[04:52:53] John L joins the room
[05:26:18] cm-msk@jabber.org joins the room
[07:14:36] Kurt A joins the room
[07:16:06] John L leaves the room
[07:19:58] Meetecho joins the room
[07:20:03] Kurt Andersen joins the room
[07:20:03] Pete Resnick joins the room
[07:20:03] Lorenzo Miniero joins the room
[07:20:03] Paolo Saviano joins the room
[07:20:03] Dave Crocker joins the room
[07:20:03] Alexey Melnikov joins the room
[07:21:21] Alexey Melnikov leaves the room
[07:21:28] Alexey Melnikov joins the room
[07:21:34] Seth Blank joins the room
[07:21:35] Alexey Melnikov leaves the room
[07:21:48] Alexey Melnikov joins the room
[07:22:16] Alexey Melnikov leaves the room
[07:22:21] Kurt Andersen leaves the room
[07:22:41] Alexey Melnikov joins the room
[07:22:53] Kurt Andersen joins the room
[07:23:05] <Lorenzo Miniero> I can hear you
[07:23:22] <Seth Blank> I heard you
[07:23:24] <Alexey Melnikov> Thank you!
[07:23:39] Kenneth Murchison joins the room
[07:24:21] Barry Leiba_959 joins the room
[07:24:40] Harald Alvestrand joins the room
[07:25:57] <Seth Blank> Alexey- your screen isn't visible yet. It just says a screen share is being started.
[07:26:02] <Seth Blank> Ah, there it goes. I can see it now.
[07:26:28] Bron Gondwana joins the room
[07:26:41] <Kurt A> Is there any audio yet? I'm not hearing anything
[07:26:46] <Lorenzo Miniero> Yes, that placeholder appears while the screen/application picker is there, so until something is chosen
[07:27:04] <Lorenzo Miniero> Kurt: no one is talking at the moment
[07:27:10] <Kurt A> ack
[07:27:27] Ned Freed joins the room
[07:27:47] Kazunori Fujiwara joins the room
[07:27:52] Peter Koch joins the room
[07:28:51] Alwin de Bruin joins the room
[07:29:34] bhoeneis joins the room
[07:30:20] Trent Adams joins the room
[07:30:23] Peter Yee joins the room
[07:30:32] Robert Stepanek joins the room
[07:30:37] <Kurt A> yes, seeing the cover slide
[07:30:41] <Alwin de Bruin> yes
[07:30:51] Bernie Hoeneisen joins the room
[07:31:23] Jim Fenton joins the room
[07:31:50] Neil Jenkins joins the room
[07:32:10] JcK joins the room
[07:32:25] Marco Davids joins the room
[07:32:28] Loren McIntyre joins the room
[07:32:42] Marco Davids leaves the room
[07:32:47] Marco Davids joins the room
[07:33:07] Jim Fenton leaves the room
[07:33:13] cm-msk@jabber.org joins the room
[07:33:18] John Klensin joins the room
[07:33:21] cm-msk@jabber.org leaves the room
[07:33:44] Jim Fenton joins the room
[07:33:50] Murray Kucherawy joins the room
[07:34:04] Loren McIntyre leaves the room
[07:34:07] Loren McIntyre joins the room
[07:34:37] Vittorio Bertola joins the room
[07:34:46] <Dave Crocker> The slide font is quite small.
[07:35:06] John Klensin leaves the room
[07:35:24] <Kurt A> Seeing & hearing you Pete
[07:35:37] John Klensin joins the room
[07:35:42] <Murray Kucherawy> Audio good here too.
[07:36:26] Hernâni Marques joins the room
[07:36:57] cm-msk@jabber.org leaves the room
[07:38:13] Hernâni Marques leaves the room
[07:38:23] Hernâni Marques joins the room
[07:39:27] Jacob Rideout joins the room
[07:40:39] <Jim Fenton> restricted changes to terminology is part of the WG charter.
[07:41:07] <Harald Alvestrand> This WG's attendee list makes me nostalgic......
[07:41:24] <Alexey Melnikov> Do people have an opinion on whether trace header fields are used in some specific way in other RFCs? E.g. "all trace header fields need to be removed" in certain context?
[07:42:03] <Ned Freed> My mic does not seem to work.
[07:42:05] <Ned Freed> As usual.
