15:11:28 ... The other one needs input 15:11:31 -> https://github.com/w3c/dapt-reqs/pull/23 Add links to MAUR requirements w3c/dapt-reqs#23 15:12:10 Nigel: This PR needs a review. It addresses issue #5 that Cyril opened 15:12:11 Present+ Gary 15:13:04 Subtopic: DAPT User Group 15:13:59 Matt: Our workflow requires a post-production script, they come in as Word documents, but we'd like it as DAPT 15:14:16 ... We'd upload them to Limecraft, align them and produce a basic subtitle caption file 15:14:33 ... We'd add in any sound effects, etc. 15:14:54 ... We want them to support DAPT, and they'd like to connect with others working with DAPT 15:15:15 atsushi has joined #tt 15:15:20 Present+ Atsushi 15:15:41 Cyril: Can you clarify what changes Limecraft want in DAPT? 15:16:47 Matt: It's not so much that they want changes. It would be good to have a mechanism to have dialog to ensure that aren't lots of different variants 15:17:11 Cyril: Last time I explained that Netflix intends to release open source software to produce DAPT from TTAL scripts 15:17:32 ... An option could be to use TTAL software to produce DAPT, would that help? 15:17:49 Matt: Yes, but for me it's about opening up the discussion beyond this WG 15:18:00 Cyril: I can see if we have contact with them 15:18:25 Nigel: There may be implementers who aren't W3C participants who want to compare notes. I can think of at least 3 options, 2 of those exist 15:19:10 ... Existing are: the Audio Description Community Group, which could be expanded in scope, or create a new CG. Or if people don't want to come to W3C, it could be in scope for the CCSUBS group 15:19:44 ... Thirdly, we could find a home for a DAPT user group. That might be preferable, so it's clear. There's an IMF user group, and DASH-IF 15:19:50 ... They need funding to operate 15:20:17 ... Any of those could work. I see the need to compare notes on interoperability 15:20:33 Matt: Yes, something beyond sharing the spec, which can raise questions that need addressing 15:21:00 Cyril: At minimum, a mailing list for exchanging bug reports, without fear of sending reports to the spec authors 15:21:23 ... Publish guidelines, similar to what DASH-IF has done. A Community Group could be perfect for that 15:21:52 Nigel: The CG provides a mailing list. Could start there, and see if that works for people 15:22:17 ... It might need a meeting structure, not frequent, every 3 months, but to show that other people are there. 15:22:44 ... How to progress this? I can send messages out, but I wonder what's the most effective way 15:23:09 Matt: I'll be at the CCSUBS meeting in Berlin, could request an hour there? 15:23:53 Cyril: Creating a CG is easy. Could get that ready and create following CCSUBS 15:24:22 Nigel: If we create a new one we should close the existing CG to avoid confusion. Or repurpose the existing one. I don't know if the group can be renamed 15:25:01 Cyril: A timed text scripts CG. Having a CG named audio description might be misleading 15:25:08 Nigel: Yes 15:25:28 ... Let's think about it, and see what we can set up. It's a good idea 15:25:59 Subtopic: Test suite 15:26:22 Nigel: I haven't seen any movement on the issues about tests. I haven't made PRs to add tests 15:26:38 ... Ideally we have someone else look at my analysis to see if it makes sense 15:26:46 ... I can create tests if it looks about right 15:26:54 Cyril: I'll try to do that tomororow 15:27:36 Subtopic: Explicitly permit daptm:represents on tt, body, div, p and span w3c/dapt#294 15:28:43 Nigel: Andreas raised the issue, but he's not here today. If anyone wants to look, it's a minor issue about the data model 15:29:56 Nigel: Anything else on DAPT? 15:29:58 (nothing) 15:30:02 Topic: IMSC 1.3 15:30:12 Subtopic: Liaison response received from EBU 15:30:22 -> Member-only link to liaison response: https://lists.w3.org/Archives/Member/member-tt/2025May/0000.html 15:30:54 Nigel: They asked for DAPT metadata to be allowed in IMSC documents. I don't think anything prevents that 15:31:26 ... Do we need a call out to say something about compatibility with IMSC? 15:31:56 -> https://www.w3.org/TR/ttml-imsc1.2/#foreign-elements IMSC 1.2 Foreign Element and Attributes 15:32:18 Cyril: From IMSC perspective DAPT is a foreign namespace 15:33:13 Nigel: If you want to turn DAPT into captions, to what extend do you preserve existing metadata. A question about good practice 15:33:59 ... By preserving translation information in the player, you can use that to drive spoken subtitles 15:34:12 ... Data that's always been missing so far, but how we have a scheme for describing it 15:34:48 ... DAPT data can tell you if the subtitle text was translated from a different language, so you can use that for spoken subtitles 15:35:16 Cyril: I don't follow 15:36:27 Nigel: If there's markup to show which subtitles are translations, the player, if it wants, can trigger text to speech on those specific subtitles, to get an audible translation 15:37:34 ... I heard this in action in Holland, 10 years ago. They do it using broadcast subtitles, and special set top boxes. A Dutch news programme, in Dutch, cuts to someone speaking in English, and the translation was spoken by the STB. That needs the translation data present 15:37:55 