Internet Engineering Task Force 104, 26 Mar 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: IETF104-EXTRA-20190326-0900

Description

EXTRA meeting session at IETF104
2019/03/26 0900

https://datatracker.ietf.org/meeting/104/proceedings/

A

All righty good morning, everybody, according to the clock, it is actually time to turn around and do the thing welcome to the second extra extra session.

A

Yes, I wasn't going to go there, but someone.

A

B

I'll say it as I say it note the note well well, yeah.

A

Well there, it is again the note well yesterday, our agenda for today is imap4 Rev Thank, You Lexi Lucius are going around Alex. You can talk about I'm at 400 of we're going to talk, possibly some more about each other and that's everything on the agenda today. So, let's right over to your Lexi.

B

Note that there are two to a DS on the author.

B

If, if you, if you want to be an ad, just ask us to be a co-author on this time,.

A

Hey is it a cherub dispatch.

A

B

A

B

You out to pasture you go to chair.

A

B

A

C

Right, I'm not going to already right through goals, I think we're nearly done with the process. That's you know, although it's an interesting thing, because I obviously had you know some ideas by the way of trying to optimize for clients or for the server and then team of the send a bunch of comments, and you can see that both might be.

C

I'm just talking blah blah blah blah blah well above are some doesn't.

A

Turn into this one's good, you can be smug.

C

Right so, okay, let me just watch.

C

Do you hear me it boxing a bit a little bit? Okay, so pretty quickly, hello goals, my client so I devote you know, client developers, life he's here.

A

C

Very common extensions format for everyone into the base back, hopefully be able to run two on the same.

C

Port right so last time, I had two slides sort of you know things to add and set things to remove and clarify the good thing that one of them is empty. Now, so I think most of clarifications are done. So we agreed last time to add the move command to the base back. It's quite commonly used, clarified and solicited fetches that they might continue IDs.

C

Its simplifies, quite significantly else up, was taken out, search response, recent flag, unseen response code and select so all the staff that people either there are better ways of doing this now or they're, just pointless things that service need to do and Glines don't really care about these days, I also found looking at RFC 22 31 about parameter. Continuations I found some text updating, I Maps back, and this was actually never mentioned an IMAP. So this was a like RFC who are updating. Rc 2060, which is imap4.

C

I, know that many service do this already so I just made it clear.

C

Things remaining I think more examples explaining use of body structure with various complicated messages, especially with nested messages, and for that would be very useful. The still a significant amount of bugs and various implementations in the yesterday's again, if you have great examples, send them to the mailing list. That would be very helpful, but if not I think we'll come up with vari and I will come up with something yeah.

B

Do we think, am I coming through the speakers? Okay,.

A

B

We think we can actually explain this in text better or that we just want to do it with better examples or more examples, because yeah I know the nested messages in particular, especially complex ones, have just been a nightmare. Yeah.

C

Mustard messages they have special numbering so they're always a special case. I can tell you so I'm. Recently, writing I'm a client again and I have to rewrite code three times to actually got all the cases, and this is me so you know this is pretty complicated.

B

C

Would start with examples, because the document has one example, but it doesn't demonstrate lots of diverse complexity. Every time I need to test it. I test it against my own server, which you know similar to Saracen right yeah, so I would say, add a couple more examples, and then we can see if we can improve text.

C

The only other major additions are edit addition of list extended, as we already agreed, allowing status in list and.

C

One clarification that needs to be done is binary fetch, so we discussed it last time and a little bit on the mailing list. I think we agreed to add the fetch side of it so that you can fetch by any body parts in binary, but we decided not to add app and side.

C

So when you can upload them, it keeps it a little bit simpler um and Chris raised the question which I had on to do list, but it's not reflected in the message in their in their latest revision is we need to be clear whether binary fetch only applies to live body parts over that applies to intermediate nodes. So what does it mean? If you fetch binary a multi-part mixed?

C

Do we want it to work? You know so.

B

The intent of binary is to retrieve a binary body part not to retrieve in binary mode a segment of the message. So I think we should just stick with that and say that binary is a single body part, but and yet a single leaf body part does anyone. We have a real use case where they need to use binary to retrieve a portion of the tree that isn't just a leaf body. Part.

C

Well, II, I suppose what I'm thinking about is. If you have afforded a nested message, you want to start with this message: Eric see body part and download everything below it. Basically, and if it's in binary, you want to get it in binary.

