13:55:25 RRSAgent has joined #tt 13:55:25 logging to http://www.w3.org/2015/08/13-tt-irc 13:55:27 RRSAgent, make logs public 13:55:27 Zakim has joined #tt 13:55:29 Zakim, this will be TTML 13:55:29 I do not see a conference matching that name scheduled within the next hour, trackbot 13:55:30 Meeting: Timed Text Working Group Teleconference 13:55:30 Date: 13 August 2015 13:55:38 chair: nigel 13:55:41 scribe: nigel 13:55:50 Regrets: Frans 13:55:52 Present: nigel 14:02:55 tmichel has joined #tt 14:03:22 mike has joined #tt 14:03:29 pal has joined #tt 14:03:49 atai has joined #tt 14:03:56 Present+ Mike, Pierre, Andreas, Thierry 14:05:24 Topic: This meeting 14:05:41 nigel: Agenda for today has a few action items and maybe TPAC plus the HTMLCue proposal. AOB? 14:05:47 group: No AOB 14:06:17 Present+ Courtney 14:07:02 Topic: Action Items 14:07:15 action-409? 14:07:15 action-409 -- Nigel Megitt to Discuss liaison with unicode with richard ishida -- due 2015-07-09 -- PENDINGREVIEW 14:07:15 http://www.w3.org/AudioVideo/TT/tracker/actions/409 14:07:50 nigel: We sent the message to Mark Davies of Unicode, and haven't had any response yet. 14:07:55 courtney has joined #tt 14:08:24 pal: Something else we can also do is to file an issue on their tracker, or to supplement the existing one with the new text. 14:08:38 ... I'm not sure which one is better. If we haven't heard back in a week maybe we can do that. 14:09:01 nigel: Okay that's a good idea, let's revisit that if we've had no response by next week. 14:09:18 close action-409 14:09:18 Closed action-409. 14:10:06 Action: nigel Add Unicode liaison to weekly meeting agenda 14:10:06 Created ACTION-414 - Add unicode liaison to weekly meeting agenda [on Nigel Megitt - due 2015-08-20]. 14:12:13 Topic: TPAC 2015 14:12:18 action-412? 14:12:18 action-412 -- Nigel Megitt to Add draft topics for tpac to wiki page as discussed today and last week -- due 2015-08-13 -- PENDINGREVIEW 14:12:18 http://www.w3.org/AudioVideo/TT/tracker/actions/412 14:12:32 https://www.w3.org/wiki/TimedText/tpac2015 14:13:08 nigel: I've done this action - the wiki page holds the topics we have so far. I've also added my 14:13:16 ... arrival and departure times and hotel. 14:13:26 ... If you're attending please could you add your name to the page? 14:13:43 close action-412 14:13:43 Closed action-412. 14:14:09 nigel: I've done that, so closing it. But this is an opportunity for more topics. 14:14:21 Courtney: We're close to having a 1st draft of the mapping document and would like to be able 14:14:34 ... to discuss it in person at TPAC - it might take a few hours to half a day. 14:16:05 nigel: Okay, I and I guess David Singer will have to work out how to allocate the time over the 14:16:17 ... two days to all the topics we have. I'm not sure what David's plans are but if he's able to be 14:16:34 ... present then we'll need to hook up to work out how to share the time. 14:17:33 nigel: Any other requests? 14:17:47 tmichel: I guess the Timed Text working charter is linked to the IMSC 2 deliverable. The current 14:18:01 ... charter doesn't mention IMSC version numbers, so we'll need to discuss that. I don't know 14:18:10 ... if that needs a charter extension - it's an open issue. 14:18:44 nigel: Thanks, yes that's a good point. I can't remember the discussions we had when we chose 14:19:01 ... to split IMSC up into v1 and v2 but the result was that we didn't make any charter changes. 14:19:20 tmichel: We can open up discussions with plh when he's back from vacation, in early September 14:20:54 nigel: If there are no other proposals right now for the agenda, just a reminder to register 14:20:59 ... and book hotels etc. 14:21:32 tmichel: If you register before Sep 26 with the hotels then you get the discount rate. 14:21:59 tmichel: And the advance registration for TPAC closes on Oct 7 - until then its 85USD per day 14:22:08 ... and after that 170USD. This is all mentioned on the wiki page. 14:22:12 http://www.w3.org/2015/10/TPAC/#registration 14:23:14 Topic: TTML2 Editorial Actions 14:23:19 action-413? 14:23:19 action-413 -- Nigel Megitt to Update the ednotes.xlsx to remove/mark up completed work -- due 2015-08-13 -- PENDINGREVIEW 14:23:19 http://www.w3.org/AudioVideo/TT/tracker/actions/413 14:23:48 nigel: I did check that this is up to date - completed work is marked as C, open as O. 14:24:01 close action-413 14:24:01 Closed action-413. 