14:00:08 RRSAgent has joined #me 14:00:12 logging to https://www.w3.org/2025/06/03-me-irc 14:00:17 ohmata has joined #me 14:00:26 cpn has joined #me 14:00:33 present+ Chris_Needham 14:00:45 nigel has joined #me 14:01:14 present+ Kaz_Ashimura, Hiroki_Endo 14:01:29 present+ Hisayuki_Ohmata, Tatsuya_Igarashi 14:01:39 present+ Kensaku_Komatsu 14:01:48 igarashi has joined #me 14:01:55 present+ 14:02:12 present+ Nigel_Megitt 14:02:13 Present+ Nigel_Megitt_BBC 14:02:48 endo has joined #me 14:02:52 Agenda: https://www.w3.org/events/meetings/1d6d52b3-5688-4bd4-b481-ecb5a1e772d0/20250603T150000/ 14:03:13 present+ Rob_Smith, Bernd_Czelhan 14:03:22 rrsagent, make log public 14:03:26 rrsagent, draft minutes 14:03:27 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:03:45 scribe+ nigel 14:04:28 Slides: https://www.w3.org/2011/webtv/wiki/images/c/c1/W3C_MEIG_Meeting_3_June_2025.pdf 14:05:03 Topic: Welcome 14:05:16 cpn: Welcome to our monthly call. Apologies we didn't hold a call last month. 14:05:19 .. Agenda today: 14:05:34 chair: ChrisN, Igarashi 14:05:39 .. Authentic Web Workshop meeting on C2PA 14:05:48 .. Media Metadata Japanese Community Group update 14:05:54 .. TPAC 2025 Planning 14:06:01 .. Sync on the Web Community Group update 14:06:04 i|Agenda today|-> https://lists.w3.org/Archives/Public/public-web-and-tv/2025May/0007.html Agenda for today| 14:06:07 rrsagent, draft minutes 14:06:08 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:06:15 .. AV Media Formats for Browsers 14:06:15 .. DataCue 14:06:20 .. MSE issues next steps 14:06:27 .. Anything else to add to the agenda today? 14:06:41 no other business 14:06:50 Topic: Authentic Web Workshop: C2PA 14:07:07 cpn: Idea was a workshop style of meetings addressing the general problem of misinformation on the web. 14:07:19 .. Motivation was seeing a number of technology solutions being developed and proposed 14:07:45 .. so they wanted to facilitate conversations within W3C to see if anything should be taken forward 14:07:47 .. for standardisation. 14:08:29 .. I've been on the programme committee for this group, along with Dom, and others 14:08:39 .. Two meetings. 14:08:54 .. First was an introduction, second was about C2PA itself. 14:08:58 meeting: Media & Entertainment Interest Group 14:09:00 rrsagent, draft minutes 14:09:01 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:09:18 -> Session 1 (March 12) Introduction and Framework: https://github.com/w3c/authentic-web-workshop/blob/main/minutes/2025-03-12AuthWeb.md 14:09:41 -> Session 2 (May 6) C2PA: https://github.com/w3c/authentic-web-workshop/blob/main/minutes/2025-05-06AuthWeb.md 14:10:05 cpn: Upcoming are Trustnet and other topics. There then may be an in-person meeting 14:10:20 .. to go through the relative advantages or disadvantages of each and look more closely at what 14:10:26 .. W3C may want to standardise on this area. 14:10:47 .. Focusing on C2PA, many of you will have heard Leonard presenting this more than a year ago. 14:11:06 .. He talked about the overall design characteristics and what they are targeting, 14:11:22 .. provenance metadata and signals for how to use the content, like opt-outs for AI scripting 14:11:27 .. "do not train your AI models". 14:11:34 RobSmith has joined #me 14:11:38 .. They're looking at a solution for audio, and live video. 14:11:51 .. They're also working on text and HTML, I don't know details of how they're thinking about that. 14:12:03 .. Interesting area. Also developing UI level guidance. 14:12:23 .. Then Brendan from IPTC talked about work they're doing on publisher identity and creating trust lists 14:12:27 .. for publishing organisations. 14:12:40 .. A number of GitHub issues were raised with specific questions. 14:12:49 .. E.g. use of tech in open source software, 14:13:02 .. use by different user groups like consumers and fact checkers, and how the signals are useful 14:13:08 .. and may be interpreted. 14:13:24 .. Some questions about extended use validation certificates. 14:13:37 .. Links on the slides. 14:13:52 .. Open Source is interesting: how do you keep your signing keys secret? 14:14:12 .. Anyone can fork the project so how to make sure the signing keys are assigned? 14:14:24 .. Seems they're thinking at a user level rather than a software version level. 14:14:41 .. Related to that, connecting a user's social media identity to the signing certs, or using verifiable credentials. 