A

This is the the case where it confuses clients have to check. It is a message rch, it's who did I need to download it. It's not fine or otherwise. I need to download this binary to get the binary representation, because the binary representation of a message are CH you to use just the vines.

B

I suppose my sense would be that okay are people really doing that, as opposed to my sense? Is that if you're gonna do that kind of thing, you just download the whole message in the first place, you just do a fetch, RFC, 822 yeah.

C

Except for whether it's message, digests of you know, hundreds of other messages, I know I know it's a bit contrived. Example: it's rather complicated.

B

Use case, but maybe we can say it applies only to leaf nodes and to message. Rfc 822 part sources, Ned.

A

Would like to say to the microphone message: slash global yeah.

C

Yeah, that's the being.

B

A big part message, body parts well actually.

C

Actually, there is extra issue with message global, because message: global now allows nested in codings and I actually haven't looked at the text to see what makes sense.

A

Kimmo would also like certain microphone men problem with multiple parts or multiple messages is handling unknown, City response code. Since you don't know what exactly it applies to right.

B

Yeah I mean that's one of the novel's with fetching multiple chunks.

C

B

All right, well, fine! If there are use cases for other than leaf nodes, then we should allow that.

C

Sorry I was really playing devil's advocate. This just came up um yeah. Maybe we should think a little bit more.

D

Yeah I'd like to say you know the increments.

E

D

Chris Newman sorry yeah implementing this for a leaf node a leaf part is pretty straightforward on the server, but implementing this to support yo, a subset of the message gets more complicated and if you actually wanted to implement, you know decoding the nested encoding case message: global, which I never would then it gets even more complicated, so I I would say we you know I would say the trade-off. Is we kind of want I met four rep to to actually deploy, so we want to minimize barriers to getting deployed within reason.

D

So that's why I would pretty strongly prefer leaf. Node. Only okay.

C

Now III think actually the complexity argument right, I think you won over me because yeah you need to potentially you know, do I, don't know decode them into separate files and merge.

B

This together, this is the together yeah.

C

B

Yeah, that's why I was saying only leap are side. Same thing to Chris said why don't we put only the leaf body parts in the spec and as we review it, people can say whether this whether they suddenly discover well.

C

Yeah I I think possible outcome of this discussion is we need to review how message global works with binary anyway? So maybe we'll do we'll do revision of the binary expansion, which includes other bits. You know later on sure Nate.

A

Would like to sale if only it's fine for him as a starting point: okay,.

C

So well as long as we tell it, you know it so.

B

Let's put leaf only in the spec, with a note that says: please review this part carefully and make sure that you don't have real use cases for other for other answers.

C

Yeah there are a couple of remaining bits from I think we so bran did a couple of polls last time about various bits: I'm, not sure that it is any strong feeling. Any of this, so I suspect. Probably this will be dropped.

C

But if you have an opinion.

D

F

Newman I just.

D

I really like search res. It's it's very simple to implement. You know nice little spec, so you.

C

Don't don't get a problem for me, I'm the author, yeah.

D

Okay, fine I think you know one less one less option since its you know wide. It's easy to implement it widely.

C

Okay, if we can minute this yeah.

B

Nobody's gonna make a case for status deleted right well beyond.

C

A

C

This is, this: is a it came sort of related to quota. You know how much storage you can free if you expand this mailbox type thing, but so.

B

We can just do it.

B

C

Depends on your quota resource type, yeah, yeah, yeah, happy to dot the second month, okay right! So going back! There are a couple of changes to relate it to list extended that we need to do clarify binary fetch where it such as.

C

We should be able to do this, material I would say, and then I thought you know there is nothing else and then Tim assigned a long message yesterday, thinking up on all the things that I forgot about, which is actually was very good. So let's just go that I didn't have slides for this Elizabeth.

C

Yeah just keep scrolling gathered if this is straightforward. um I.

A

Had a bit of a complaint about this, that you actually have to keep track of every single yard, I believe you've ever told a client ever with that client might reconnect.

C

Right so just keeping existing language yeah, it's an well-known deficiency, Ian Cyrus based implementations, at least not beyond our, so it doesn't increment huge evil ADT when the rule is renamed.

B

Past it too quickly what what it is, what.

C

Is so the issue is.

