15:02:34 RRSAgent has joined #tt 15:02:34 logging to https://www.w3.org/2022/10/13-tt-irc 15:02:36 RRSAgent, make logs Public 15:02:38 Meeting: Timed Text Working Group Teleconference 15:02:45 Present: Andreas, Hew, Nigel 15:02:48 scribe: nigel 15:02:52 Chair: Nigel 15:02:55 Regrets: Gary 15:03:06 Agenda: https://github.com/w3c/ttwg/issues/229 15:03:34 Previous meeting: https://www.w3.org/2022/10/06-tt-minutes.html 15:04:25 Reminder that the meeting is on Zoom today... 15:04:41 Cyril has joined #Tt 15:05:38 Present+ Pierre 15:06:16 atai has joined #tt 15:06:24 Present+ Atsushi 15:07:01 Topic: This meeting 15:07:24 Nigel: Today we have DAPT issues, IMSC-HRM (one pull request/issue) 15:07:34 .. and any update on the charter FO. 15:07:45 .. Is there any other business, or anything to make sure we cover? 15:08:16 Hew: I can give some feedback on DAPT-REQs 15:08:28 Nigel: That'd be great, thank you 15:08:38 group: no other business 15:09:06 Topic: IMSC-HRM - Coalesce empty ISDs into non-empty ISDs w3c/imsc-hrm#50 15:09:07 https://github.com/w3c/imsc-hrm/pull/50 : Coalesce empty ISDs into non-empty ISDs 15:09:18 github: https://github.com/w3c/imsc-hrm/pull/50 15:10:47 Nigel: This PR is primarily to deal with the clears and short gaps between ISDs - is that right? 15:11:03 Pierre: We're trying to address the industry practice of 2 frame gaps between successive 15:11:05 .. subtitles or captions. 15:11:15 .. Today the HRM imposes a very high cost to those short gaps. 15:11:19 .. This PR addresses that. 15:11:46 .. I think we're down to just one question: 15:11:53 .. Should there be any cost to displaying an empty ISD? 15:12:02 .. i.e. painting it and presenting it 15:12:16 .. I am arguing that because the model assumes that subtitles are painted and presented on 15:12:31 .. a graphics plane that is then overlaid on to the related video object, then displaying nothing 15:12:40 .. costs nothing because you don't display that pane altogether. 15:12:44 Andreas: Makes sense to me. 15:12:55 Pierre: Another goal of this PR is to make as little change as possible. 15:13:11 .. That's where my mind is, why I'm arguing that there is no cost with presenting and painting empty ISDs. 15:13:20 q+ 15:13:33 Andreas: As I understand there's no algorithm that would calculate a clearing action, right? 15:13:51 .. What it means to have an empty ISD that follows an ISD with content, that would then be a clearing action. 15:14:05 Pierre: My suggestion is that clearing happens only in the buffers where things are going to be painted. 15:14:16 .. Before drawing a new non-empty ISD you have to clear the back buffer, and that has a cost. 15:14:31 .. You first clear that, then draw the non-empty ISD, then when the presentation time comes in you 15:14:36 .. flip the buffer and display it. 15:14:54 .. If you clear it before then there's no cost because you just don't display the buffer. 15:15:15 .. Clearly the "clear" is related to the cost of clearing the buffer onto which an ISD is painted. 15:15:39 .. What I'm suggesting is that in fact there is no need to paint empty ISDs so there's no cost to not displaying 15:15:43 .. the graphics plane, period. 15:15:46 q? 15:16:14 Nigel: That's an argument I hadn't understood before, so good to know. 15:16:15 ack n 15:16:31 Pierre: By the way I'm just trying to come up with something that requires as little change as possible 15:16:44 .. to the model. I hadn't thought exactly this 2 weeks ago for instance. 15:17:15 Nigel: Good to clarify that you're not saying that the cost of CLEAR is zero. 15:17:28 Pierre: Yes, the cost of CLEAR is not zero, but there's nothing to clear for empty ISD. 15:18:53 Nigel: Then the question is if we always need at least one buffer to be composited, and 15:19:05 .. therefore that we need to count the cost of clearing a buffer at least once for each clear. 15:19:14 .. Or can we just say "stop compositing" at no cost. 15:19:30 .. I think the conservative approach is to say that we assume there is always exactly one buffer 15:19:45 .. being composited, and therefore we need to count the cost of preparing an empty buffer if there is 15:19:53 .. to be an empty ISD after the current one. 15:20:12 q+ 15:20:24 Pierre: Within the model that's not possible with only 2 buffers. 15:20:45 .. If you clear the back buffer for the empty ISD and then present it then you have to wait 15:20:59 .. for the empty ISD to be presented which defeats the purpose. You would need a third buffer. 15:21:14 .. What I have in mind is that the screen is being refreshed by the video. 15:21:26 .. There's no cost to not painting on the next refresh. Not drawing an overlay is a no-op. 15:21:55 .. When the next frame comes in you draw neither of the two buffers. 