14:14:42 -> Session 3(June 10) Trustnet: https://www.w3.org/events/meetings/f0e40d0d-3328-4d7c-85cd-aefd67ea5f6d/ 14:14:58 cpn: W3C could be looking at what a browser integration might look like. 14:15:09 .. Increasing amount of content on the web with C2PA signatures. 14:15:19 .. How is that surfaced to end users, and how do they interact with it? 14:15:32 i|Open Source is|-> https://github.com/w3c/authentic-web-workshop/issues/17 Issue #17 - What sorts of software need to be able to produce "authentic[ated]" content?| 14:15:38 .. Currently via JS libraries and interfaces provided by the site you're looking at. 14:15:51 q? 14:15:56 .. If the browser is your trusted UA then would be good to prevent website spoofing it. 14:16:03 .. That's the idea behind a browser API. 14:16:19 .. The conversation is with Jeffrey Yasskin, one of the technical leaders on the Chromium project and the 14:16:33 .. Google Chrome browser. He pointed to a project called signature based integrity. 14:16:50 .. There's also subresource integrity that allows sites to validate that e.g. script code loaded into the page 14:17:07 .. matches a hash or a signature. Signature based integrity extends that by using signing certs. 14:17:21 .. Somewhat related technologies in development, interesting to compare with C2PA. 14:17:35 i|Related to that|-> https://github.com/w3c/authentic-web-workshop/issues/19 Issue 19 - How can we express a signer's identity to an end-user, in a way that's not vulnerable to the problems that doomed EV certificates?| 14:17:36 .. A point that came up was that perhaps the browsers would be less trusting of what C2PA calls 14:17:55 .. soft browsing. E.g. take an image, read the watermark and then recover the metadata. 14:18:10 .. Images get manipulated or resized and then it is easy to lose the train of signatures and therefore 14:18:23 .. having a mechanism to recover is seen as useful within the C2PA community. 14:18:35 .. Depending on how robust it is, may or may not be something browsers want to adopt. 14:18:44 i|W3C could be looking at|-> https://github.com/w3c/authentic-web-workshop/issues/20 Issue 20 - Browser API for C2PA| 14:18:50 rrsagent, draft minutes 14:18:51 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:18:57 .. I was encouraged by the conversation with Jeffrey on this because he was interested to engage 14:18:58 .. in analysing the technology and providing feedback. 14:19:17 .. Doesn't translate into Chrome support, there's a long way to go, not clear if they would support it 14:19:22 .. and if anything is needed to enable it. 14:19:28 .. Good to start the conversation though. 14:19:40 .. Any opinions or questions around this? 14:19:43 q? 14:19:46 q+ 14:20:00 kaz: Thank you for your report, very useful. 14:20:13 .. I was wondering if the MEIG might want to talk about some of those issues in detail 14:20:22 .. during the early discussion? Or maybe not. 14:20:38 cpn: I don't know, I'm hopeful that the workshop meetings will come back to this topic, and perhaps 14:20:43 q+ 14:20:48 .. do it there. I suggest let's see how the workshop meetings progress 14:20:50 ack k 14:21:23 .. There are other things like audio and live video. The C2PA community have dedicated work within 14:21:38 .. their consortium to develop those things. It would be interesting for us to hear how that is going. 14:21:40 q+ 14:21:54 .. I remember when Leonard presented to us in this group, he invited us to contribute to that. 14:21:55 Chrome project ? https://chromestatus.com/feature/5032324620877824 14:22:14 .. A useful thing to do might be to focus on the audio and live video cases, 14:22:22 .. and at least hear about their latest developments. 14:22:42 kaz: That's true. On the other hand, maybe re coordination, and how to integrate various different 14:22:59 .. video streaming service with a consolidated metadata service, user credentials might come into it, 14:23:10 .. so in 6 months or so we could revisit the results from the workshops. 14:23:23 cpn: Good suggestion, let's plan to do something like that. 14:23:35 s/Crome/Chromium/ 14:23:39 q? 14:23:39 .. Hopefully in that timescale the workshop will have produced more concrete suggestions. 