Session Date/Time: 26 May 2026 19:00
Spencer Dawkins: Well, this may be a very short meeting.
Robert Sparks: You say that like it's a bad thing.
Spencer Dawkins: We should give it a couple of more minutes.
Robert Sparks: Yeah.
Jérôme Martinez: Check, oh.
Robert Sparks: Hey Jérôme, how are you?
Jérôme Martinez: Hello, fine, thank you.
Robert Sparks: Excellent.
Robert Sparks: I see Michael popping in also. So this is most excellent.
Jérôme Martinez: Dave rice cannot join us today, so he's excused. So I think we can start.
Robert Sparks: Okay. I'm making one slight addition to... Okay, cool. And we have Charles.
Robert Sparks: Oh, excellent. So all the cool people are here now. This is excellent. Excellent. So we are just calling ourselves to order and I can sign in our Area Director, so... Oh, okay. Whoops, sorry. Okay, so Charles is signed in. And so we are meeting and we remind ourselves that this is an IETF activity covered by the IETF Note Well, which is in the agenda and various and sundry other places. Contributors, please be aware of obligations under the Note Well. We did have previous meeting minutes from our meeting in March. And do we have any corrections that we need for that?
Jérôme Martinez: They are fine for me.
Robert Sparks: Okay, thank you. Um, and, uh, just just so that this is obvious, I actually do review these after the meeting. So, uh, but of course, I'm the one typing them. Uh, so it's good to have somebody else reviewing them before we approve them. But, uh, I also look at them. Um, we get to our working group documents. Uh, we have cellartags now in the RFC Editor queue. We will stop for a short celebration. And thank you, Charles, for your support with that one. And, uh, just to double down, uh, we have, uh, the cellar-codec draft that is now in, uh, IETF last call. Uh, and I have not seen any feedback on that yet. Uh, I don't know how quickly we start expecting, uh, review team input. I'm trying to remember.
Robert Sparks: Well, it usually is driven by the last call deadline. So, I would expect it within, you know, five days before the last call expires.
Robert Sparks: Cool. So, we'll continue to watch for the, uh, anything that pops in on the mailing list. Um, excellent. Um, do we have any updates about, uh, cellar chapter codec or FFV1 V4?
Jérôme Martinez: Nothing for chapter codec. So, FFV1 V4, there is an edit PR for the CFA buyer, RGGB support. And so, I didn't finish yet, uh, for having the floating point support. So, I will send soon a new PR for the floating point support. So, the floating point support is already in FFmpeg code. And RGGB support, buyer, uh, there is still a draft on FFmpeg Dev about how to implement it. And there is a discussion about how to implement it. I will I will write the link to the PRs.
Robert Sparks: Okay, excellent.
Jérôme Martinez: So, it is moving on that side.
Robert Sparks: Perfect. Uh, so that is happening, uh, now. Um, is this is this still, uh, Quarter 4 2026, at the earliest?
Jérôme Martinez: I still hope so, yeah.
Robert Sparks: Okay, excellent. Um, Charles, uh, just just, uh, touch base on this. Um, this is another one of the, uh, this is another one of the technologies that the IETF took in, uh, spent, uh, that we developed, uh, informational specifications to capture the specifications as they exist in the wild, and then, uh, working on, uh, working on, um, standards track documents that were, you know, developed, uh, completely within the IETF process. Uh, Matroska was that way, uh, and, uh, the FFV1 codec also, uh, being that way. Um, thank you also. I Charles, uh, please feel free to keep dropping in on these meetings, as long as we're, especially as long as we're starting you out by saying, thank you, thank you, thank you. Uh, thank you for, uh, helping us with our recent errata for RFCs that we, um, have been processed since the last IETF. Um, I am noting that we have a reported errata that had not been, uh, processed, um, and so I just stuck the link in there for that, and I'll, I'll work on, uh, proceeding through that. Um, and that starts, that starts to be, uh, relevant when we get to, um, our next topics on the, uh, on the discussion. Um, Steve has a PR, uh, that would create a, um, EBML, uh, corrections draft, um, which I think, uh, among other things, would would, uh, capture, um, all of the helper helper document update, and, uh, approved errata that we've, that we, that we, uh, are doing there. Um, and Steve was asking about, um, going ahead and submitting that, uh, that specification, um, and we've been thinking going ahead and doing that as a working group draft. This is basically corrections to, uh, stuff the working group has done. So, uh, that, that seemed like a good plan to me.