B

Must be greater must be.

C

Different okay, if you go.

A

Yeah, if you post a client.

C

You don't I, don't think clients enforce check that the new value is greater than just check. It's not the same as the last one. They know about I.

B

C

I just need you to read it though yeah so I think Braun argued that we should keep it, as is basically.

C

Yeah I agree about this must not you know, literally close. The connection without by is yeah.

B

C

Not very well described and yeah, it should be assured no yeah, and we probably should talk about more. You know what are the cases exactly.

A

Yeah or should send the buy it.

B

Because there are times when you just you can't you.

C

Just have to kill the connection now, it's must not without sending by so what.

B

I'm saying sometimes you sometimes you can't do that something right.

C

He does I have to.

B

Kill the connects yeah. So that's, why should not is the.

D

A

B

D

Chris Newman I agree with, should not the two cases I've encountered where it's very important to just hang up is like? Is you know in the middle of? If, if the client starts delaying in the middle of less negotiation, you can't send it by you. Cannot the only option is closed detection and then you'll also, you know the other example was special fetching large message actually.

C

And you send a small message about in the middle of TLS: I think that that's quite might be very useful to do it to the document.

B

Yeah I mean the fact is, if you're killing it in the middle of a large fetch. Technically, you can send the by, but it doesn't do any good clients, not gonna, get it before. You kill the connections. I.

A

Do any other thing I would argue put here is: let's get rid of a few of the negatives from that, and just say, should send a bye if you know that early closing the connection yeah than should not, if not not not,.

A

A

C

And now clarifying, courted nil, vanillas at a time and special form. That's fine.

C

C

Well, basically, it was just a statement of fact that some server Street mailbox names has kids a sensitive because they might be using file system on the OS, which is itself case, insensitive and ASCII range. So this is client. Clients cannot rely on this one way or the other, and you know its server implementations have a choice which way they want to do it small statement of fact than you know right.

B

So the typical way of doing that is the server preserves the case, but doesn't care yes.

C

But this effects create in a sense of if you try to create mailbox in different case it will fail the letter one. That's.

B

Correct and and.

C

That's the only thing it is trying to say.

B

Right yeah I would rather go that way. Then.

B

Yeah, what I mean I'ma do is recommend that either you that you always preserve the case of the created mailbox and that the server may choose to ignore the case in future. References to that mailbox.

B

You know I'm, saying yeah.

C

Well, give me suggest some text, because this.

B

House I'll suggest this.

C

Was not intended to be any level of requirements is just showing two alternatives that are common in the wild right.

B

But but I think we do need to be prescriptive about it, so that we make it clear that if you do the wrong thing, you're gonna screw up interoperability in funny ways. I'll come up with text chart when I do when I do the list extended editions all of.

A

This okay shifting from Java here, one from Ned, saying problem with calling for specific terrorist behavior. Is it as typically up to the telleth implementation you're using I'm, not the.

B

Only buddy Christine say anything about specific to you. All I was saying was highlight that as a reason that you might have to close the connection without there's.

A

No other choice to do things. Timur says I may be more worried about the part of the utf-8 normalization than case sensitivity. Yeah.

C

I was case sensitive in ASCII range. That's what deliberate yeah.

B

But I will look at this whole thing about mailbox naming and make a pass through it and then we'll see where we copy.

C

C

Yes, this section needs your work. ah What happened is Chris contributed text to move modified, you gsm, seven, to appendix sort of as an optional backward compatibility thing, and this wasn't cleanly edited. So it might give you a wrong impression that.

C

This is supported, so I think we just need to review and a the move, the section or edited slightly yeah.

B

I think this is gonna, be a landmine for servers that are trying to support both rf1 and rf2, dealing with both UTF modified UT, f7 and utf-8 for the same mailbox names I, don't know, Chris is gonna. Tell us how to do this. Yeah.

D

I've implemented this it. There are some subtle cases, but it's but I was able to. But it's not too bad yeah subtle case is. You must track on the server side, whether or not client has enabled either I met for rep to, or you know or the utf-8 extent. Smtp utf-8 is right.

F

D

It's enabled it then the the behavior of a bear ampersand is different, because if the client hasn't enabled the bear ampersands an error because it's it has to assume the MMU tf7, if it's enabled, then a very ampersand yo is allowed, and if your internal storage is mu, t f7, you turn the bear amperes and into a frozen so yeah. This is. This is implementable.