[07:42:22] <Ned Freed> The question here is whether we want to call everyone that's added as trace field.
[07:42:23] <Alexey Melnikov> Ned, if you type here, I will try to relay
[07:42:47] <Ned Freed> I'm pretty sure some of the crap that's added strethes the meaning.
[07:45:16] <Jim Fenton> @Alexey There is some discussion elsewhere on message repudiation, which might motivate removing some header fields.
[07:46:00] <Seth Blank> @Jim can you add a reference in codimd or start a thread on the list?
[07:46:03] 木村 大和 joins the room
[07:46:51] 木村 大和 leaves the room
[07:47:12] 木村 大和 joins the room
[07:47:13] 木村 大和 leaves the room
[07:47:13] <Jim Fenton> Will add something in codimd. Much of the discussion has had to do with DKIM signatures (not a trace header field, right?) but I can see it spilling into Received: header fields and others.
[07:47:16] 木村 大和 joins the room
[07:47:28] <Dave Crocker> Quoting or not quoting is not supposed to have 'semantic' meaning.  It's only intended as an encoding mechanism.  Absence clear, concrete, and compelling reports of significant problems, don't change the spec.
[07:47:30] Masaki Kase joins the room
[07:47:30] Masaki Kase leaves the room
[07:47:33] Masaki Kase joins the room
[07:48:10] <Dave Crocker> go read it
[07:48:17] <Kurt A> DKIM signatures do gain some tracing when used along with ARC
[07:48:33] Fraser Tweedale joins the room
[07:50:14] 木村 大和 leaves the room
[07:50:24] jtrentadams joins the room
[07:50:36] <Dave Crocker> The issue is not that it's suboptimal or unappealing, but that it is not known to cause serious problem.  I thought the goal for this round of effort with the specs was to make the minimum changes required.  This one isn't required.
[07:51:15] <Murray Kucherawy> DKIM says signatures SHOULD be treated as trace fields.
[07:51:32] Fraser joins the room
[07:52:10] <Ned Freed> This is a layering violation. Since we don't specify the prociess for canonicalizing addresses, we don't have a way of telling what's legal for a given ADMD and what is not.
[07:52:23] Fraser leaves the room
[07:52:27] Jiankang Yao joins the room
[07:52:42] <Ned Freed> It is perfectly permissible for an ADMD to allow an empty local part, represented syntactically as a pair of quotes.
[07:53:08] <Ned Freed> Or not.
[07:53:19] <Dave Crocker> +1
[07:54:27] <Kurt A> testing would be a bad idea - we have no way of running a comprehensive test across every system and client
[07:54:34] <Ned Freed> The issue of quotes being borked by the use of exciting cominbations of Unicode stuff is a separate one. And I have to say it's becoming a tremendous disincentive to using quoted addresses, display strings, and so on at all.
[07:54:35] <Dave Crocker> +1
[07:54:38] <Ned Freed> But it's a separate issue.
[07:55:10] <Dave Crocker> (My +1 was meant for Kurt's comment.)
[07:55:35] <Dave Crocker> Meta:  Is this topic really worth this much time of this meeting?
[07:55:47] <Ned Freed> Hey Dave, that problem would have been solved by having a reaction mechanism...
[07:56:00] <Pete Resnick> heh
[07:56:04] <Dave Crocker> :)
[07:56:36] <Alexey Melnikov> We have to spend time on this topic at some point anyway.
[07:57:09] <Dave Crocker> Alexey, the 'cost' of meeting time is quite a bit higher than the cost of mailing list time.
[07:57:10] Vittorio Bertola leaves the room
[07:57:16] Vittorio Bertola joins the room
[07:57:21] Hernâni Marques leaves the room
[07:57:21] Marco Davids leaves the room
[07:57:26] Hernâni Marques joins the room
[07:57:27] Barry Leiba_959 leaves the room
[07:57:35] <Harald Alvestrand> are we still doing obsolete-syntax?
[07:57:39] Fraser Tweedale leaves the room
[07:57:42] Marco Davids joins the room
[07:57:44] Fraser Tweedale joins the room
[07:57:56] Barry Leiba_258 joins the room
[07:58:10] <Alexey Melnikov> Harald: yes
[07:58:30] <Dave Crocker> As for obsolete-syntax, is there any clear indication that it has proved useful over the last 25 years?  Have any of the entries there actual become obsolete, as in, not used?
