12:54:56 RRSAgent has joined #me 12:54:56 logging to https://www.w3.org/2022/10/04-me-irc 12:55:09 Meeting: MEIG monthly meeting 12:55:13 scribe+ cpn 12:59:34 ohmata has joined #me 13:00:14 present+ Hisayuki_Ohmata, Ryo_Yasuoka, Tatsuya_Sato, Chris_Needham 13:01:13 present+ Kaz_Ashimura 13:01:21 present+ Hiroshi_Fujisawa 13:01:24 tatsuya has joined #me 13:04:18 tidoust has joined #me 13:04:28 present+ Francois_Daoust 13:07:39 present+ Chris_Lorenzo 13:07:51 present+ Kinji_Matsumura 13:08:14 agenda: https://lists.w3.org/Archives/Public/public-web-and-tv/2022Sep/0023.html 13:08:26 scribenick: ka 13:08:28 scribenick: kaz 13:08:37 hfujisawa has joined #me 13:08:43 topic: Agenda 13:08:45 - Welcome 13:08:45 - Brief recap of media discussions at TPAC : 5 mins 13:08:45 - Hybridcast Updates with NHK: 10 mins 13:08:45 - NHK work and proposals for Application Development for Consumer Products TF: 15mins 13:08:45 - Q&A and discussion: 30 mins 13:08:46 kinjim has joined #me 13:08:46 - AOB 13:08:56 zakim, who is on the call? 13:08:56 Present: Hisayuki_Ohmata, Ryo_Yasuoka, Tatsuya_Sato, Chris_Needham, Kaz_Ashimura, Hiroshi_Fujisawa, Francois_Daoust, Chris_Lorenzo, Kinji_Matsumura 13:09:16 topic: TPAC 2022 13:09:27 cpn: during the MEIG meeting 13:09:36 ... discussion on TV application development 13:09:49 ... media device specific APIs 13:09:57 ... would move forward 13:10:11 ... should invite interested people 13:10:18 ... will talk more about that 13:10:20 ... then 13:10:43 ... CTA WAVE HTML Application TF and Web Meda API snapshot updates 13:10:50 ... then 13:11:05 ... presentation from Fraunhofer on next generation audio codec support 13:11:36 ... new activity starting up 13:12:00 i/new/... during the joint meeting with WebRTC and Media WG 13:12:15 ... new media architecture in browsers 13:12:42 ... then the new CG on Audiovisual Media Formats 13:13:01 ... everybody is welcome to join to participate in the discussion 13:13:04 ... then 13:13:08 ... Media WG meeting 13:13:16 ... MSE feature prioritization 13:13:31 ... pre-emptive eviction for low memory platforms 13:13:46 ... also DASH emsg support 13:14:16 ... follow up on DASH-IF. still too early with implementations 13:14:23 ... then 13:14:34 ... EME v2 13:14:47 ... aiming for FPWD during the Charter period (May 2023) 13:14:59 ... then 13:15:09 ...Media session 13:15:24 ... agreed to add actions to support slide presentations over WebRTC 13:15:38 rrsagent, make log public 13:15:42 rrsagent, draft minutes 13:15:42 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 13:15:46 ... Audio Focus API 13:15:56 ... music vs notification sound 13:16:13 ... interest from Apple and Mozilla for API 13:16:24 ... use cases for proposed API 13:16:29 ... then 13:16:41 ... WebCodecs and associated metadata 13:16:53 ... analyze the video for, e.g., face detection 13:17:01 ... use a registry approach 13:17:21 ... introduction for document PinP 13:17:29 s/PinP/PinP API/ 13:17:44 ... use case might be sharing your custom control 13:17:58 ... that is a brief summary 13:18:05 ... not everything, though 13:18:19 q+ 13:19:03 Kaz: Regarding, consumer device TF discussion, as I mentioned at TPAC it would make sense to collect current practices, including from NHK 13:19:08 ack k 13:19:23 i|Regarding|scribenick: cpn| 13:19:31 rrsagent, draft minutes 13:19:31 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 13:19:52 topic: Hybridcast updates 13:19:58 @@@ slides tbd 13:20:06 i/scribenick: kaz/ 13:20:42 sato: ToC 13:20:46 ... Updates on IPTVF 13:20:48 ... then NHK 13:20:58 ... [slide 3] 13:21:02 ... [slide 4] 13:21:13 ... Hybridcast diffusion status 13:21:36 ... annual shipments and total ones of Hybridcast TVs 13:21:55 ... available on over 300 TV models 13:21:59 ... [slide5] 13:22:08 ... NHK's services 13:22:11 ... [slide 6] 13:22:22 ... Media service for various devices 13:22:38 ... [slide 7] 13:22:39 ... Video streaming 13:22:46 ... (diagram of their system) 13:23:02 ... broadcast station -> Hybridcast TV 13:23:05 ... [slide 8] 13:23:12 ... Software tools for video streaming 13:23:27 ... developed opensouce softwares 13:23:39 ... Bajoo.js, Relay-miffe, Delivery-miffe 13:23:48 ... [slide 9] 13:23:53 ... Device collaboration 13:24:23 ... Broadcast station -> Hybridcast TV -> Hybridcast Connect -> Devices 13:24:29 ... [slide 10] 13:24:39 ... Linkage between Hybridcast and WoT 13:24:54 ... WoT-enabled TV services using Hybirdcast Connect at TPAC 2022 13:25:04 ... demo during the WoT breakout 13:25:12 ... [slide 11] 13:25:20 ... OSS for Hybridcast Connect lerners 13:25:37 ... opensource software for Hybridcast Connect as well 13:26:00 ... Hybridast Connect TV emulator, SDK, Client sample application 13:26:24 ... can be connected to the Hybridcast TV 13:26:36 ... [slide 12] 13:26:36 ... resources 13:26:42 ... NHK OPEN HOUSE 2022 13:26:55 ... Application GitHub repository 13:26:59 ... tx! 13:27:02 q+ 13:27:07 rrsagnt, draft minutes 13:27:11 cpn: really interesting 13:27:25 ... can you explain a bit more about the WoT demo? 13:27:30 ... what's happening here? 13:28:25 sato: (shows their slides on the demo itself) 13:28:41 @@@ slide tbd 13:29:00 sato: presentation during the WoT demo breakout by Endo from NHK 13:29:44 ... demonstration including succor program on TV and an air cleaner 13:30:25 ... air cleaner's noise would be lower depending on the succor game's status 13:30:29 cpn: understood 13:31:36 s|@@@ slide tbd|-> https://github.com/w3c/wot/blob/main/PRESENTATIONS/2022-09-hybrid-f2f/2022-09-14-WoT-F2F-TPAC2022_Breakout-Endo.pdf NHK's slides on their demo during the WoT demo breakout session| 13:31:49 sato: (shows the demo content within the slides) 13:32:28 ... event message sent from the TV side to the air cleaner side depending on the succor game's situation 13:32:51 cpn: is this MTE-based? 13:32:54 sato: yes 13:33:21 ohmata: originally based on the event message from the TV program 13:34:00 ... trigger messages, e.g., start, goal, replay 13:34:22 ... handled by WoT and Hybridcast protocol 13:34:28 cpn: make sense 13:34:51 ... interesting to see the detail of the mechanism 13:35:05 ... don't have enough time today, though 13:35:11 q+ 13:35:14 ack c 13:35:36 ... do you have any description to control the TV? 13:36:01 ... wondering if we should discuss the detail during the future calls 13:36:07 s/any description/a WoT Thing Description/ 13:37:10 ohmata: WoT protocol and Hybridcast protocol handle the message 13:37:31 kaz: maybe you can show the p9 from Endo-san's slide set to describe the mechanism 13:38:17 Kaz: The TV also has a WoT Thing Desciprtion. Two devices using the WoT mechanism 13:38:29 i/The TV/scribenick: cpn/ 13:38:30 q? 13:39:30 Kaz: Good start for collaboration with MEIG and WoT. As discussed with WebTransport people, a use cases is surveillance cameras, adding metadata annotations, handling related devices using WoT and MTE 13:39:37 ... That kind of collaboration would be useful 13:39:37 ack k 13:40:11 cpn: ok. let's have another detailed discussion for this direction again 13:40:18 i/ok/scribenick: kaz/ 13:40:44 topic: NHK work and proposals for Application Development for Consumer Products TF 13:40:56 cpn: next proposals for Consumer Products TF 13:41:20 yasuoka: Case studies of Hybridcast application development... 