Cyril: I see 15:38:27 Nigel: It would be useful to call out somewhere, whether in DAPT or another document 15:38:38 Cyril: Could be in an appendix 15:38:51 Nigel: Anything else on IMSC? 15:38:55 (nothing) 15:39:03 Topic: TPAC 2025 Planning 15:39:39 Gary: It's the same as previous years, ask how many meeting slots you want, and which groups to have joint meetings with or avoid overlaps with 15:40:06 ... For us this year, it's more complicated as it's in Japan. If members aren't there in person we might not have quorum 15:40:34 Nigel: I don't know about my attendance right now. We have to fill this out in about a month 15:40:57 ... We'll get a lot of Japanese attendees, would be useful to have their input 15:41:30 ... I'd like to be able to discuss with people involved in ARIB. Going there in person might help us resolve the IMSC ARIB issue 15:42:11 ... In terms of other work, the timing might be we can confirm transition of DAPT to Rec in person 15:42:17 ... Could also have a CG meeting 15:42:31 ... I'd hope we can also share progress with MEIG 15:43:13 ... Also Media WG and MSE carrying captions. 15:43:23 Chris: Absolutely, both of those topics would be welcome, 15:43:29 .. happy to organise joint meetings to support that work. 15:43:41 .. In general, what I would suggest is request the timeslots now. It's easier 15:43:54 .. to release unneeded timeslots than to add new ones that you do need, late in the process. 15:44:04 .. Experience last year - MEIG had more agenda requests than I at first expected. 15:44:18 Gary: That makes sense, plan for it to happen, then cancel if it turns out nobody can come. 15:45:20 Nigel: In the past we've created an agenda page. We can do it in the TTWG GitHub, as a wiki or issue or discussion. I can transcribe what we've just said into that tomorrow 15:45:34 ... Which days would we want to meet? 15:46:00 ... There's often pressure to meet at the beginning or end of the week 15:46:25 Chris: I thought similar to last year - MEIG could have a few slots on Thursday, 15:46:40 .. then Thursday for the Media WG, but it ends up being the event team that reconciles all the slots. 15:46:51 .. The other thing to bear in mind is that this year, because there was so much interest 15:47:01 .. in breakout sessions there are more slots during the week dedicated to breakout time, 15:47:16 .. so the format for the Wednesday is a bit different and there are breakouts on the other days too. 15:47:28 .. So there might be less time allocated for groups to fit their sessions in. 15:47:52 .. I think between those groups it might be good to come up with an overall structure of the week, 15:47:59 .. because there will be overlapping participation. 15:48:10 .. The other thing I would suggest is if we want to coordinate joint time with the APA WG. 15:48:26 .. We did a 3-way meeting with APA, MEIG and TTWG the past couple of years which was good 15:48:42 Gary: They were always interesting. Might be easier in November because I don't think it 15:48:56 .. overlaps any other major events especially for likely attendees of our meetings. 15:49:51 Nigel: Could be MEIG Monday, TTWG Tuesday, Media WG Thurdsay, then joint with APA on Friday 15:49:56 Chris: Yes that's the kind of thing I was thinking of 15:50:40 Cyril: I'd prefer having all the timed text related meetings all at the beginning or end of the week, not spread out 15:51:05 ... It's preferable to only come for just a few days rather than the whole week 15:51:17 ... But I'll follow what the group decides 15:51:32 Nigel: Anything else on TPAC? 15:51:38 (nothing) 15:51:54 Topic: AOB - TTWG Charter Review 15:52:25 Nigel: The charter review closed on the 19th. There were no objections, some proposed changes, that Gary and I are happy with 15:52:37 ... It looks like a formality that the group will be rechartered 15:53:09 Atsushi: We're asking reviewers about the proposed changes. We'll proceed with the rechartering after 1 week if there are no comments from reviewers 15:53:29 ... So, to be announced next week 15:57:24 Topic: Meeting close 15:57:28 Nigel: Next meeting is 2025-06-05 15:58:09 .. Thanks everyone [adjourns meeting] 15:58:14 rrsagent, make minutes 15:58:15 I have made the request to generate https://www.w3.org/2025/05/22-tt-minutes.html nigel 16:03:31 Chair: Nigel, Gary 16:03:45 Regrets: Andreas, Pierre 16:04:04 s/DAPT profile registry/TTML profile registry 16:05:46 s/tomororow/tomorrow 16:06:06 s/response:/response 16:06:36 s/but how we have a scheme/but now we have a scheme 16:08:56 s/... Also Media WG and MSE carrying captions./Nigel: Also Media WG and MSE carrying captions. 16:09:32 rrsagent, make minutes 16:09:33 I have made the request to generate https://www.w3.org/2025/05/22-tt-minutes.html nigel 16:10:23 scribeOptions: -final -noEmbedDiagnostics 16:10:27 zakim, end meeting 16:10:27 As of this point the attendees have been Nigel, Matt, Chris_Needham, Cyril, Gary, Atsushi 16:10:29 RRSAgent, please draft minutes v2 16:10:30 I have made the request to generate https://www.w3.org/2025/05/22-tt-minutes.html Zakim 16:10:36 I am happy to have been of service, nigel; please remember to excuse RRSAgent. Goodbye 16:10:37 Zakim has left #tt 16:10:42 rrsagent, excuse us 16:10:42 I see no action items