21:59:02 <RRSAgent> RRSAgent has joined #mediawg
21:59:06 <RRSAgent> logging to https://www.w3.org/2024/12/10-mediawg-irc
21:59:06 <Zakim> Zakim has joined #mediawg
22:01:02 <cpn> Meeting: Media WG
22:01:08 <cpn> Chair: Chris
22:01:19 <eugene> eugene has joined #mediawg
22:01:45 <eugene> present+ Eugene Zemtsov
22:01:50 <cpn> Present+ Chris_Needham, Eugene_Zemtsov, Guido_Urdaneta, Bernard_Aboba, Jasper_Hugo, Francois_Daoust
22:02:05 <tidoust> RRSAgent, make logs public
22:02:11 <cpn> Present+ Nishitha_Burman
22:03:28 <cpn> Present+ Nic_Champagne_Williamson, Jer_Noble
22:03:47 <cpn> Agenda: https://www.w3.org/events/meetings/318343ad-12d0-4b11-bb99-c2a049be70bc/
22:05:41 <cpn> scribe: cpn
22:05:55 <Nic> Nic has joined #mediawg
22:06:46 <jernoble> jernoble has joined #mediawg
22:07:03 <cpn> Topic: WebCodecs metadata issues
22:07:30 <cpn> Subtopic: EncodecChunk metadata WebCodecs#245
22:07:46 <cpn> Bernard: Guido's been working on proposals for adding metadata to video and potentially audio
22:08:12 <cpn> ... This is an old issue, is it still relevant given Guido's more recent issues?
22:08:48 <cpn> Eugene: The metadata being added is for VideoFrameMetadata, and here it's encoded chunks. Not clear that the metadata Guido is adding is helpful there
22:09:13 <cpn> ... It's not something attached to the EncodedVideoChunk itself, it's in the output callback from the VideoEncoder
22:09:31 <cpn> ... There's lots of data there, e.g., SVC data, possibly alpha (not in current implementations)
22:09:56 <cpn> ... But if the encoded chunk is from other sources, e.g., over the network, then no metadata is attached to the EncodedVideoChunk, so not so useful
22:10:34 <cpn> Bernard: Makes sense. Issues says we'll harmonise, but don't think that's the current direction. We just want constructors
22:10:51 <cpn> Guido: That's the proposal Harald is working on, to construct one from the other, in both directions
22:11:21 <cpn> Bernard: Do we need EncodedVideoChunk to have metadata support WebRTC? The RTC encoded chunk doesn't have to be the same
22:12:43 <cpn> Eugene: I agree, constructing EncodedVideoChunk from RTCEncodedVideoFrame can and should be done. Don't think we need yet another EncodedVideoMetadataChunk to encode everything in RTCEncodedVideoFrame, those are different
22:13:05 <cpn> Bernard: Wanted to confirm the direction we're going
22:13:29 <cpn> Eugene: There'll be constructors, and nothing further for video chunk metadata
22:13:43 <cpn> Bernard: Makes sense, and I can link to Harald's proposal
22:14:35 <guidou> guidou has joined #mediawg
22:15:07 <markafoltz> markafoltz has joined #mediawg
22:15:13 <markafoltz> present+ Mark_Foltz
22:15:14 <cpn> scribe+ markafoltz
22:15:15 <tidoust> scribe+ markafoltz
22:16:30 <markafoltz> Topic: w3c/webcodecs#601 Expose in VideoFrameMetadata some fields from VideoFrameCallbackMetadata - youennf
22:16:39 <markafoltz> GitHub:  Expose in VideoFrameMetadata some fields from VideoFrameCallbac
22:16:47 <markafoltz> GitHub: w3c/webcodecs#601
22:17:17 <markafoltz> Topic: w3c/webcodecs#813 Add captureTime, receiveTime and rtpMetadata to VideoFrameMetadata - guidou
22:17:27 <markafoltz> GitHub: w3c/webcodecs#813
22:17:37 <markafoltz> RESOLVED: Merge #813
22:18:00 <markafoltz> RESOLVED: Merge #813 to close issue #601
22:19:22 <markafoltz> Topic: #855, Audio Metadata
22:20:06 <markafoltz> GitHub: webcodecs/#855
22:20:28 <markafoltz> guidou: Adding similar fields for audio metadata.  Some of them already have them.