14:23:41 ack Ro 14:23:57 RobSmith: Thanks Chris, on your comments about audio and video, also metadata, thinking form 14:24:14 .. a news reporting point of view, if reports come in claiming to be from a location, verify that it is at 14:24:23 .. that location, using metadata embedded into the video. 14:24:27 s/form/from 14:24:34 s/form/from 14:24:40 s/from/form 14:25:03 .. You're not limited to a particular technology but it would be valuable if harnessed correctly, 14:25:07 rrsagent, draft minutes 14:25:09 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:25:15 .. given the ease of filming with smartphones etc. 14:25:34 cpn: Yes it allows people to make claims but doesn't necessarily say if they are truthful or not. 14:25:55 .. If you're looking to verify that a video was actually taken at the claimed location that would be an extra step. 14:26:16 RobSmith: I'm involved with a group called GeoPose which is not just about where you are but also where you're looking. 14:26:38 .. That's how it works - look at key markers in the video image to determine where the video was taken. 14:26:58 .. I can imagine independent fact checkers confirming location and signing it has been verified. 14:27:06 cpn: That's interesting. 14:27:33 RobSmith: It may not have been deliberately obfuscated because if people film important events that 14:27:42 q? 14:27:44 q+ 14:27:44 .. could be in a crisis zone etc the visual reference points you're expecting might not be there (any more). 14:27:59 .. You may still have enough visual markers remaining to confirm location though. 14:28:01 present+ Piers_O'Hanlon 14:28:06 rrsagent, draft minutes 14:28:08 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:28:27 cpn: If you've been following the workshop meeting, this would be good to discuss in that context. 14:28:33 .. The layering approach makes a lot of sense. 14:28:45 .. It's part of what the Chrome proposal being discussed next week may speak to, 14:28:59 .. if it is intended to enable independent verification to determine veracity. 14:29:12 .. The other thing is if the OGC has a relationship to C2PA. 14:29:29 .. I don't participate in C2PA so I don't know, but maybe they could give valuable input through 14:29:31 .. coordination. 14:29:41 s/GeoPose/GeoPose in OGC 14:30:01 .. We could contact Leonard for introductions. 14:30:02 ack iga 14:30:45 q+ igarashi 14:31:22 q- 14:31:59 kaz: Rob, thank you for your comment - it would be important for the Smart Cities discussion also. 14:32:01 komasshu has joined #me 14:32:08 .. Let me talk with you offline 14:32:12 ack ka 14:32:32 rrsagent, draft minutes 14:32:33 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:33:21 About the Signature based integrity, the difference from C2PA is if the certificate is embedded in a media. Is there any discussion on the difference? 14:33:47 cpn: Thank you, I don't know the detail on this. 14:34:21 .. I would recommend to look at the GitHub conversation which is 14:34:25 .. issue 20 https://github.com/w3c/authentic-web-workshop/issues/20 14:34:34 https://www.geopose.org/ 14:34:43 .. I don't know the difference in how it works, would need a more detailed comparison to understand that 14:34:50 .. I haven't gone into that detail myself. 14:34:51 s|issues/20|issues/20 Browser API for C2PA| 14:34:54 rrsagent, draft minutes 14:34:55 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:35:02 igarashi: OK 14:35:12 Topic: Media Content Metadata Japanese CG 14:35:32 endo: Hello, my name is Hiroki Endo from NHK, and I am chair of Media Content Metadata Japanese CG 14:35:37 .. or MCMJ CG 14:35:46 .. Thank you for the opportunity to share our recent activities. 14:35:57 .. We are focusing on interoperability of media content metadata, 14:36:05 .. starting with the Japanese media industry. 14:36:16 .. Today I would like to share two key updates. 14:36:21 i|Hello|-> https://www.w3.org/community/mcm-jp/ Media Content Metadata Japanese CG| 14:36:22 rrsagent, draft minutes 14:36:23 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 14:36:37 .. In March 2025 we published a draft report titled : “Case Studies on Media Metadata Interoperability” 14:36:42 .. in Japanese and English 14:36:54 .. Some sections are still in Japanese only, but we are working on improving it. 14:37:25 .. The report includes case studies from broadcasting and publishing. 14:37:36 .. We would appreciate your feedback as we continue to refine the report. 