C

Is then advise that you think we can add to this.

C

C

Haven't thought the full the whole thing throat to be honest, I was thinking about you know so you're saying you're rejecting would if we're ampersand, if in well.

D

Here's the issue once you allow utf-8 in the protocol, you may as well allow utf-8 in the protocol period whether or not you've enabled it yeah.

C

D

You know it's actually harder if you disallow utf-8, unless you've enabled it so I. Allow you to f8 all the time yeah. But then you have then. The way you deal with utf-8 is different, depending a little bit different for mailbox names, depending on whether or not you've enabled because in.

F

The encoding case.

B

I'll see if I can come I understand what Chris is getting at so again, I'll see if I can write text we'll see.

D

How that comes out.

B

Alright to Kate I completely agree with this. We should not. We should take out digest, md5 and sure s cram soffit 256 is a fine substitution. I think.

C

I didn't move that just md5 to historic anyway. So that's that's right!.

B

About that I'm just wondering if we should have a mandatory to implement one I call scram mandatory to.

C

B

Or something like that, because otherwise a client could implement a perfectly good sasso mechanism and the server could implement apparently good others hassle mechanism and then they're stuck with plaintext passwords, because they don't have anything good. In common.

D

Chris Newman that the fact the matter is they're stuck with plaintext passwords anyway, so the as much as I'd like to promote scram, sha-256 I, think promoting it with as a mandatory to implement, isn't the right way to do it. Yeah it's yo, I think it needs to be encouraged by the community, but I, don't think you know I. Think people will just ignore that and and we're trying to impose our our wishes on. You know on implementers that can't follow that.

D

Basically, the issue is, if you tie into an identity system and any medium to large deployment, has to tie into an identity system yo, you can only implement what the identity system supports. So yo. If we put in a mandatory to implement we're, saying we're saying something that we know, people will have control and can't deploy, but.

C

I mean I think this text is long, the lineup should it doesn't quite.

B

As a I was suggesting mentor and implement, but so Chris the you say, we're stuck with plain text passwords anyway,.

F

B

You mean we're stuck with implementing them or were stuck with you using them.

D

B

D

The most part both I mean you, can you can create enclaves where you can do other things, but right.

G

There enclaves.

D

I'll just one more note: I I came up with an idea for new sasso mechanism that I'm going to work on that for bootstrapping of I specific tokens, so I'm gonna try working pick, I'm gonna be working.

D

B

Just for the record, I suspect that we might get some pushback from the sec ADEs about the about not having a mandatory damn. It's awesome at what we'll burn that we'll burn that bridge. If we get to it well sure this bridges, throat anyways, but that's still not mandatory to implement and they often want mentor template we'll see where we go. That's.

G

C

Right, our favorite topic I forgot about the reference argument on the list. The.

B

Reference for those who don't know the history, the reference argument is there as like a CD change directory command within the the system, and it was put there because of a particular back-end that used UNIX file system. Is that obsolete now? Should we not use.

C

That, well, we don't want to change syntax right, you know at the moment it has two parameters basically, and there are complicated rules for combining them. So.

B

C

Keeping the parameter syntax.

B

Wise you just stick in a null string. We can certainly deal with the syntax, wise I guess. My thought is that if well, there's two things: if you need it: syntax, wise and you're arbitrarily sticking in a null string. Why not stick in an actual string, but there is some complexity to implementing a non null string there, as I found out when I first implemented it.

F

B

Think we should just leave it the way it is yeah.

C

I'm tempted to leave it the way it is I'm, happy to say, clients should use just empty reference than again well.

B

C

B

I'm happy to say that why does it matter if everybody has to implement it and make it work anyway, then it doesn't really matter well.

C

It just helps unless you're writing special client, it just easier yeah. The.

B

Right by that is.

C

Complexity on the server domain- yes, I.

B

Would not want to put it as a normative statement, but I would I would put it as an explanation that this is why it's there, if you don't, have that particular use case you're better off just using a null string and not going timo. Do you have anything further to say about this.

A

Canvases I think some clients have used the reference string and then it works differently in different servers which confuses I.

B

Believe that's true yeah so, but but I think as alexei said. I think in some sense we really can't do anything about that at this point. Unless we want to be incompatible.

