14:00:41 RRSAgent has joined #tt 14:00:41 logging to http://www.w3.org/2016/07/21-tt-irc 14:00:43 RRSAgent, make logs public 14:00:44 Zakim has joined #tt 14:00:45 Zakim, this will be TTML 14:00:45 ok, trackbot 14:00:46 Meeting: Timed Text Working Group Teleconference 14:00:47 Date: 21 July 2016 14:00:57 mike has joined #tt 14:00:57 Present: Nigel 14:01:01 chair: Nigel 14:01:04 scribe: nigel 14:01:49 Regrets: Thierry, Frans, Pierre, Glenn 14:05:38 Present+ Mike, Andreas 14:06:24 Topic: This meeting 14:06:40 nigel: We are low on numbers today for various reasons. What are we quorate to discuss? 14:07:36 mike: Perhaps a brief update on the IANA registration 14:09:10 Andreas: TPAC 14:09:48 Present+ David_Ronca 14:10:06 Andreas: Also the lineHeight, background height issue and the need for a solution. 14:10:39 nigel: Okay, I'd also maybe like to come back to the Safe Crop Area proposal. 14:11:12 Topic: TPAC 2016 14:11:33 nigel: Note that the hotel closest to the venue is getting low on availability: I just tried booking and failed! 14:11:43 ... So it's worth booking accommodation soon. 14:12:14 Andreas: Regarding the conflict (for some) with the Web & TV IG meeting on the Monday 14:12:25 ... and the TTWG meeting on the Monday and Tuesday. There is some overlap of interest 14:12:38 ... so this is unfortunate. It may be difficult to change this, but we should ask the Web & TV IG 14:12:48 ... at least about their agenda so we can have a basis for setting our agenda topics. 14:12:59 ... Also we thought about a combined meeting with the Web & TV IG. I'm not sure if this 14:13:09 ... is still of interest for others. There's clearly an overlap in what they are working on, 14:13:20 ... and also the members are I think interested in progress on the subtitle front. It could 14:13:31 ... be a possibility to discuss the issues about WebVTT and TTML and the adoption of 14:13:33 ... IMSC. 14:14:46 nigel: For the last few years we have provided an update to the Web & TV IG who are 14:14:57 ... indeed interested in our progress, so I think it makes sense to ask for some joint time 14:15:07 ... with them so that we can do the same this year. I'll give myself an action. 14:15:25 Action: nigel Contact the chair of the Web & TV IG to ask about schedule and joint meeting time. 14:15:25 Created ACTION-475 - Contact the chair of the web & tv ig to ask about schedule and joint meeting time. [on Nigel Megitt - due 2016-07-28]. 14:15:50 mike: That's a good idea - I won't be there by the way. 14:16:06 davidr: Dae will be there for Netflix - I don't really have an opinion on it. 14:16:32 Topic: Profiles registry and TTML IANA registration 14:17:02 mike: If you go to the interaction with IANA link, there's the old one a year or two ago, 14:17:16 ... but nothing in there that indicates that we have any open engagement with IANA. I 14:17:57 ... thought plh had done that but it looks like not. We've had no response from IANA either. 14:19:46 nigel: Okay, it looks like something's "stuck". I have prompted plh but it's unclear what's happening. 14:19:51 ... Ideally he'd be on the call... 14:21:07 ... I've sent him a message to see if he is available to join. If he can let's come back to this. 14:22:29 Topic: TTML issues 14:22:30 irc_w3_org has joined #tt 14:22:48 nigel: Andreas, I think you wanted to mention lineHeight etc? 14:23:06 andreas: We discussed this for 3 weeks with a lot of discussion and comments, then 14:23:15 ... other discussions took over. I think it's important that we look for a solution because 14:23:26 ... HbbTV has now published an update to their spec and they have some normative 14:23:43 ... text that clarifies this issue and I think they also are waiting for clear guidance from 14:24:00 ... W3C or EBU that they could reference, but because they have to publish sooner they 14:24:09 ... needed to include some text. I think it's important for us to check this change and 14:24:18 ... also to work on the solution and decide if there's an errata needed for TTML1 or an 14:24:28 ... update to IMSC 1 and how in general this should be solved for TTML2. 14:24:46 ... It may also be important but out of scope for W3C that HbbTV starts a liaison with W3C 14:24:48 ... on this issue. 