Steve Lhomme: It's not actually corrections. It's right now, the document, it's only additions. Okay. And so, basically, last time we discussed about additions to Matroska, but to do that, we need additions in EBML, which is not in the format itself, but in the schema, to to define the format, which is also part of our RFC, uh, 8794. And so, uh, the idea, like for Matroska, is to do, uh, the same thing, and update one version of the document that would have a few attributes to the schema, uh, and errata. But, there could, uh, I don't know if we want to include some erratas. I don't know. It works if we keep them separate, or if we put it in that one.
Robert Sparks: Please, please combine them in with these docs.
Steve Lhomme: Okay. We could also add that, and I think there was some, uh, other pending stuff that we could add as well. But, for now, I just concentrate on the part that I need, to actually do the the other document, because, as it's a schema, it's used to validate, uh, the the changes. Right now, it doesn't validate because the old schema doesn't have the new attributes. Yeah, yeah.
Robert Sparks: So, yeah, Charles, the the EBML schema is what is what we use for Matroska. So, uh, we need to be able to keep those in sync. Uh, so, that's what's going on with that.
Robert Sparks: So, it I can ask Charles Eckel if it's all right with you, um, please, please do. There are these two documents that are being proposed as working group 00s. Steve's actually queued up submission for one of them, and it's waiting for us to say yes, please, let it into the repo. Both of them take us to the edges of our current charter. Do you have any objection to us going ahead and accepting these as 00 documents, anticipating that we will recharter in a way that makes it very clear that they're within our chartered scope?
Charles Eckel: Uh, yes. I, I believe so. Uh, but, one question I did have, is I'm, I'm still a little unclear if, in both of these cases, um, so the first one, just based on how it was named, the, uh, EBML, uh, corrections, I thought for sure would be a Bis version. And it sounds like if it pulls in errata, it will be. Um, but then, the Matroska one, if it's, I wanted to understand, is it purely additions? So, it's not even an update. It's just a...
Robert Sparks: Yeah, so it's, it's an extends. And the the other one, are just extensions to EBML. I'm not expecting that that would be a Bis document either. But, if there are, if there are, if we do fold in errata, that um, address, um, well, I can't think of any way around this, it would also become an updates, right, if we're folding in the errata. But, I don't think it's a replaces.
Charles Eckel: Okay.
Steve Lhomme: No, but it's like we discussed the last time, like, with the replace, that means we have this huge document that needs reviewed, that, of, for 95% have been reviewed already. Uh, there's no point in going through that again, I think. Right.
Charles Eckel: Okay. Um, all right. And, and then with Matroska, if it really is just, there's no outstanding errata that we'd be pulling into that one. This really is just purely additions.
Steve Lhomme: Uh, there might be, but they were mostly, bogus text or things like that.
Robert Sparks: And, an errata should, might show up while we're working on this document, and, yeah, convenient for us to address it while we have the open work item.
Charles Eckel: Okay, so we just need to be clear in the document, you know, there would be the, if there is an updates, there would be a succinct explanation of updates to, you know, uh, and the rest of it should have clear wording that it is just extending, you know, the schema and then uses, using the extended schema in the other document, as I understand where we're going.
Steve Lhomme: Okay.
