21:02:04 RRSAgent has joined #mediawg 21:02:08 logging to https://www.w3.org/2026/06/09-mediawg-irc 21:02:41 Zakim has joined #mediawg 21:03:17 Meeting: Media WG 21:03:20 Chair: Chris Needham 21:03:25 Regrets: Marcos Caceres 21:03:42 Present+ Xiaohan Wang, Greg Freedman, Francois Daoust, Chris Needham 21:03:50 Agenda: https://github.com/w3c/media-wg/blob/main/meetings/2026-06-09-Media_Working_Group_Teleconference-agenda.md 21:03:59 present+ Gabriel Brito 21:04:23 GabrielBrito has joined #mediawg 21:06:47 Topic: EME editor 21:06:59 scribe+ 21:07:19 cpn: Thanks Xiaohan for stepping up as EME editor. 21:08:08 ... We'll focus on issue #545. Goal is to have an editing session. Minuting will be lightweight. We'll just record decisions if any. 21:09:09 -> https://github.com/w3c/encrypted-media/issues/545 Issue #545 Notes contain normative requirements 21:11:36 present+ Nishitha Dey 21:14:26 About the note that talks about "proprietary format", people present in the meeting understand it to mean formats that are not in the initdata registry. Chris will clarify the note. 21:18:15 present+ Jer Noble 21:24:40 The note came from discussion in https://www.w3.org/Bugs/Public/show_bug.cgi?id=27093 21:29:32 The group discusses the note in step 3 of section 3.2.2.2, and will look into turning it into a matter of fact that the implementation may reject the configuration later on. 21:31:35 In step 21., the "MUST" could be dropped to also turn the note into a statement of fact. 21:34:42 In section 6.1 MediaKeySession Interface, the note about keyStatuses seems to define algorithmic steps to keep the map consistent. Is that already encoded in the Update Key Statuses algorithm? 21:38:52 That algorithm (in section 6.6.2) also contains a note with a "MUST NOT" that could be turned into a "will not" because what it states is a consequence of the algorithm running in a task. 21:40:37 Back to the note in 6.1, for the same reason, "The map MUST NOT ever be inconsistent or partially updated" can be rephrased into "The map will never be inconsistent or partially updated". 21:46:43 The requirements on Key IDs can also be reformulated: "Key IDS are not removed if they become unusable". Same for the rest. The note just states facts handled by other normative algorithms. 21:54:15 On the other hand, the following note that has "the map is populated with a single pair containing the one-byte key ID 0" describes how things are done in implementations in older platforms. Ideally, that would be encoded in an algorithm that describes how keyStatuses are computed. The note can probably be dropped entirely because all newer 21:54:15 platforms expose key IDs. 21:56:50 Chromium implementation now requires newer Android platforms in particular, so the code could be removed from the implementation, and the note as well. 21:59:17 cpn: Are such editing sessions useful? Or would async work better? 22:00:10 ... Plan: create PR for the notes discussed today, and sync up within the next couple of weeks. I'm happy to create the pull requests. 22:01:12 RRSAgent, draft minutes 22:01:13 I have made the request to generate https://www.w3.org/2026/06/09-mediawg-minutes.html tidoust 22:01:34 rrsagent, make log public 22:02:57 i/... We'll focus/Topic: Issue 545 Notes contain normative requirements/ 22:03:08 rrsagent, draft minutes 22:03:10 I have made the request to generate https://www.w3.org/2026/06/09-mediawg-minutes.html cpn