A

B

Tell clients it's a bad idea to do it unless you have a really good reason to, but not at a should level. Just that's.

E

B

I else Elsa, but.

C

Elsa Elsa is gone yeah. Well, there is no this. The comment was about no selects when you do select and string right, no list empty string, and then you get a hierarchy separator. If you're writing it from scratch. I would have fixed it, but I'm tempted just to leave it, as is because clients still need to handle it. The way it is so. Yes, it's a glitch, but I. Just yes wouldn't be well.

B

That's again an artifact of a particular back end that was unable to put mailboxes and messages in the same mailbox.

B

Is really where it came from, but.

G

B

You create, if your mail, if your back-end structure, says that if X contains other subfolders, you can't also contain messages. Oh no, no.

C

No select, that's where no, no, no I think this is talking about list quote, quote quote quote you get this hierarchy separator yeah to find out. You know how to parse Lister mailboxes or do it for the namespace I think so.

B

I'm, just I'm not talking about this text. I'm talking about teemo's comment. Demos comments are a Oh select Cluj, that's I.

C

Might have misinterpreted his comment, then yeah yeah and we're getting no select in general as much as it's. You know it's pretty similar to non-existent in some Wow and in other cases it's not but yeah I. Don't think we can do anything about this kimber.

A

Says to come out was about L sub, empty string percent. That's.

C

What I thought he was talking about, but well, so how does that? It returns special list response? Yes, that just returned you hierarchy, delimit era, but it doesn't correspond to.

A

Existing mailbox name Kim earth in thing, but you were saying list extended, is coming anyway, so then else that would be removed. It's.

C

B

That's why I said it's. That's.

A

Really an issue.

B

I was thinking of notes. I was looking at the new Select in general, so anyway, let's move Oh actually.

C

Sorry I might have intended for it to be gone, it might still remain in the document. Then it needs to be taken out.

A

B

When I put in list extended them, I'm gonna go through it. Yes, that.

C

B

C

C

Yeah, this is very interesting requirement.

C

In general, using different hierarchy, delimiters in different namespaces is I, wouldn't say it's a well. It's almost like a bad idea, but I understand why people do this. It.

B

Doesn't even have to do with different hierarchy: delimiter, yes, just saying create, creating a new mailbox in a personal namespace when there are multiple personal namespaces, yeah I agree using the first one.

C

Well and the delance clan can do this. No.

B

I haven't seen only different, didn't say that we I don't think we should be using normative language to talk about what a user interface should do.

B

So maybe what we should say is just what what what the alternatives might be and let the client decide whether to put up a pop-up asking for which namespace or just use the first one. But here we're saying yes idea.

E

C

Don't use allow the user in the answer agree.

B

And check definitely, we should drop yes, yeah.

C

Yeah thing is I, haven't actually seen, I might know once over the does multiple namespaces, but all of the same type. So it's not a very common option but yeah. It's quite a lot of complexity. Shouldn't have a shake right and check command.

C

B

C

I think it has slightly different semantics in practice. All the service I know through check and Noah as the same I'm. Fine either saying they are the same or a moving check. I, don't mind I.

B

Suggest that we remove check because I've never encountered a client that sent it anyway, I think.

C

The check is more expensive one. The right supposedly like.

B

Well, ready yeah, if it isn't just to know up, then it's more expensive, yeah. Okay,.

C

Yeah I think it's also more complicated to write a client to figure out which one you need to use where when and so yeah.

B

Yeah, any objections to dropping check speak up. You'll have another chance when you review the document. Okay,.

D

C

Well, I think we need specific text proposal yeah so.

B

I: okay, here's the thing about search, I, know at least four different ways that search has been implemented in different servers and I know of no interoperability, issues that that causes. That search is consistent in how its implemented for the things that matter like flag searches and such oh.

A

B

Header field searches, but for Strings in the body of the message there's no, inter about interoperability issue, it's a user experience issue and I like I, said an.

D

Old still efficient version, yeah and.

B

Users of those systems just get used to what they get and I, don't think. We need to worry about.

A

So, what's the proposal here that we remove the exact specification of how body search works and have it behave like search fuzzy? Yes, but.

B

Not it's not a habit, behave like search fuzzy. Some servers will make it behave like a search, fuzzy. Some will do substring matches. Some will do other things. I would.