Charles Eckel: Yeah, I think it would be helpful if, at the time that these are accepted, as, uh, adopted by the working group, if we're, we're clear on, for EB, uh, EBML, if it is going to be handling, uh, the errata like we believe, and if so, then, uh, making sure that initial version, um, says that it updates, uh, the RFC. That, that would be a good thing to have. I didn't actually check and see if it has that. And then, for the Matroska one, I, I guess it would just be in the text saying that these are, these are additions, and there is no update relationship, at least at this time. Uh, it'd be adopted that way. If some errata does come along, then, we could always revisit that.
Robert Sparks: Yeah, sure.
Steve Lhomme: By the way, I renamed the draft for EBML, uh, update one, like Matroska, but I realized it's, it's actually called corrections. So, I suppose I should change that for now.
Robert Sparks: So, cancel the existing submission and make a new one?
Steve Lhomme: Yeah, except I don't know how to cancel it, because once it's sent, uh, like...
Robert Sparks: Okay. So, Spencer just just reject, just reject. Yeah, reject. We'll we'll reject it as chairs. Yeah. Yeah. IETF, you can count on your IETF management to say no.
Steve Lhomme: Yeah, it's just that from my side, I, I see nothing, so I don't know if it's possible.
Robert Sparks: Yeah, yeah, exactly. So, that, so, uh, am I getting correctly that both of, uh, the result for both the EBML and Matroska, extensions, um, will be, uh, will be, uh, updates documents?
Steve Lhomme: Yes.
Robert Sparks: Yeah, let's just put in updates, anticipating that we'll be addressing errata. If we get to the end of the game and there's nothing that actually updates the doc, it only extends it, we can remove the tag. Yeah. Perfect. Okay. Um, so, this is stylistic, but I've got to, I've got to ask. Um, are we doing Matroska V4 extensions or additions? I thought these were additions.
Steve Lhomme: Uh, yeah, they are additions. Right now, the document is called, uh, update one. So, Yeah. Yeah. But, yeah, that's it.
Robert Sparks: I'm, I'm, and I'm actually thinking more the title of the document, rather than the file name. The file name matters also, because people think they know what's in it, because of the file name. But, just for the, just for the, um, just for the, uh, just for the, uh, good title.
Steve Lhomme: It's called, Matroska Media Container Version 4 Update One.
Robert Sparks: Mhm. Okay.
Steve Lhomme: And so, basically, for EBML, I don't remember what I put. Uh, Extensible Binary Meta Language Corrections for RFC 97- 8- uh, 94- 87 94, but, actually, we should be update one, for RFC 87 94.
Robert Sparks: Okay. So, um, tell me if I've got this, uh, I, I don't know for sure that everybody is looking at, uh, the notes at this very second. Let me just, uh, share my screen for that, um, So, this is, this is, this is what, this is what I'm showing. And please feel free to, uh, type corrections if, if, uh, if that's a good thing to do.
Steve Lhomme: Yeah, it's correct. Uh, I don't know if you want to put the updated name for the EBML one or it doesn't matter in the discussion.
Robert Sparks: Okay. Yeah. Oh, you want me to write it down? So, um, I think I can call that, I think we can, I think it's fine to call these updates, um, and I think it's fine to call the Matroska one, Update One, um, but, basically, the idea is that both of these will end up with a, uh, updates tag, um, at the, in the RFC Editor. So, uh, again, both of these will kind of look the same. Um, and that should actually be, I'm just changing the case on this, so that this is somewhat more clear. I'm not talking about the file name. I'm talking about the, um, the, uh, the tags at the, at the top of the RFCs, or of the Internet Draft. Um, cool. So, I feel like we're getting very close to the point of being over-wrought on our naming discussions. Yeah. Yeah. Yeah. Yeah. We can just let it, let it go. So, Steve, the submission expired. And, I'm surprised that the syst- the, the, your previous submission of Draft IETF Cellar EBML Update One, um, canceled itself on the 17th. Um, surprised that the system lets that happen, when it's pending chair approval. But, it did. We'll just take the track issues for that. But, the point is, is that there's nothing that anybody needs to do at this point, except that you need to submit a new 00.
Steve Lhomme: Yeah. Okay. I'll do that.