14:25:50 nigel: Can you recall what HbbTV did? 14:26:21 andreas: They made a lightweight addition, based on a request that a transparent gap 14:26:37 ... should not appear between line backgrounds. So the solution is that for lineHeight <= 125% 14:26:57 ... when the background is set on consecutive rows there should not be a gap between 14:27:08 ... the line backgrounds, without explaining how the renderer should make that happen. 14:27:23 ... The solution for how this conceptually and theoretically would be done in TTML is still 14:27:29 ... outstanding and the task of this group. 14:28:16 nigel: Even this group may not be the best place to fix it - sometimes we have to do so 14:28:29 ... when nobody else will fix it, but it seems like it should be fixed in CSS and then we should 14:28:36 ... align with that, otherwise it's a bit backwards. 14:29:09 andreas: It would be good to have a pragmatic solution as soon as possible so that the 14:29:25 ... different standards are compatible with us, e.g. by an errata with TTML1. Going back 14:29:35 ... to the discussion we had there was I think not a really clear view of the group how this 14:29:45 ... case should be handled so there seems to be a gap in TTML1 how it should be handled. 14:30:00 ... If this is the case that there is some room for interpretation then there can be an errata 14:30:11 ... in TTML1. If this is not the case then it would be good to have a solution in the next 14:30:21 ... upcoming spec, which could be an update to IMSC 1. 14:31:34 davidr: I've seen that separation between backgrounds of multiple lines of text is a problem 14:31:57 ... on iOS with WebVTT. 14:32:10 ... For our TTML clients we don't specify backgrounds. 14:32:24 nigel: Our BBC subtitle guidelines take the opposite approach - a black background is 14:32:25 plh has joined #tt 14:32:38 ... required and there must not be gaps between lines, so this is a real issue for us. 14:33:08 present+ 14:33:09 nigel: Thank you for raising the profile of this once again. 14:33:14 Present+ Philippe 14:33:43 Topic: Profiles registry and TTML IANA registration 14:34:35 plh: I started the review - IETF says we have to start the review on their types mailing list 14:34:49 ... first. You have to do a complete submission every time. 14:34:51 s/types/TIES 14:34:59 ... Once we're done we send the request to IANA. 14:35:45 nigel: Is the IETF review period open ended or will it close? 14:35:58 plh: It is open ended; it's been more than a month but I would think that is enough. 14:36:10 ... I would not worry about the time period. But I would ask if we addressed the comments? 14:36:28 mike: The status is: we received 2 comments from W3S; one was rejected by the WG 14:36:52 plh: I'm not aware of W3S comments on the IETF list. 14:37:00 mike: I'm not aware of any comments from the IETF list. 14:37:56 ... There was a comment from Eric Prudeaux and Paul Libbrecht who posted to the public TTWG reflector 14:38:02 ... and those are the only two comments. 14:38:21 plh: [looks on the IETF list] If the comments are not on the IETF list then they are not a concern. 14:38:51 https://lists.w3.org/Archives/Public/public-tt/2016Jun/0012.html 14:39:06 ... That's my request on June 16, which I sent to the IANA media types list, as required. 14:39:30 https://lists.w3.org/Archives/Public/public-tt/2016Jun/0014.html 14:39:40 https://lists.w3.org/Archives/Public/public-tt/2016Jun/0015.html 14:39:55 plh: So one comment was rejected? 14:40:16 mike: We decided not to proceed with the Mac clipboard type request. 14:40:17 answer to Paul: https://lists.w3.org/Archives/Public/public-tt/2016Jun/0018.html 14:40:48 mike: Nobody expressed any interest or practice that we could codify. 14:40:57 plh: So did you do what is needed to address the comments? 14:41:10 mike: Eric's comment was editorial, but I didn't want to edit the document until the review 14:41:13 ... period is over. 14:41:26 plh: At this point I need the document to be updated so I can submit the registration to 14:41:29 ... IANA itself. 