Internet Engineering Task Force Interim Meetings, 22 Aug 2023

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: IETF-CELLAR-20230822-1900

Description

CELLAR meeting session at IETF
2023/08/22 1900

https://datatracker.ietf.org/meeting//proceedings/

A

A

B

So can you folks hear me.

C

I can hear you.

B

Excellent how's everybody doing.

C

B

Not where you are too, oh, no.

C

It's everywhere does.

A

My microphone work.

B

A

A

My uh this setup, I used earlier, didn't work today, so I tried something new.

B

B

B

So I think I should probably wait for uh Michael to come back for in a minute.

C

I think the date on the notes is incorrect. I was wondering if it was the right document.

B

ah And I'm sorry, which, uh which date.

C

The one on the top of your screen.

C

This is April 25th, oh.

B

Yeah uh uh actually yeah uh I did not change that one oops sorry ah happy happier.

B

B

So uh Michael we are all signed in.

D

B

You think we need to wait for a minute for other people to show up.

D

I would say: Dave, maybe I, don't know who else.

B

If I was smarter than I am I would have contact info for him to message him.

D

Yeah I don't either, but you know maybe he's not on any of the 300 instant messaging systems.

B

D

Yeah, exactly you know: I need to write an app. You know to keep track of who to mess who which IM system to to uh use for any one person.

B

C

If only they could work all together.

D

Yeah imagine that eh.

B

Thank you inconceivable.

D

A

D

Got your your your message here ready to discuss thanks.

D

D

D

Didn't copy the minutes into the thing this time I referenced them. That was the.

B

Wrong and that and that's never wrong in the mind dispenser at least.

D

But he hasn't cleared the discuss yet.

D

C

Right well, let's start.

D

Pardon me cool.

C

Yeah so he said he cleared the Roman.

D

Roman may have, but John hasn't.

C

D

Any comments about the minutes, first of all, I guess, welcome everyone. All four of us.

A

Excellent I didn't read them, but I I just saw what you scrolled through, and it looks like what I remembered so I think there I think they're fine.

B

D

Okay, uh I think I got the status correct.

A

D

A

D

Okay, so um John still has a discuss open um but you're saying that it's probably closed. This is on here right here.

D

So are we, is he satisfied? The telechat happened. Didn't it.

B

uh I'm I am looking uh and if that happened, uh see because.

D

The rest of it is.

D

This is an update. My.

B

Previous ballot.

A

So I haven't said.

D

Everything below this is um already been dealt with.

D

Well, we can't verify the Errata. Only 80 can do that.

B

uh So, if we're doing, if we're having that conversation, let's see uh I can be taking notes, and for this uh thing uh we could say basically uh uh uh John so the erratic. So nothing has happened to the uh Arana, so uh they're still all reported.

D

Yeah well I think we verified them all since pretty much all been reported by us, okay right, but our ad hasn't marked them as verified.

B

D

Because you know 80s busy right, I, don't know what are we supposed to do? We found we found a mistake in our old document. We report an error on it, uh Errata on it.

B

D

B

Think I think that this is a uh emailed the iesg just following up on stuff does.

A

B

Like a plan or do you or do you want to do you want to uh give.

D

B

Minute to uh redeem himself.

D

No, no I think that we should follow up and we should say well they're, almost all reported by the working group. We reported our Errata, so the world can see and our A.D should mark them as verified.

D

So they should have the force of law since they'd be verified. He doesn't know if he.

B

Doesn't know if.

C

D

Is a good use of a rata, I I, don't know how else we would do it.

D

We, you know, thought about revving the document but that'll take a while.

B

Yeah well I mean we're back to the thing about having the uh having the Errata system only makes sense if it if the Errata system makes sense right so or if the actual Retta in the system makes sense. So.

B

And we want this uh worked as verified.

D

Yeah we want marked as verified, and we want John to please clear his disgust.

B

Yeah, we think that's the only thing that John is holding out on.

D

I, don't know I mean matroska tags. He says here right. Do we think this is there? Well, we thought we designed the whole thing to be informative. You do not need to read matroska's tags to implement matroska unless you want to implement matroska tags right so.

B

uh I understand I, understand his question so asking the dumb question myself: does he uh if, if if you or if you, if you implement matroska and you don't implementation tags is what you have is useful.

D

C

Should I reply to him.

B

Well, uh that that that and- and that would be fine too, uh the uh the other thing is uh that uh I just want to like I said I want to make uh sure that we send send him all the uh all the all the questions we have to get everything resolved. Does that make sense.