Robert Sparks: And, and, uh, Robert, can we reasonably do that with both of these documents?
Spencer Dawkins: That was my understanding with our conversation we just had with Charles.
Robert Sparks: Perfect. Okay, excellent. In that case, um, I think we are, I'm just typing, we will, um, submit, uh, -00 working group drafts for both of these. Yeah. And at some point, we need to propose text for rechartering with Charles. And, my attempts, my first runs at that, I got very frustrated. So, um, anybody has solid ideas, now that we've had these conversations about what that text should look like, I would appreciate a first stab.
Steve Lhomme: I don't even remember who wrote the first charter. Do you remember, Jérôme?
Jérôme Martinez: No. No, I don't remember the name.
Spencer Dawkins: I'm sure we got the question. Steve, try again.
Steve Lhomme: Jérôme, do you remember who wrote the first charter for the group?
Jérôme Martinez: No. No, I don't have the name.
Spencer Dawkins: I think it was, just at a base guess, I think Michael was the pen holder.
Robert Sparks: I, I think, I think that, um, I think we've rechartered once previously, and if that's true, it's probably Michael. I don't, I don't, Michael was not involved at the beginning. Uh, he was not one of the chairs. So, uh, but, I think we have, I think we've rechartered once since the, uh, original chairs. So, um, is is it is it so, we've, uh, we've got a working group wiki. Uh, is that a decent place for us to, that everybody should have access to, honestly. Um, is that a decent place for us to talk about, uh, proposed text?
Steve Lhomme: It's probably more convenient that GitHub repository, where you have access and edit online, it's very convenient.
Spencer Dawkins: I would just use a Notes Page, myself.
Robert Sparks: Okay. Cool. We will do that. And, um, is there any other business?
Steve Lhomme: No.
Jérôme Martinez: No.
Robert Sparks: And, I think, I think we can have a short meeting this week. Um, actually, um, I should ask, um, is anybody working on PRs that, uh, or issues that they need to discuss at this meeting?
Steve Lhomme: Uh, for now, I was mostly blocked on the EBML one, and there's a lot after that, that work can some of the work already prepared, but has to go in order, so.
Robert Sparks: Cool. So, we will plan for our next meeting to last more than, uh, than, uh, 30 minutes. Or maybe it's straightforward. Well, perfect.
Charles Eckel: And, and, let's plan on having, either time in that meeting or before, to, uh, work through the charter, uh, that charter. Yes. Yeah, we I think it's, it's important to get that done, and we don't want to block any work. But, it, it, if it's not a big change, but, we should clean that up before we we get too far.
Spencer Dawkins: For what it's worth, I'm quite likely to miss the June meeting.
Robert Sparks: I am planning on being present. We will, we will...
Spencer Dawkins: Uh, the date?
Robert Sparks: It's currently June 23rd.
Spencer Dawkins: Mhm. And I will be, um, somewhere random in New Zealand on that day.
Robert Sparks: Oh, good for you. I'm going to try. Would you, uh, what if we had a what if we had an in-person meeting that week? Never mind, never mind, never mind. I'll just be jealous sitting here. Um, but cool. So, I will take, I will have that as the action to discuss, uh, proposed charter text, um, for the next meeting.
Steve Lhomme: Also for the charter, is the mailing list a good place to discuss that?
Spencer Dawkins: Yes.
Steve Lhomme: That's what I thought, but it's better to ask.
Robert Sparks: Cool. Excellent. And we're still getting out of here in less than 30 minutes. Oh, fantastic. Thanks, chairs, for a well-run, uh, uh, meeting. Yeah, and, uh, uh, thank you for being here, Charles, and, uh, we will not tell anybody that this meeting ended early if you wanted to if you need to go grab lunch or something. Yeah, thank you. I'm finishing it up here on the side. So, yes. This is great. We're there We're there We're there for you, guy. Take care. Bye-bye. We'll see you We'll see everybody next time. Bye-bye.