13:41:24 ... [slide 1] 13:41:29 ... Hybridcast 13:41:42 ... already in service around 10 years 13:41:46 ... [slide 2] 13:41:55 ... Stakeholders and their relationship 13:42:13 ... IPTV Forum Japan generates Hybridcast standards 13:42:18 ... and operation guidelines 13:42:29 ... also runs conformance tests 13:42:58 ... for various TV manufactures and application developers 13:43:02 ... [slide 3] 13:43:12 s/3/4/ 13:43:14 s/2/3/ 13:43:16 s/1/2 13:43:29 ... Performance difference 13:43:56 ... performance difference among various devices is a big issue 13:44:09 ... [slide 6] 13:44:22 ... Case 1: Animation rendering performance 13:44:29 ... focus frame movement 13:45:01 ... need to reduce CSS transition property to avoid the delay 13:45:28 ... asked the TV manufactures to improve the device, but couldn't 13:45:48 ... so had to reduce some of the properties 13:46:00 ... to improve the performance 13:46:05 ... [slide 7] 13:46:17 ... Case 2: Process when switching services 13:46:30 ... requirements for linear video services 13:46:45 ... broadcast => streaming 13:46:57 ... vice versa 13:47:22 ... quick switching after getting the trigger is expected 13:47:42 ...but difficult to control what will be displayed on the screen 13:47:59 ... due to the performance difference 13:48:11 ... [slide 9] 13:48:23 ... Performance test activities (by the IPTVF-J) 13:48:37 ... UX improvement for Hybridcast services 13:48:59 ... test categories: screen rendering, app loading, etc. 13:49:03 ... [slide 10] 13:49:10 ... Test 1: Animation by sprite sheet 13:49:24 ... 8-sec horizontal moveent of an image 13:49:36 ... measured 10 times on 6 different TVs 13:49:44 ... results vary... 13:50:11 ... expectation was 8 sec every time 13:50:27 s/time/time, though/ 13:50:37 ... [slide 11] 13:50:44 ... Test 2: Browser launch time 13:51:21 ... time span between launch button pressed and invocation 13:51:25 ... [slide 12] 13:51:36 ... conclusion and discussion 13:51:52 ... past and ongoing challenges on performance issues 13:52:15 ... how can the application developers and TV vendors contribute to the TF? 13:52:23 ... what can they expect? 13:52:40 ... can provide use cases based on our experience 13:52:51 s/use cases/our use cases/ 13:53:03 ... questions? comments? 13:53:09 rrsagent, draft minutes 13:53:09 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 13:53:12 cpn: tx! 13:53:42 ... ChrisL, any comments? 13:53:58 cl: interesting discussion on pushing Web and browsers for applications on TV 13:54:03 ... major challenge 13:54:17 ... need to increase the performance for TV apps 13:54:36 ... for transitions and animations, the performance is not enough 13:54:46 ... and need to reduce the animations, etc. 13:55:07 ... not sure if you're aware of Lightning platform which uses WebRTC 13:55:18 ... let me post the resource 13:55:24 ChrisLorenzo has joined #me 13:55:37 cpn: aspects here, we've not considered for the TF discussion 13:55:40 Developer Experience - https://github.com/w3c/media-and-entertainment/pull/92/files?short_path=753fd1d#diff-753fd1d9b2d6cb55d78b5c965fca3387ca62e617d22e89a0121d86ebdf23059b 13:55:51 ... very much consistent with what ChrisL has been mentioning 13:56:00 https://github.com/w3c/media-and-entertainment/blob/master/app-development/charter.md 13:56:11 ... one of the techniques we use is 13:56:16 Framework for performance on TV apps - https://lightningjs.io/ 13:56:23 ... enabling features on TV to be tested 13:56:40 ... framework to allow conditionally able/disable the feature 13:56:55 ... if less capable, disable the feature 13:57:11 ... the other thing interesting is start up performance time for Hybridcast applications 13:57:27 yasuoka: (shows the Test 2 results again) 13:57:43 Can we get the link to the slides on here? 13:57:44 cpn: any insights to what the cause of the delays here? 13:57:55 ... how does it vary with the browser implementations? 13:58:25 yasuoka: it's difficult to describe the detail 13:59:06 cpn: any additional measurement where the TV is spending the time? 13:59:20 ... e.g., the browser itself is a large application binary 13:59:29 ... and needs to load big memory at once 13:59:45 ... way to instrument the details? 