D

B

So um so, if the question uh is the is, is the metronica tags being normative.

B

Yeah but but was that what he was asking? Yes.

B

B

um So if I say one can implement uh matroska.

D

A useful version of matroska without tags, yeah.

B

um So we think this is informative and needs to stay. Informative, okay, cool.

B

um So I don't know uh Michael I, don't know how well you know John, but uh if, if if I was him, and somebody came back and said that uh I would I would uh have my head's been around once? uh Is there any way to characterize what a useful version of matroska without tags might be good for.

D

C

C

For me, the like it's it's very two separate things like the Metro scan documentaries.

C

If you look at XM and it's like describing XML and the mattress gas tag is just a one use of XML for something it's yeah I mean the mattress attack depends on the other one, the Metro scale document, but not the other way around. It's the.

B

Oh okay, so that so that's so that's the answer. That's the answer is that the batroska doc specification is normative for otroscot tags, but not the other way around correct.

C

B

That sounds I I'm, not the I'm, not the ad uh interpreter anymore, but uh that's that's a lot more! That's a lot more! That makes a lot more sense um to me just off the top of my head. If you know, if I'm imagining John reading my the email.

C

D

As I understand as I understand tags, okay.

D

um You know you could take a a video of a concert. Okay and you could tag in that video where each song starts right. Yes, you know, so you have your Grateful Dead concert. You can tag okay, each time they change, songs, right! Well, yeah! You read a if you build a if you writing a player and you.

B

Don't Implement tags then.

D

You just can't jump from song to song. You just have to watch the whole thing through or whatever right.

C

Yeah, but in our case that's chapters and tags are still something different.

C

It's like metadata that describes what's in the file, if you remove the metal.

D

You could use chapters for the different songs. Okay, so you know uh I could tag uh some of the songs as covers and some of the songs as Originals all right right. That would be a use of a tag.

D

Here's when the Grateful Dead played their own song, Here's Where, The Grateful Dead played a Rolling Stones cover.

B

C

B

C

The concept was on that date and the time of that song happened that time, for example,.

D

C

You absolutely don't need it for playback, but it's useful information.

D

Right, it's totally reasonable to build a player that doesn't know anything about these tags.

C

Right and in fact most players don't support metro scare tags at all, so.

D

Right because it's meta information, it's not information. It's you know, lets you catalog the item. You know it's more interesting for archivers than it is to anything else. Yes, okay, here we have, you know, uh um beats per minute right as a tag yeah.

D

You know, maybe that's a good one to give them. So we could. You know you can identify which songs in the thing are appropriate for the Disco and which ones are appropriate. For you know, Thursday.

B

Night, easy listening right.

B

So we're saying most media players that uh can use matroska uh encapsulations, can't uh don't support uh metronska tags.

D

They don't need to to play things back. Yeah.

C

Yeah and what I was going to say is also it's. Actually, we could remove all the Metro scare tax documents reference and you could still just Implement tags from metroska from the Metro scare documents. The mattress catex document only provides a list of known tags, so everybody uses the same ones.

D

Right and I I think they're in they're in strings even right. Yes, yeah.

D

So they don't even need to go through uh Ayana process. They just yeah they're just strings all right. Let's send that email and get them to.

B

Quiz is there: is there anything else that we need to ask them about right.

D

I, don't know to clear his disgust, that's it really. That was the the item here.

B

B

B

D

All right anything else from matroska.

C

No uh glad we clear the encryption box because yeah I didn't understand the whole thing at first, so I can dig a little and now it's clear.

D

Okay, good all right, so let's talk about. uh Let's move on to Flack, then there's two unresolved issues and.

B

That's what that's one! That's one issue uh twice right.

A

Yeah well, the issue here is: what should I do with those two references? References should I make them normative or should I remove them.

B

So so uh did did you did you see my did you see my? uh Did you see my note about this from yes? Yes, whatever.

A

Day that was I'm still, not really sure what to do. Are you explained what the issue is here, but um I I still don't know what to do. I can remove the implementation status section, but I think it's a nice addition. So.

D

Usually, implementation status uh gets removed um by the RFC editor.

A

Yeah I read that yeah.

D

A

D

And so you know you can go in the document. It can go in the shepherd report right up. Okay, um so, but we have a reference to 7942. Oh okay, this is silly but okay.