22:20:38 <markafoltz> ...Also have them for the raw version.  No RTC equivalents.
22:20:56 <markafoltz> adoba: Comes up with lip sync, if you only have it for video
22:21:11 <markafoltz> Eugene: Arguments for video apply to audio as well
22:21:18 <markafoltz> ACTION: guidou@ to prepare a PR
22:21:54 <markafoltz> chrisn: Take the same approach as video frame metadata, use a registry
22:22:30 <markafoltz> ...Some slight duplication between audio and video fields.
22:23:59 <markafoltz> Topic: Guidance for user-defined VideoFrameMetadata entries #849
22:24:09 <markafoltz> GitHub: https://github.com/w3c/webcodecs/issues/849
22:24:24 <markafoltz> chrisn: Should there be guidance for users adding their own fields?
22:24:53 <markafoltz> ...namespacing, look at how they are returned.  They are always copied.
22:25:19 <markafoltz> adoba: Should implementations only deal with the dictionary entries, or take everything?
22:26:23 <markafoltz> Eugene: No way for developers to know which fields are supported or are not, if we copy the known entries
22:26:41 <markafoltz> ...Prefer to keep around the metadata entries which are known and supported for UA but see both sides
22:27:12 <markafoltz> ...But see argument from the developer PoV
22:27:54 <markafoltz> ...People already can add whatever they want to VideoFrame
22:28:39 <markafoltz> adoba: If there is a custom field, some browsers may not return it because it's not in the registry
22:29:06 <markafoltz> chrisn: Does it clone the metadata?
22:29:18 <markafoltz> Eugene: Yes, what is there is copied and returned by metadata.
22:29:37 <markafoltz> ...currently, need to copy everything and return it.
22:31:26 <markafoltz> Eugene: Is there a precedent in other standards?
22:32:28 <markafoltz> ACTION: Eugene to research if other specs do something like this with registries.
22:33:37 <markafoltz> Topic: Opus Packet Loss Concealment
22:33:48 <markafoltz> GitHub: https://github.com/w3c/webcodecs/issues/558
22:34:00 <markafoltz> adoba: Opus is adding advanced PLC ("Deep Red")
22:34:06 <markafoltz> ... Make sure that WebCodecs supports it
22:34:18 <markafoltz> ... ffmpeg does not support need for concealment, pass an empty frame
22:34:41 <markafoltz> ... Other codecs don't have concealment (not generalizable)
22:34:45 <cpn> cpn has joined #mediawg
22:35:11 <markafoltz> eugene: Having gaps in timestamps seems reasonable.  If we lost network packet, then we have the next packet with a future timestamp.
22:35:13 <cpn> https://www.w3.org/2024/09/26-webrtc-minutes.html#618b <- TPAC discussion
22:35:33 <markafoltz> ... Keep feeding packets with timestamps, the opus decoder should be able to see the gaps.
22:36:14 <markafoltz> ... Up to decoder to decide behavior, might skip, or emit silence, or guess missing audio
22:37:20 <markafoltz> adoba: Specific to opus. For others, do concealment/recovery externally.
22:37:35 <markafoltz> Eugene: May want to spec what happens for chunks with gaps.