C

Like to keep you know, if you don't know what you're doing do substring or something I, don't know not exactly this written this way, but if.

B

You haven't thought about optimizing.

C

It start with this a.

B

Graduate student who's writing an IMAP server will need better wording than what we what I'm suggesting yeah, but any anybody who's really knows what they're doing with mail store will do whatever they do with mail stores. So maybe just a few words in here that say.

B

Substring matching is a good default. If you don't have anything else, you know you don't have a reason to do anything else.

B

Maybe I'll take an action to look at the text here and figure something out when I do my update yeah.

A

Yeah, the one thing I'd like to see is less prescriptive about exactly way you have to do correct and because that's not being followed right.

B

Nope nobody what nobody worried about that.

B

The macros are handy and clients use them. Yeah.

C

I'm tempted to leave them, as is, though, first.

B

C

Kind of the funny.

B

One, but yes, don't.

D

B

That I've ever seen that but I've seen all in full.

A

B

As well, keep them all I.

C

Mean it does not add complexity. So it's a simple bitmask and you know couple of you write it simple. So.

B

While I in principle, I agree with Teemo that it's probably time to do that, it doesn't matter enough to break interoperability for it. Yeah yeah.

A

And then things like the binary and body requiring cupcake only turn off the auto setting. It see.

B

The body I I, don't have enough experience with binary to say, but body is certainly it's kind of odd that it sets the scene that, for some partial body retrievals and the main thing is, if you, if you have a 2 gigabyte body part and you retrieve the first 3 bytes of it, does it make sense to set the scene bit and if the answer is no.

C

But this is a user interface. It makes.

B

Sense, it's user interface issue not entirely because the scene that affects other clients that are accessing the same mail store same some part of the.

C

Person a hate out the setting scene, but I'm just trying to find out if there are clients that relies on it. So I don't want just to take this.

A

C

Then I suspect, maybe we should again encourage use dot beak but don't take it out. I'll.

B

Buy that so a bit of discussion that says setting the auto scene flag automatically setting the scene flag on partial searches has come out as shown to be problematic.

B

Clients should consider whether they should have turn off the automatic setting using dark beef.

C

But it was something like that: yes, if you can just do I problem I take other than this is annoying. You know, just not quite the same, but problematic is unexpected.

B

C

That's true if.

B

You're using, for instance, if you're using a partial, a partial fetch in order to populate a one-line summary of the message in the message list and that causes the message to be seen. That's unexpected behavior, yeah,.

C

I, if you are writing the client like this I assume users will stop screaming at you. So.

B

That's some kind of text like that that says dot peak is a good idea, for these situations sounds good.

C

Next one, yes, this is a mission, will be fixed, yep.

B

F

B

Unfortunately, I've seen a lot of use of the RFC 822 versions of these I would just.

F

Like this might.

B

Be the time to start deprecated because RFC 822 is has been obsolete for.

C

15 years now there is actually comment in that at the beginning, even in 3501 saying despite saying RCA to tool, it doesn't actually mean that I understand.

B

But I just would rather not have that be a protocol element anymore. Well, it's a it's. A mime element still so yeah yeah well I'm. So, though, about properly deprecated, which means leave them in but say that they may be that the.

F

Moles in a nexus image may.

C

B

In the next version, so you should start using the I'm.

C

Okay, that it's sort of like, like must insecurity area you know, yeah, must implement now might be removed yeah anyway, we I think.

B

It's text in for that, but yeah I agree it's time to deprecated.

C

Yes, mentioning capability that might return on login and authenticate. Yes, yeah.

C

Right namespace response, extensions and I was thinking about this yesterday and I thought I agree with you with demo and then I think I remembered one case where it's actually used, which is a language extension localized namespaces per language.

C

B

Point is that you should tell people where to find what that means. It.

C

Is defined in a B and F just there is no example of what it is. Well, it's an extension, but I think probably need to add some text what the intent of it is. Yeah Chris, it's namespace, one of yours.

B

C

B

Mars or any namespace response extensions currently I think.

C

Language does have one.

B

Botanical yeah, then then we can put it in an example should do it.

B

A

All right there Beth separate up.

B

The second one is interesting: I.

C

Don't think the second one is correct.

C

Inbox cannot have a separate separate day. The only way it can have if it's its own personal namespace right.