B

Yes, yes, it is um so the so the uh so the so the answer, so the answer is uh in in the mine dispenser is that the first one that describes uh how we uh request uh uh Diana registrations for what we're registering uh I, think you can remove that one now, okay,.

A

Remove that whole paragraph I can just remove the text that says, uh in accordance with the procedure set force in RFC 6838, the document registers one New Media type. We can remove this whole. It's.

B

uh I I think you could even say this document uh registers, uh one media type.

A

D

Don't know why I don't know why this I don't know why that either them wouldn't be informative anyway,.

B

D

B

D

Don't need to read that you don't need to read the procedures in order to implement the the protocol. You need to read the procedures in order to publish the document. Yeah.

A

Right I can just rephrase that and remove the first section and say this document rests just one New Media time: okay,.

A

I'll remove that.

D

Okay, so that was section.

A

Three, this.

D

One and same thing for 79.42.

B

uh Almost uh remind me what the first RFC number was.

D

B

That we were removing.

D

I'm sorry, the 838 right.

B

B

D

uh I, don't I, don't know. I I actually really disagree with Marie on this special approach to your following. That doesn't mean it's normative anyway, I don't anyway, I think he's anyway, it doesn't matter. Let's just remove it.

B

uh Well, so so let me let me, let me be clear clear. Let me make sure I understand what's going on so um if we have a, uh if we have an implementation uh status section, then uh somebody will ask: did you use the boilerplate and the boilerplate uh ref?

B

You know the boilerplate uh references, the RC that the board of players taken from so if we copy the boilerplate in that would give us a reference there, and so, if we still got the implementation status section, we should uh we should have the reference in there someplace and.

B

B

If, if we're having this, so let me let me let me ask the other question: uh we've gotten all the way through, we don't know we haven't um to me. uh Michael tell me if this makes sense uh to me. It makes sense to uh leave the uh implementation status section in the draft and because Maria is thinking that the reference that RC needs to be normative, we would move it to normative.

B

Is that does that make sense.

A

Yeah I I think there's no problem in making it normative. I just wasn't sure whether that would be having any other consequences.

B

Well, so the other thing is.

D

You know it's a it's a moot point. If, if you actually follow the whole thing which says you should remove the section before publication yeah and it's a moot point whether or not it's informative or normative, yeah.

A

D

B

Only and the only reason that it matters um that any of this matters is that's removed on publication by the RC editor, but it's not removed before then.

B

um You know as we're going through publication requested uh processes. So people will, you know, look at that and say: did they use the boilerplate or you know things like that uh and also we will uh also the boilerplate or below the boilerplate. uh It says also put a note into the RC editor saying: please remove this. uh Please remove this section before publication and uh please remove the corresponding reference.

B

That is on. That is only used in this in the section, and uh so then it will be where it needs to be for any reviews, any review teams or any other uh ad ballots, and then it will magically disappear uh before it's published as an RFC.

B

Does that make sense.

D

Yes, so just add the sentence here, edit RFC Editor, to remove the entire section before publication.

A

Okay, yeah, okay,.

D

A

Because uh yeah well, people will find the there's no problem, this RFC being from before, linking to GitHub with updated application stuff.

B

And I apologize I would like to apologize to the entire team for uh that we have these two references that we are handling differently than we would uh handle normative references or informative references normally because they are expect they are both different special cases. So I apologize to the team.

D

Yeah I don't understand why you would complain about it being normative or informative I anyway, I just you know, whatever.

A

D

Yeah: okay: let's: let's move.

A

On here, okay, I I think the reference to Org the AUG specification is much more important. For example, that's.

D

Yeah yeah formative.

A

Reference but well, if you really want to implement flocking Oak you, it need to need to be a normative reference, but orc mapping is well. You.

D

Need it's optional, yeah.

A

Yeah it's optional and the problem is that's an informative RFC anyway, so I couldn't link it normative without uh downdraft, so yeah I think that would make a lot more sense to me have Fizz about, but yeah.

D

We should be arguing over that point. Not this one. Yes, I agree with you completely okay, so uh the IEC document we've established that it's not necessary to read it.

A

D

Okay, good, so let's go on to the other point.

D

A

This is a very.

D

Old close-up business just means the end of day, but anyway, uh that was Spencer said anyway,.

B

D

uh That was from you, okay, yeah.

A

um We've already been this, this emo I think has been marked as well uh handled. Okay, all the points in this have already been processed. You know.