15:21:57 ack a 15:22:10 Andreas: The HRM is a theoretical construct to calculate presentation cost of ISDs. 15:22:21 .. That doesn't mean that is how it is implemented in practice. 15:22:37 .. You could view it like you do but of course in practice it could be like Nigel says when empty ISDs come in. 15:22:49 .. Follow a logic that is not yet implemented in the HRM. 15:22:55 Pierre: Sure, exactly, I agree. 15:23:06 .. I think what we're trying to do here is to tease out where the main complexity is, how it scales. 15:23:24 .. I think I'm arguing, I've convinced myself, in general, you could have a cost to not displaying pixels, 15:23:40 .. you could have a third buffer, maybe your implementation is on still images and clearing has a cost, 15:23:55 .. but in the case of video the cost of not displaying an ISD is not substantial and can therefore be safely 15:23:56 .. ignored. 15:24:59 Nigel: Sudden moment of clarity for me that the current pre-PR HRM is what you have to end up with 15:25:03 .. if you have a 2 buffer model. 15:26:39 .. I think what we're discussing here is that the idea is the implementation flips from compositing 15:26:44 .. buffer A to B to A to B etc 15:26:55 .. but in my mind I'd imagined that there was a compositing plane C 15:27:14 .. and the task is to blit A to C, B to C, A to C etc which would have a very different set of constraints. 15:27:18 .. I'm not sure which it is right now. 15:27:32 Pierre: The presentation buffer Pn-1 is directly connected to the display in figure 2. 15:28:24 .. This was designed with a simple TV architecture in mind, where one or the other is displayed on screen, 15:28:31 .. or read directly by the display circuitry. 15:29:14 Nigel: I see what you mean about fig 2. 15:29:51 .. But again, fig 2 has no switch about whether the presentation buffers are connected to the display or not. 15:30:02 Pierre: We could add that to the model, that with empty ISDs nothing is displayed. 15:30:40 .. The PR has text that goes beneath that figure, but we could modify the figure to make it clearer. 15:32:11 Nigel: Yes. I'm just wondering which is preferable, to insert a switch to the display 15:32:21 .. or to posit a 3rd buffer. 15:32:31 Andreas: A new buffer would add more complexity to the model, right? 15:32:38 Nigel: I don't really think so. 15:32:55 Andreas: For me the question is if for some implementations there would be some cost when an 15:33:07 .. empty ISD is going through the change, and if this cost is high enough to be taken into account 15:33:13 .. for the calculation of complexity. 15:33:16 Pierre: Exactly. 15:33:27 Andreas: The cost is negligible and not go into the calculation. 15:34:35 Nigel: I don't think that introducing the third buffer and counting the cost of a CLEAR would not 15:35:16 .. fail anything that passes today, but there are some documents that could potentially fail it, 15:35:20 .. and we should catch those. 15:35:31 Pierre: Concerned about the editing complexity that would introduce. 15:35:48 Nigel: I haven't done the exercise, it seems like it _should_ be simple, but maybe not. 15:36:01 Pierre: If you want to try making the edit, go ahead. I'm not sure what it will look like. 15:36:11 .. You need to introduce the cost of drawing into the back buffer. 15:36:31 Nigel: I would have to make the assumption of a cost-free blit into the third, composition buffer. 15:36:47 Pierre: If you want to give it a shot, as we pointed out, adding the cost of a clear for an empty ISD 15:36:59 .. will land up with results that are closer to what we have today. It's not going to 15:37:09 .. invalidate documents that are valid today. That's not my concern. 15:37:17 q+ 15:37:22 .. It's updating the text in a way that does not cause more trouble than it solves. 15:37:36 .. I'm happy to modify figures - I have the master files, though they may be checked in. 15:37:38 ack at 15:37:52 Andreas: I want to echo a bit Pierre's concern more about the complexity of the specification and 15:37:59 .. unexpected consequences of adding a new concept. 15:38:12 .. The issue is that the whole model is already not so easy to get, from non experts 15:38:17 .. or people who really need to implement it. 15:38:25 .. Adding a new component just increases the complexity of the model. 15:38:31 .. From an understanding point of view. 15:38:44 .. When we consider adding it, we should compare the benefit of adding it to the 15:39:01 .. benefit we get from getting closer to reality for this particular case. 15:39:13 .. We heard that even if we add a component it would not really change the results very much. 15:40:00 Nigel: Either proposal is a change to the model. 