14:24:17 nigel: I also uploaded a README.txt document for anyone wanting to make editorial changes. 14:25:28 nigel: I also pushed a couple of examples, for padding and textAlign on span elements, but 14:25:40 ... as far as I can see Glenn hasn't yet regenerated the ED and pushed it, so you can't see them yet. 14:26:24 nigel: If you want to see them then pull from Mercurial, run ant to build the HTML and have a look. 14:26:33 ... Let me know if you have any difficulties doing that. 14:27:23 Topic: HTMLCue Proposal 14:27:30 action-406? 14:27:30 action-406 -- Andreas Tai to Follow on creating a strawman re: html cue extension specification -- due 2015-07-16 -- OPEN 14:27:30 http://www.w3.org/AudioVideo/TT/tracker/actions/406 14:27:58 atai: I sent round the proposal and there's already been some discussion. The idea is to 14:28:14 ... draft the proposal and then approach developers etc and see if there's any feedback. 14:28:28 ... I would propose to use this and think about to whom we want to contribute the summary, 14:28:34 ... and inform about our planned activity. 14:28:46 https://lists.w3.org/Archives/Public/public-tt/2015Aug/0009.html 14:29:33 atai: The proposal is very straightforward, which is what we discussed, that we want an 14:29:47 ... implementation of the Text Track Cue interface, with HTML data instead of VTT. 14:29:57 ... And I added some notes to explain why we think that's useful, to address the feedback 14:30:13 ... from previous discussions, that we should highlight why we need an extension, and 14:30:22 ... possibly already have a proposal for how this works out. 14:30:54 ... It would allow more than one format to use the Text Track Cue interface in browsers. The 14:31:18 ... solution would be to map everything to VttCues, to define a cue type for every format, or 14:31:32 ... to define a generic cue. I describe why a generic HTML cue would be a good solution, 14:31:47 ... coming from the feedback from Opera, Glenn and our discussions. The cue data is HTML 14:31:54 ... and it's easy for browsers to render. 14:33:04 nigel: I'm trying to think ahead for what the spec might look like. We have the idea of a cue 14:33:21 ... containing HTML. Then there would be some kind of handler for an onenter event - that would 14:33:37 ... put the HTML payload of the cue somewhere - how would we specify where? 14:33:44 ... And would it append or replace? 14:33:58 atai: I think that's the proposal, and your questions are the task to solve. 14:34:27 ... I want to ask everyone else if they think that's what we discussed and how realistic it is to 14:34:33 ... implement it in browsers. 14:34:55 pal: I don't have an opinion on that other than that to date there's no generic way in HTML5 14:35:09 ... to assign HTML to time events, which is essentially the proposal. This has been discussed 14:35:29 ... in the past and I think Silvia stated factually the feedback in the past. 14:35:58 nigel: My reading of Silvia's emails is that the world has moved on and it may be worth raising 14:36:12 ... this proposal once again even if it was previously considered and not taken any further. 14:36:47 atai: I think it's worth taking this and starting some quiet discussions. 14:36:59 pal: If I'm not mistaken Glenn already proposed a patch on webkit. 14:37:08 atai: I think it was a TTML implementation proposal. 14:37:23 pal: You might be right - in that case it would be worth highlighting that this is something different. 14:38:01 atai: The problem with that the browser devs didn't like XML. 14:38:14 nigel: Specifically it was the Google blink community. 14:38:33 atai: I think the proposal from Giuseppe Pascale was to raise this and see what feedback we get. 14:38:44 pal: Maybe an initial channel to try would be the Web & TV IG. 14:39:01 atai: That may be one, yes. We can send something officially across groups. Giuseppe also 14:39:17 ... proposed contacting the WHATWG. Those would be official channels. We can also make 14:39:35 ... more informal contact. 14:40:15 atai: Giuseppe mentioned WHATWG because a lot of developers are on the reflector for that group. 14:40:56 tmichel: I can investigate the options for contacting WHATWG. 