14:37:58 -> https://w3c-cg.github.io/mcm-jp/reports/cg-report.html?lang=en Draft CG Report on GitHub 14:37:59 cpn: Do you want general feedback or are there particular questions for more specific feedback? 14:38:24 endo: We would like information we should add into the report such as related initiatives and specifications. 14:38:44 cpn: There may be more like from TV standards type bodies. 14:38:49 .. I see you have DVB-I mentioned. 14:38:58 .. We would need to study your use cases. 14:39:09 .. I'm thinking of TV-Anytime, an ETSI specification. 14:39:24 .. Is the approach you're taking to use this document to start drafting specifications, 14:39:45 q? 14:39:49 .. or is the scope of this more about the use cases and you would work on specifications in a new document? 14:40:06 endo: This report only includes case studies, and after it is completed we will identify 14:40:14 .. the technical requirements. 14:40:22 q+ 14:40:37 cpn: Thank you for sharing this. 14:41:03 endo: The main content about use cases from industry is written only in Japanese now. I will update and 14:41:18 .. in a future meeting share again with MEIG. 14:41:35 cpn: Thank you, that's very welcome. I think we can review this anyway using a translator. 14:41:55 .. I guess the question I would have is whether other companies have use cases they would want to add to this? 14:42:45 endo: Yes, the authors are from various industries. Shogakukan is a publishing company. 14:42:54 .. Omron is a media editing software vendoe. 14:43:03 .. M Data is a metadata company in broadcasting. 14:43:19 .. Jorte Inc is an event metadata platform. 14:43:33 .. They have mainly contributed use cases from their own industries. 14:43:40 cpn: What kind of event data is that? 14:44:28 endo: Such as festivals. They have an event publication platform, and provide an application called Eventia. 14:44:42 .. Users can build various events based on map information. 14:44:49 cpn: Interesting, thank you. 14:45:00 q+ 14:45:05 .. I can see the use cases where you have media content in a video platform that you can link 14:45:19 .. to related real world events and surface all of that to the user. 14:45:21 ack k 14:45:32 kaz: Thank you Endo san and Chris. 14:45:55 .. They have been collecting best practices for real events in Japan, and identifying pain points also. 14:46:07 .. At some point after translating into English we can talk about their pain points and potential 14:46:12 .. requirements for web standards. 14:46:14 cpn: Absolutely. 14:46:17 q? 14:46:31 .. I hope that they bring that to the CG to continue that work. Sounds good. 14:46:37 ack igarashi 14:46:58 igarashi: Thank you Endo san. I'm looking at your draft report and for me it is a little bit unclear 14:47:14 .. what interoperability means here. You describe the exchange of metadata. 14:47:30 .. The word interoperability has a slightly different meaning from these use cases. 14:47:41 .. The use case should not be metadata interop but metadata exchange. 14:47:58 .. The issue you want to address is interop between metadata, so you should describe the problems 14:48:17 .. you want to address - this would be more helpful to cover with the CG. 14:48:25 endo: Thank you very much, I agree with you. 14:48:53 .. In our CG, interoperability solutions are mainly for exchanging metadata but 14:49:17 .. each industry cannot identify the existence of each industry metadata includes their own industry metadata, 14:49:45 .. so first, why we want to service interoperability and why we cannot service interoperability now 14:50:06 .. in case studies we identify the existence of each industry's metadata and how to exchange each metadata. 14:50:18 .. So data exchange is the main part of this report. 14:50:47 .. I want you to read the examples of data exchange. 14:51:01 igarashi: I am not an English speaker, but interoperability means that we have a standard but they 14:51:27 .. have some issue implementing the spec. It feels like the interoperability is a standardisation issue. 14:51:48 .. Does the CG want to define a common metadata format or just identify the different kinds of metadata, 14:51:56 .. or an ontology? Which kind? 14:52:05 endo: The later one [an ontology] 14:52:11 q+ 14:52:54 igarashi: Ontology for media contents is not something we have a standard for in industry use cases. 14:53:17 s/igarashi/endo/ 14:53:39 kaz: Just to make sure, this is a CG and I have been attending their meetings. 