B

I'm SID, but you can have an inbox in each namespace.

C

Only one of them is magic, and only one of them is magic correct. So so maybe it doesn't matter doesn't.

B

Matter: okay, I'll buy that.

C

Reflects permanent lights be sent after new keyboard, I think dance is know, yeah the reason. Well, so we have slash star special marker, saying I, support 88 right new keyboards and can store that if the server turned this marker, you store a new keyboard. There is no need to announce it because and the command succeeded it stored it. There are some implementations that have a limited number of the key words for them. At some point.

C

They need to announce that slash star is no longer there at this point you know so I think that's, there's no Timo.

B

Do you have any comment about the use case for what you're talking about here about the flags permanent flags? He.

C

Wrote the text according to the spec and some service break. It.

A

Yeah humor says: I, don't really care, it's just different servers work differently. Okay,.

B

C

All right, the last one I'm not entirely sure, is this just clarification such that something is.

B

Just pointing out some discussions that happened on the list sometime.

B

Yeah I, don't I'll check his wiki. We should look at the explanations for search text in search body and make sure that they're clear about what the differences are right.

B

The other thing is, I did do they work differently, as as far as drilling into body parts and decoding the body parts to do the search and all of that now.

C

I think what is more, one is more inclusive, I.

B

Think body is more inclusive for that, but text is more inclusive for header fields and stuff. So you know there are differences and we need to make sure that the text clarifies it. That's all.

C

Okay, so thank you to Timo for this. This is actually walking group law, school level, comments, I'm, very glad you send them early, so that makes it so much easier to deal with. Well, thank you for that.

C

This is oh, this is just bus changes, yeah.

A

All right, thanks, Lexi.

A

We are pretty much exactly on time and are we at working group left coast edge? Absolutely not? No. We.

B

But we should be Oh, certainly be by Montreal and I hope sooner.

A

B

Maybe two revisions, Alexei, says and I agree with that: I have to make a pass and then Alexei will have to make a pass again and if he doesn't come up with anything significant after that, then we should be ready for us all. So yeah.

A

That is all we had on the agenda today, other than kicking our master right.

A

Like everybody trying to do stuff with this or at that time,.

A

Gotta try to get busy three times in years.

A

Not community says it's it reliable, Network.

C

I was trying to remember this is one of my this year right, so I put the fetch preview on the first teller jet in April, so um I nearly put it before but transition, but it we had quite a few very busy torture. So.

B

Okay, so the only thing we we ought to be able to adopt these documents. The two milestones that say adopt: we ought to be able to get that done right.

A

B

A

King, the Delfino two documents should be fine, all right.

B

A

Just a matter of finding out the culprit: oh don't.

B

We meet that we we mark those milestones done when that happens, and we create new milestones for dates for those yep, so the the imap4 Rev to change the milestone to say May, and we should be okay with that or maybe June say June we can always. We can always beat it cool.

A

And updating the kado is to look at during next month's. We have we have to charter actions from yesterday, one of which was to update a charter for this. The other ones propose a new charter before updating up 3, 21 22.

B

Yeah so you're that that's the question of the dates for that is to you: can. Can you meet those days and.

A

B

Charter, the.

A

Charter fairly broke your problem, yeah.

C

A

Yep cool- we don't have anything else on our list. I, don't believe that milestone sent me not from today's session.

A

Planning for Montreal will definitely I expect one to meet there again, we'll have a new charter and a new work to look at wood. We don't need to go over the Charter again stay unless anyone has anything, no one pose for that dinner at tonight. I didn't call for show of hands. Anyone who didn't already say they were coming at the Jay map session yesterday or a mummy got up and walked perfect three more I will I will change from twelve to fourteen excellent.

A

That is all I've got. Anyone have anything else they want to discuss for this h-here regarding extras.

B

F

B

A

All right thanks everybody! Oh yes, yes, good point! I will make here at 6:50. It is only about five minutes. Walk here is downstairs in the lobby between the lifts open, the name out the least. Let's.

B

Look for a crowd of socially awkward people standing together, no.

A

That doesn't work, it shapes up to broad. Yes, no, it still doesn't help.

A

All right, thank you and if you haven't signed the blue, yet I think they're floating around the road somewhere. Please grab me sign them.

E

F

All right, hello, you know yep.

E

I've pulled up, yes, we'll have a page here so means you.