D

Okay, fantastic version.

A

Eight of the document.

D

So it's all done in version eight, that's what we should just tell the 80 uh we need to do. We need to do. We need to do the removals. Okay, so we're gonna have version nine yeah, no.

A

No, the email you were looking at was from the Shepherds review, and that was done in version eight and.

D

A

Just discussed were from the idea, review and I published version 10, but I will publish another version 11 and that should address all comments.

B

And we're on 10 now I'm just trying to keep up right.

D

A

D

Perfect anything else.

A

Not for flag, no I I think we can manage now.

C

Nothing for me what's up.

D

All right, well I, think we might be done. The meeting.

B

uh We can ask, is there any other business?

B

D

There any other business.

C

A

Not that I don't know.

D

D

When is the next meeting? Oh.

B

I think we're back to normal in September.

D

D

Okay, uh and uh are we going to ask for a uh slot for ietf 118 in Prague, which is at the same time as um um you know,.

C

um You're done to it.

D

No time to wait, thank you. I had the initials, but not the other thing, um because you're going to be across town, so I, don't I, mean I. Think it's useful. Should we just I, don't know I, don't know what to do exactly there um like.

D

B

Don't know so so uh this is. This is only me talking out loud, but we have talked about this before so if we know a reason to say, oh we're going to do something at uh no time to wait uh at this time on this day in this place, uh it makes perfect sense for us to.

D

B

It's perfect for yeah to announce it and um flip the okay.

D

B

uh Do that, if it's, if it's you know, basically that people are just.

D

Outing out, it's an out of Hotel uh side meeting.

B

D

B

D

Okay, I think that's fine all right, so we won't ask for an agenda slot on the official thing, but we will let people know via the side meeting and whatever stuff. That is also happening.

B

Excellent, okay um and- and uh just just so everybody knows because I don't know who knows what um the uh there is a there is a uh wiki page uh if I'm remembering correctly, that that's what it is for uh side meetings for each ITF meeting, so we would basically just need to put an entry in for that.

B

um And if you, if we wanted to have people, uh if we wanted to have people from the ITF possibly show up over there,.

D

B

It's uh and we could. We could do a four minute presentation on uh as part of the hot RC thing on uh on Sunday night. Okay,.

B

If, if people think that would be useful.

B

D

You know, do you know the chair of that thing.

B

um My wife sleeps with him.

D

Oh three generous of you to let her do that. You're.

B

Welcome, oh, the pleasure is all mine. Did I say that out loud anyway, yeah.

D

There we go anyway, okay, all right anything from Michael who may not have a microphone I, don't know.

D

D

D

All right, sorry, anything else.

D

We said that already going twice all right good day good evening to you: okay,.

B

uh And uh Martin I've got I've got I. I did actually put the uh like specific text in the uh uh in the notes for the what to do with uh 79.42.

A

Yes, thanks: okay.

B

Perfect. Thank you. Okay, excellent bye,.

B

um You guys rock, and uh thank you especially for your help in uh helping me focus on what the last curiosity thing for Murray is.

B

um If you you, when you make the when you make the changes to uh uh submit Dash, 11.

B

uh I would like I I should read through it one more time uh because I have been reading through it, but I've slept since then, and we've been still changing, making changes and then I can then I can tell Murray that uh it's. You know that we've addressed the all the issues uh from his evaluation and that it's ready for uh it's ready to go for last call.

A

Okay, I'll, let you know.

B

Thank you so much.

A

Would you like a div or um just the plain documents.

B

I think I I think I. Think I. Think part of the answer was that the uh what is it like, the diff from eight to nine or whatever it was well? It had a lot of spelling Corrections in it. Yes,.

A

Yes, that's correct.

B

Yeah, so I, don't I, don't think I actually need that um because I'm going to be, you know basically I need to if you, if there are enough, if they're, if there are enough differences, I'm going to be reading the whole thing anyway, so I, I, honestly I, usually when I'm looking at this I, usually just use the diffs to tag. What sections to look at and then I just read the new stuff. Oh.

A

B

A

I'll send something to everyone.

B

That would be perfect.

A

Okay, oh um have a good day. I'll go I'll, go I, have to um or go to bed.

B

Hey technically I'm at 3gpp meeting in uh in uh gottenberg so uh this week. So my my that's: why I'm more incoherent than usual? My days are starting about midnight. My time have a great one.

A

Yeah have a good meeting.

B

Yeah take care.

B

B