15:40:05 .. I'm concerned about clarity of that change. 15:40:14 .. The risk is that it's so subtle that it is not actually noticed properly. 15:41:30 .. I'm not sure if the mathematical results would be identical. 15:41:41 Pierre: They would not be because you'd count one extra CLEAR per sequence of ISDs. 15:41:53 .. Neither proposal would invalidate a set of valid documents, which they don't want to do. 15:42:03 s/they don't/we don't 15:42:49 Nigel: I'm interested to know what is the edge case between the two ideas where 15:43:00 .. one would validate a document and the other would say it is invalid. 15:43:26 Pierre: The cost of that extra clear would be included in the cost of painting the following non-empty ISD, right? 15:43:29 Nigel: I think so. 15:43:33 Pierre: Let me try something. 15:43:43 .. Scenario: non-empty ISD, empty ISD, non-empty ISD 15:44:08 .. The model limits the complexity by saying painting the second non-empty ISD takes some time 15:44:18 .. and the system has some time to do that, and what we're saying now is that the 15:44:32 .. time available to paint the second non-empty ISD is the time between the presentation time of 15:44:56 .. difficult to say without graphics on screen - [shares screen] 15:50:54 .. [discussion of 2 buffer model plus 3 buffer model by looking and pointing at the rendering time figure] 15:51:14 Pierre: We could schedule some time for a call next week? 15:51:32 Nigel: That's a good shout - I should be able to do that. 15:51:43 Pierre: I'd really like to get to the point where we ask for feedback, and 15:51:52 .. we need to complete this issue before we do so. 15:51:56 Nigel: Yes 15:52:20 Pierre: I am available at the same time next week 15:52:25 Nigel: Me too 15:52:41 SUMMARY: Discussions to continue offline and possibly in additional call 15:54:05 Topic: DAPT-REQs 15:54:36 Hew: Our AD operations and product owner reviewed DAPT-REQs and felt 15:54:47 .. that it captured their current workflows and that there would be no problem 15:54:52 .. capturing what they do, within it. 15:55:02 .. They raised possible future questions. 15:55:21 .. For example richer pan options if you want to get creative in e.g. 5.1 audio or other distribution approaches. 15:55:37 .. They wondered about extensible synth voice options to include tone. 15:55:44 Nigel: Emotional tone maybe? Pitch is already available. 15:55:58 Hew: Yes, that could well be what they meant. 15:56:12 .. If you think about dramatic AD particularly they try to carry the tension or emotion of the programmes. 15:56:15 s/s// 15:56:27 Hew: That was it. 15:56:39 Nigel: Great, thank you, always very helpful to get reviews. 15:56:49 Topic: Charter FO Status update. 15:57:09 Nigel: As discussed last week I submitted the Chair's report, and I got an acknowledgement back. 15:57:20 .. I expect the Chair's report to be appended to the document. 15:57:26 .. Atushi, anything to add? 15:57:32 s/Atushi/Atsushi 15:57:51 Atsushi: The report will be made available to the FO Council in the next couple of days. 15:58:16 .. Stepping down from Council is ongoing - I am not sure when they will actually start to discuss it. 15:58:33 .. I believe they have 2.5 weeks for stepping down prior to Team report so they should start shortly. 15:58:51 Nigel: Thanks for the update. Any questions? 15:59:03 .. I see nobody on the queue. 15:59:23 .. Q from me: did the Objectors respond to the draft Team report? 15:59:39 Atsushi: I have not seen a response. WG participants can see the updated report at the same URL. 15:59:45 Nigel: Thank you. 15:59:56 Topic: Meeting close 16:00:19 Nigel: Thanks everyone! [adjourns meeting] 16:00:31 rrsagent, make minutes 16:00:31 I have made the request to generate https://www.w3.org/2022/10/13-tt-minutes.html nigel 16:08:54 s/tenion or emotion of the programmes/tension or emotion of the programme 16:13:09 s/I don't think that introducing the third buffer/I think that introducing the third buffer 16:15:43 rrsagent, make minutes 16:15:43 I have made the request to generate https://www.w3.org/2022/10/13-tt-minutes.html nigel 16:16:24 scribeOptions: -final -noEmbedDiagnostics 16:16:27 zakim, end meeting 16:16:27 As of this point the attendees have been Andreas, Hew, Nigel, Pierre, Atsushi 16:16:29 RRSAgent, please draft minutes v2 16:16:29 I have made the request to generate https://www.w3.org/2022/10/13-tt-minutes.html Zakim 16:16:32 I am happy to have been of service, nigel; please remember to excuse RRSAgent. Goodbye 16:16:36 Zakim has left #tt 16:40:04 rrsagent, excuse us 16:40:04 I see no action items