14:53:56 .. I believe Endo San's objective is not to create an ontology but to identify best practices and pain points 14:54:11 .. and then work with the other SDOs or WGs or IGs and think about how to achieve the expected 14:54:18 .. standardisation with those stakeholders. 14:54:22 cpn: Makes sense, thank you. 14:54:25 q? 14:54:26 ack k 14:54:35 .. Finally, there's an upcoming showcase at the Interop Tokyo event next week. 14:54:43 Topic: TPAC 2025 planning 14:55:00 cpn: If you can, please look out for the showcase event. 14:55:11 .. TPAC this year is in Kobe. 14:55:23 .. Each group needs to make its requests for meeting time in the next few weeks. 14:55:28 .. Deadline 20th June. 14:55:39 .. The question for us is should we meet, and if so, how much time to allocate? 14:55:48 .. My proposal is to do the same as last year, 14:56:08 .. time for MEIG, time for joint meetings with groups like Timed Text, or the APA WG for accessibility topics. 14:56:19 .. Once we have time allocated we can decide how to use that in terms of agenda and so on. 14:56:36 .. I think we don't need to decide agenda right now but we will be planning that in the next months. 14:56:50 .. Does anybody think we need more time than we had last year, or was last year okay? 14:57:02 .. I can certainly request more time should we want it. 14:57:14 .. Or does this proposal look good to you? 14:57:14 .. Any thoughts? 14:57:18 -> https://github.com/w3c/media-and-entertainment/issues/110 ME Issue 110 - TPAC 2025 planning 14:57:39 cpn: That's a GitHub issue that we will use to plan the agenda. 14:57:48 .. I invite all of you to suggest the topics we want to cover. 14:57:54 .. We will use that to plan the agendas. 14:58:06 .. Following last year as the example we would continue that. 14:58:17 .. If no questions... 14:58:32 Topic: In Brief 14:58:42 Subtopic: Sync on the Web Community Group 14:58:56 komasshu: Based on the last discussion at TPAC I created this community group at the beginning of this year. 14:58:57 -> https://www.w3.org/community/sync-on-the-web/ Sync on the Web CG 14:59:12 .. I am now the Chair of the CG. 14:59:24 s/komasshu/komatsu/ 14:59:27 .. This is focused on Media over QUIC, which has a bunch of fantastic features. 14:59:58 .. One of them is synchronisation, because Media over QUIC can handle frame by frame, which is different 15:00:00 .. to HLS etc. 15:00:28 .. I am really happy for you to join the CG and talk about your use cases. 15:00:48 .. Next IG meeting I will show you typical use cases for synchronisation on top of Media Over QUIC. 15:01:04 cpn: Thank you, that would be welcome. 15:01:13 .. Congratulations on getting the CG up and running. 15:01:27 .. I remember last TPAC you generated quite a lot of interest in the community. 15:01:47 .. We're out of time now but I would like to invite you back next time to talk about this in more detail. 15:01:51 .. This is good news. 15:02:02 Subtopic: DataCue 15:02:12 cpn: We have been doing work on the DataCue API proposal. 15:02:26 .. I haven't had time to review the latest stage of discussions but it feels like we want to continue 15:02:30 .. the momentum on that. 15:02:36 i|Subtopic: DataCue|-> https://github.com/w3c/tpac2024-breakouts/issues/54 FYI, TPAC 2024 breakout 54 - Sync on Web, now and next of realtime media services on web| 15:02:42 rrsagent, draft minutes 15:02:43 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 15:02:49 .. If it would be helpful to organise a technical discussion under WICG terms we coupd do that. 15:02:49 Rob: We should discuss in WIG 15:02:53 w/WIG/WICG/ 15:03:10 s|w/WIG/WICG/|| 15:03:14 s/WIG/WICG/ 15:03:45 Topic: Meeting close 15:03:59 cpn: Thank you everybody for joining, particularly those who presented updates. 15:04:15 .. We will meet at the beginning of July. In the meantime, please share your thoughts on TPAC preparations 15:04:39 .. on the GitHub issue, review the report on metadata and send feedback, and apologies for running 15:04:42 .. 5 minutes over! 15:04:47 .. [adjourns meeting] 15:04:54 i|Topic:|-> https://github.com/WICG/datacue/issues/35 WICG Issue 35 - Proposal to expose HTML TextTrackCue constructor| 15:04:56 rrsagent, draft minutes 15:04:57 I have made the request to generate https://www.w3.org/2025/06/03-me-minutes.html kaz 17:02:00 Zakim has left #me