14:41:15 atai: We also need to take be aware that we will need a mapping document from TTML to 14:41:32 ... HTML. We would also need to be willing to draft such a mapping in case an HTMLCue were 14:41:37 ... accepted by the other groups. 14:42:09 nigel: There was also a question on the reflector about implementation in browsers and if 14:42:21 ... anyone could do that. Can anyone respond to that question here? 14:42:37 atai: No, but I think it's an interesting observation that even VTTCue is implemented by 14:42:52 ... mapping VTT into HTML. I think there are open source or other players that take TTML and 14:43:09 ... populate the cue with very HTML-like data. That could be worth investigating if we can 14:43:17 ... point to implementations that use that strategy. 14:44:47 nigel: I don't think it's a secret that EBU has a dash.js fork on github that takes exactly this 14:45:12 ... approach, misusing VTTCue because that's the only thing that's supported by browsers 14:45:21 atai_ has joined #tt 14:45:31 ... and not showing the track, but using the timing of the cues to fire onenter and onexit 14:45:46 ... events whose handlers simply attach or remove HTML from a target element. 14:47:03 nigel: In terms of taking this forward, I can as chair take on the liaison role for sending the 14:47:37 ... draft text to Web & TV IG and to the HTML WG and possibly WHATWG. The first question 14:47:54 ... is if the draft text is okay to send now? 14:48:14 pal: Have you considered thinking what the next steps would be, and for example pointing 14:49:05 ... to those areas in the code that convert VTT into HTML? 14:49:26 atai: The idea from Opera would be first to test the idea and then go into detail later. 14:49:38 ... What we discussed in the last face to face meeting is if we want to initiate a joint meeting 14:51:54 ... with HTML WG at TPAC. 14:52:14 nigel: For me I think we could add a little bit more detail such as pointing to existing code 14:52:32 ... or addressing some of the questions like what action happens on the entry and exit events 14:52:41 ... exactly, and how the target element is identified. 14:54:41 nigel: I'm interested to know what Courtney's feeling on this would be? 14:54:50 Courtney: I'll take a look at this and talk to the webkit guys about it. 14:55:17 mike: I haven't studied this either. I'm not familiar with this area though. 14:56:43 nigel: So turning this into actions, we have the one Courtney just mentioned; I'm happy to try 14:56:56 ... to edit the draft to include some of the areas that we know need addressing but haven't 14:57:12 ... wanted to spend time on until we know the idea might be adopted, and Thierry will 14:57:21 ... investigate the possibility of contacting WHATWG. 14:57:53 nigel: I'd like to revisit this next week following these actions and make some concrete plans 14:58:01 ... to contact people e.g. Web & TV IG etc. 14:59:06 nigel: On a possibly related subject is it worth a quick mention of the TTML WebVTT mapping 14:59:08 ... work? 14:59:22 Courtney: Sure. Andreas, Loretta and I have made good progress on it and hopefully can share 14:59:46 ... the draft with the group in the next few weeks. We've defined sections on layout, style and timing, 15:00:11 ... and a TTML profile that matches the WebVTT feature set. 15:00:33 had to drop for another call 15:01:38 nigel: We're out of time so I'll adjourn the meeting - if we can come back with those HTMLCue 15:01:51 ... actions next week that would be great. Thanks very much everyone. [adjourns meeting] 15:01:57 rrsagent, make logs public 15:02:00 rrsagent, draft minutes 15:02:00 I have made the request to generate http://www.w3.org/2015/08/13-tt-minutes.html nigel 15:03:26 s/Davies/Davis 15:05:37 s/But this/But this agenda item 15:06:09 s/working charter/Working Group charter 15:07:01 s/USD/ USD/g 15:07:45 close action-406 15:07:45 Closed action-406. 15:08:48 s/solution would be/solutions could be 15:10:09 s/with that the/with that was that the 15:13:08 rrsagent, draft minutes 15:13:08 I have made the request to generate http://www.w3.org/2015/08/13-tt-minutes.html nigel 15:16:08 ScribeOptions: -final -noEmbedDiagnostics 15:16:09 rrsagent, draft minutes 15:16:09 I have made the request to generate http://www.w3.org/2015/08/13-tt-minutes.html nigel 16:02:13 Zakim has left #tt 16:20:52 nigel_ has joined #tt