14:42:10 mike: Okay I'll make the changes via issues and PRs and send you an email when I'm done. 14:42:24 plh: Then let me know when you're done and email me until I've sent it to IANA! 14:42:41 plh: The form I will fill is at: 14:42:48 http://www.iana.org/form/media-types 14:43:10 plh: Sometimes the hardest question to answer is "encoding consideration" so if you think 14:43:17 ... it is perfectly answered in your document then you're fine. 14:43:24 "Encoding Considerations" 14:43:36 mike: All the comments are on the note not the media type registration. We didn't touch 14:43:41 ... that aside from adding a codecs parameter. 14:44:03 mike: I can't understand why the same answer would be a concern now when it wasn't before. 14:44:20 plh: Understood. When I send the form to IANA then, according to my most recent experience, 14:44:27 ... they're pretty fast in processing it. 14:44:45 mike: I understand the steps - I'll make the edits and send you an email later. 14:44:50 plh: Thank you. 14:44:55 nigel: Thanks both - great to unlock that. 14:46:21 Topic: TTML issues 14:46:35 nigel: I'd like to come back the BBC's Safe Crop Area proposal - noting that the people 14:46:50 ... present today are a mutually exclusive set relative to those present last week when we 14:46:53 ... discussed it. 14:47:11 nigel: I don't know if anyone has had a chance to look through it. Mike, thank you for 14:47:17 ... sending through the links to smpte documents. 14:49:59 ... I think they help to explain how the problem arises but do not solve it. 14:50:23 nigel: [outlines problem of trying to maintain positional registration and complete display] 14:50:37 mike: There's both a display problem and an authoring problem. 14:51:03 ... There used to be something like a safe area in the video, not directly solving the 14:51:42 ... problem for subtitles and captions. The active format descriptor and bar data help 14:51:58 ... to describe the status for video, and SMPTE is in the process of updating those. 14:52:09 ... It helps solve the video problem and may provide a framework for how we want to fix 14:52:26 ... it for TTML. There's also ISO/IEC 14496-30 corrigendum 1, which tries to provide 14:52:35 ... some signalling outside the document, which we should look at. 14:53:15 nigel: It was unclear to me what that corrigendum actually means for the decoder. 14:54:23 mike: It's about the aspect ratio of the TTML and the active format descriptors describe 14:55:54 ... how to ignore deliberately black areas even though they're encoded in the video. 14:57:16 nigel: What this proposal does is to defer some of the solution space but provide extra 15:00:07 nigel: information to allow decoders to do a better job. 15:01:03 davidr: At this point we don't deal with the use case of subtitle registration differing from how it is authored. 15:02:25 nigel: Okay, well I encourage everyone to look at the proposal document 15:02:26 ... https://lists.w3.org/Archives/Public/public-tt/2016Jun/0003.html 15:02:40 ... direct link to PDF: https://lists.w3.org/Archives/Public/public-tt/2016Jun/att-0003/Subtitle_Safe_Crop_Area_BBC_submission_2016-06-09.pdf 15:02:57 ... If there are no adverse comments then I intend to raise this as an issue and possibly a PR on 15:03:11 ... IMSC 1 for a future update, because that is where it would be most useful for uptake. 15:04:06 nigel: Thanks all [adjourns meeting] 15:04:13 rrsagent, generate minutes 15:04:13 I have made the request to generate http://www.w3.org/2016/07/21-tt-minutes.html nigel 15:05:21 s/Topic: TTML issues/Topic: TTML issues (continued) 15:05:37 s/Topic: Profiles registry and TTML IANA registration/Profiles registry and TTML IANA registration (continued) 15:08:34 s/Prudeaux/Prud'hommeaux/g 15:09:59 rrsagent, generate minutes 15:09:59 I have made the request to generate http://www.w3.org/2016/07/21-tt-minutes.html nigel 15:11:30 s/Profiles registry and TTML IANA registration (continued)/Topic: Profiles registry and TTML IANA registration (continued) 15:11:36 rrsagent, generate minutes 15:11:36 I have made the request to generate http://www.w3.org/2016/07/21-tt-minutes.html nigel 15:12:10 ScribeOptions: -final -noEmbedDiagnostics 15:12:11 rrsagent, generate minutes 15:12:11 I have made the request to generate http://www.w3.org/2016/07/21-tt-minutes.html nigel 17:01:11 Zakim has left #tt