14:00:28 yasuoka: green part depends on the network connections 14:00:43 s/green/firstly, green/ 14:01:22 ... need to dig into the detailed behavior of Hybridcast implementation 14:01:39 crisL: the user press the launch button is time 0 14:01:53 ... the browsers already receive the responnse 14:01:58 s/responnse/response/ 14:02:07 ... then HTTP connection 14:02:16 ... then load the page content 14:02:27 ... browsers do that quickly, actually 14:02:43 ... what the difference on A-J here? 14:02:45 igarashi has joined #me 14:02:56 yasuoka: TV model variations 14:03:00 ... available in Japan 14:03:04 crisL: ok 14:03:19 ... this is very similar with smartphones variation 14:03:31 ... A and J are very fast 14:03:39 cpn: yeah, that's interesting 14:03:49 ... are those faster ones the latest models? 14:03:55 ... or most expensive ones? 14:04:04 yasuoka: newer ones are faster 14:04:28 crisL: WAVE guys are doing similar tests 14:04:35 present+ Tatsuya_Igarashi 14:04:55 ... some TVs have powerful chips 14:05:07 ... and some don't 14:05:12 q+ 14:05:23 cpn: interesting approach 14:05:30 present+ 14:05:42 ... not sure if it's within the scope of W3C, though 14:06:03 q? 14:06:04 ... similar to the CTA WAVE's activity 14:06:57 nigel has joined #me 14:07:24 Kaz: TV device performance may be out of scope, but getting this kind of inforation from various areas and SDOs should be useful. 14:07:32 ... We can look into the detailed processing time in this kind of procedure, not the device or network part, but the browser processing time should be clarified. 14:07:43 ... If we can get that kind of clarification and see the actual cause of time delay, we can look with browser vendors to improve 14:08:01 ... As ChrisN metioned, maybe NHK could bring this to CTA WAVE 14:08:07 ... MEIG could look at the results in more detail 14:08:13 ack k 14:08:14 ... such as in TF calls 14:08:23 i/TV device/scribenick: cpn/ 14:08:34 rrsagent, draft minutes 14:08:34 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:08:59 cpn: happy to spend more time to continue this 14:09:42 i/happy/scribenick: kaz/ 14:09:56 ... one of the things interesting is 14:10:18 ... Web Performance WG's activity on measuring the performance 14:10:29 Present+ Nigel_Megitt 14:10:42 ... indication of measurement 14:11:01 ... rendering stability, how the content moves, etc. 14:11:31 ... is this something interesting from our viewpoint? 14:12:22 ... have you started to look at alternate rendering mechanism? 14:12:38 ... e.g., Lightning 14:12:45 ... how would they impact, etc. 14:13:34 yasuoka: this testing activity is done for those kind of mechanisms yet 14:13:43 s/Lightning/WebGPU and Lightning/ 14:13:55 ... would like to try them in the future 14:14:06 q? 14:14:51 ... currently, Hybridcast doesn't support WebGPU or WebRTC 14:15:45 ... we'd like to try them but it's kind of difficult at the moment 14:15:56 q+ 14:16:17 cpn: too difficult to introduce those new features. right? 14:16:25 ... what kind of help do you expect from W3C? 14:17:07 yasuoka: if those mechanisms could be actual standards for media handling 14:17:10 ... that would be great 14:17:59 ... would like to proceed with to consider those new mechanisms 14:18:08 s/to consider/considering/ 14:18:33 ohmata: Hybridcast currently doesn't define those mechanisms as mandated 14:19:21 ... but think it would be useful 14:19:46 cpn: then 14:19:51 ... the question here 14:20:03 ... how can app/product developers contribute to the TF? 14:20:18 ... would like to tackle the issues one by one 14:20:38 ... the discussion we have about launching time, etc. 14:20:52 q? 14:21:22 Kaz: We can start with gathering information on pain points and workarounds based on industry experience 14:21:48 ... We have another example from Chris Lorenzo at Comcast. We can summarise the description also and put them together 14:22:02 ack k 14:22:03 ... We can also ask someone from BBC and Fraunhofer about their experience 14:22:15 i/We can/scribenick: cpn/ 14:22:22 ChrisN: Yes, helpful to gather more experience reports 14:22:50 cpn: another question is expected output 14:22:59 i/another/scribenick: kaz/ 14:23:05 ... related to your 