22:40:02 <markafoltz> adoba: Different types of concealment, original vs deep concealment
22:40:37 <markafoltz> Eugene: Wait for Paul's opinion on this
22:41:22 <markafoltz> rrsagent, create the minutes
22:41:23 <RRSAgent> I have made the request to generate https://www.w3.org/2024/12/10-mediawg-minutes.html markafoltz
22:43:12 <cpn> Topic: AOB
22:46:47 <markafoltz> s/GitHub:  Expose in VideoFrameMetadata some fields from VideoFrameCallbac//
22:46:55 <markafoltz> s/GitHub: w3c/webcodecs#601//
22:47:06 <markafoltz> s/GitHub: w3c/webcodecs#813//
22:47:12 <markafoltz> s/RESOLVED: Merge #813//
22:47:42 <cpn> Chris: Marcos and I have been doing the self reviews for the W3C horizontal review, e.g., accessibility, internationalisation, privacy, etc
22:47:55 <markafoltz> s/RESOLVED: Merge #813 to close issue #601/RESOLVED: Merge webcodecs#813 to close issue webcodecs#601/
22:48:08 <markafoltz> s/Topic: #855, Audio Metadata/Topic: webcodecs#855, Audio Metadata
22:48:11 <markafoltz> s/Topic: #855, Audio Metadata/Topic: webcodecs#855, Audio Metadata/
22:48:19 <markafoltz> s/GitHub: webcodecs/#855//
22:48:47 <cpn> rrsagent, draft minutes
22:48:48 <RRSAgent> I have made the request to generate https://www.w3.org/2024/12/10-mediawg-minutes.html cpn
22:48:54 <markafoltz> s/Topic: Opus Packet Loss Concealment/Topic: Opus Packet Loss Concealment webcodecs#558/
22:49:12 <markafoltz> s/Topic: Guidance for user-defined VideoFrameMetadata entries #849/ Topic: Guidance for user-defined VideoFrameMetadata entries webcodecs#849/
22:49:30 <markafoltz> s/GitHub: https://github.com/w3c/webcodecs/issues/849//
22:50:03 <markafoltz> rrsagent, draft minutes
22:50:04 <RRSAgent> I have made the request to generate https://www.w3.org/2024/12/10-mediawg-minutes.html markafoltz
22:54:14 <markafoltz> s|webcodecs#813//webcodecs#601||
22:54:25 <markafoltz> s|webcodecs#601//webcodecs#813||
22:54:33 <markafoltz> s|#855//#855||
22:54:41 <markafoltz> s|GitHub: w3c/webcodecs#849||
22:54:49 <markafoltz> s|/github.com/w3c/webcodecs/issues/849///github.com/w3c/webcodecs/issues/558||
22:55:01 <markafoltz> s|<markafoltz> s/RESOLVED: Merge #813 to close issue #601/RESOLVED: Merge webcodecs#813 to close issue webcodecs#601/  <markafoltz> s/Topic: #855, Audio Metadata/Topic: webcodecs#855, Audio Metadata/||
22:55:09 <markafoltz> rragent, draft minutes
22:55:24 <markafoltz> s|rragent, draft minutes||
22:55:29 <markafoltz> rrsagent, draft minutes
22:55:30 <RRSAgent> I have made the request to generate https://www.w3.org/2024/12/10-mediawg-minutes.html markafoltz
22:56:56 <markafoltz> s|Topic: w3c/webcodecs#601 Expose in VideoFrameMetadata some fields from VideoFrameCallbackMetadata - youennf||
22:57:02 <markafoltz> rrsagent, draft minutes
22:57:03 <RRSAgent> I have made the request to generate https://www.w3.org/2024/12/10-mediawg-minutes.html markafoltz
23:04:12 <cpn> s/Subtopic: EncodecChunk metadata WebCodecs#245/Topic: EncodecChunk metadata WebCodecs#245/
23:04:19 <cpn> rrsagent, draft minutes
23:04:21 <RRSAgent> I have made the request to generate https://www.w3.org/2024/12/10-mediawg-minutes.html cpn
23:04:52 <cpn> s/Topic: WebCodecs metadata issues//
23:04:55 <cpn> rrsagent, draft minutes
23:04:56 <RRSAgent> I have made the request to generate https://www.w3.org/2024/12/10-mediawg-minutes.html cpn
23:20:54 <markafoltz> markafoltz has joined #mediawg
23:24:43 <markafoltz> markafoltz has joined #mediawg
23:42:59 <markafoltz> markafoltz has joined #mediawg