2nd question 14:23:22 ... What can app/product developers expect from the TF? 14:23:38 ... we want to continue to use the Web technology 14:24:07 ... if the performance improvement can be achieved based on the current Web technology 14:24:14 ... we should identify what to be done 14:24:36 ... this could be an area on Web Performance as well 14:25:05 ... this whole idea i around having tests by application developers and device manufacturers 14:25:24 ... on achievable performance goals 14:25:36 ... could even become a marketing tool 14:25:51 ... having a representative test suite 14:26:02 ... may be a valuable thing 14:26:03 ... and 14:26:17 ... the other area is looking at alternative rendering mechanism 14:26:27 ... instead of DOM-based rendering mechanism 14:26:48 ... we could start to look at canvas-based, WebGL-based, etc. 14:27:07 ... new rendering standard in the future could be developed 14:27:19 ... about how that easily could adopt 14:27:27 ... with HTML-based applications 14:27:38 ... support new technologies 14:27:48 ... would require new open discussions 14:28:00 q? 14:28:39 ... would suggest creating GitHub issues to capture your points 14:28:50 ... and see how to follow up each of them 14:29:01 ... would that make sense? 14:29:08 kaz: +1 14:29:17 crisL: also agree 14:29:33 cpn: so I'll propose to do that 14:29:37 ... really good input 14:29:45 ... thank you for your presentation! 14:30:04 ... something we'd like to continue for collaboration 14:30:13 yasuoka: thank YOU! 14:30:25 rrsagent, draft minutes 14:30:25 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:30:42 cpn: would invite you to the discussion 14:30:55 ... happy to have you both 14:31:12 ... would continue to work with you via Kaz 14:31:24 ... anything else for today? 14:31:31 (none) 14:31:50 chair: Chris_Needham, Chris_Lorenzo, Tatsuya_Igarashi 14:32:06 cpn: look forward to see further collaboration 14:32:25 chrisL: to make sure, please upload the slides 14:32:29 cpn: yes 14:32:40 ... please send them to me or Kaz 14:32:49 ... so that we can upload them 14:32:55 yasuoka: will do 14:33:01 cpn: tx 14:33:10 rrsagent, draft minutes 14:33:10 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:33:45 i|sato: ToC|scribenick: kaz| 14:33:52 [adjourned] 14:33:55 rrsagent, draft minutes 14:33:55 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:34:39 i/We can start/scribenick: cpn/ 14:34:48 i/We can also ask/scribenick: kaz/ 14:34:49 rrsagent, draft minutes 14:34:49 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:35:29 s/We can also ask some/scribenick: cpn/ 14:35:30 rrsagent, draft minutes 14:35:30 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:36:12 s/scribenick: cpnone/We can also ask someone/ 14:36:14 rrsagent, draft minutes 14:36:14 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:36:24 i/from BBC/scribenick: cpn/ 14:36:25 rrsagent, draft minutes 14:36:26 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:38:12 i|next proposals for Con|@@@ slide tbd| 14:38:14 rrsagent, draft minutes 14:38:14 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:39:42 s/cpn:/ChrisN:/g 14:40:01 s/crisL:/ChrisL:/g 14:40:04 rrsagent, draft minutes 14:40:04 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:40:37 s/yasuoka:/Yasuoka:/ 14:40:50 s/cl:/ChrisL:/ 14:40:53 rrsagent, draft minutes 14:40:53 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:41:11 s/sato:/Sato:/ 14:41:15 s/sato:/Sato:/g 14:41:29 s/ohmata:/Ohmata:/g 14:41:36 s/kaz:/Kaz:/ 14:41:37 s/kaz:/Kaz:/g 14:41:40 rrsagent, draft minutes 14:41:40 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:42:22 s/yasuoka:/Yasuoka:/g 14:42:24 rrsagent, draft minutes 14:42:24 I have made the request to generate https://www.w3.org/2022/10/04-me-minutes.html kaz 14:42:41 cpn_ has joined #me 15:28:48 tidoust has joined #me 16:25:02 Karen has joined #ME 16:28:32 Zakim has left #me 16:57:18 Karen has joined #ME