[07:59:28] <Alexey Melnikov> Dave: good question. Personally, I would just drop it.
[07:59:46] <Dave Crocker> +1
[08:00:02] <Ned Freed> I think it has encouraged a move away from the more problematic forms.
[08:00:38] Vittorio Bertola leaves the room
[08:00:45] <Dave Crocker> (Obs was an entirely reasonable idea, as were X- header fields, before that.  Neither proved useful, and X- actually was problematic.)
[08:01:02] <Ned Freed> Header field name length limits essentially fall out of the 78 character recommended line length limit.
[08:01:22] <Kurt A> Making it explicit will solve for lots of semi-circular arguements amongst people who have difficulty counting
[08:01:31] <Kurt A> (header field name lengths)
[08:01:42] <Ned Freed> You can't break a header field name, so the effect is you SHOULD limit things to 78 characters including the colon.
[08:02:12] <Ned Freed> Whether you want to point that out is another matter.
[08:02:19] <Dave Crocker> The latter part of Ned's text sounds useful for the spec.
[08:02:26] <Ned Freed> I think an ABNF comment would be fine.
[08:03:11] <Pete Resnick> The obs also contains things that need to be interpreted, but that you shouldn't generate.
[08:03:31] Julian Reschke joins the room
[08:03:56] <Pete Resnick> Julian, you are echoing
[08:04:26] <Kurt A> Re registry of header fields - how would this work for proprietary header fields which are not registered anywhere public?
[08:04:30] <Pete Resnick> Could one of the chairs mute Julian?
[08:04:58] <Bron Gondwana> Ta
[08:06:13] <Pete Resnick> Kurt: What should implementations do with unknown header fields anyway?
[08:06:26] <Kurt A> ignore them
[08:07:13] <Kurt A> Just like all the headers formerly known as "X-"
[08:08:08] <Ned Freed> I'm actually in favor of including more information in the registry.
[08:08:28] <Pete Resnick> Wasn't one of the proposals to simply add a column to the current header registry?
[08:08:59] <Alexey Melnikov> Pete: yes, but this means messing with the registry also used by HTTP and NNTP!
[08:09:01] <Ned Freed> Providing people with suggestions as to when to use something -and when not to - I'm having a hard time seeing this as a bad thing.
[08:10:19] <Ned Freed> As for the conflation with other standards, surely we have the flexibility to handle that!
[08:11:28] <Ned Freed> IPv6 addresses: I posted a comment about it.
[08:11:58] <Ned Freed> People need to read the justification for the limits in RFC 5952.
[08:12:28] <Ned Freed> The justification is nonsense in the conext of addresses.
[08:12:30] <Bron Gondwana> nobody uses http or nntp, so we're safe
[08:13:06] <Pete Resnick> Does 5322 have a sufficiently absurdly liberal syntax that I don't have to care about it?
[08:13:09] <Ned Freed> Please take this to the list.
[08:13:26] <Seth Blank> +1
[08:13:32] <Dave Crocker> +1
[08:14:11] <Ned Freed> And please read the justification for RFC 5952's limits. And think about whether or not they make sense in an address.
[08:15:47] <Pete Resnick> Oh, good, no changes for 5322: Visible US-ASCII characters between two square brackets. I hereby go back to ignoring this. :-)
[08:16:31] <Ned Freed> It's not about syntax per se. It's about canonical forms.
[08:17:11] <Fraser Tweedale> Why is a canonical form needed?  If it unambiguously represents a single IPv6 address, it is good enough.
[08:17:31] Julian Reschke leaves the room
[08:17:39] <Ned Freed> There has been some discussion of this on the list.
[08:17:49] <Ned Freed> "this" -&gt; resolvable names.
[08:17:59] Loren McIntyre leaves the room
[08:18:06] <Ned Freed> I agree with Alexey.
[08:18:23] <Ned Freed> There are too many private use cases.
[08:18:42] <Seth Blank> +1
[08:19:17] <Dave Crocker> Oddly, having it say 'resolvable' without giving details to surround it, seems quite good in this case.  It needs to work where it's being used, but we are specifying the details of what makes it work.
[08:19:28] <Fraser Tweedale> It could be good to add informational reference to https://datatracker.ietf.org/doc/draft-ietf-dnsop-private-use-tld/
[08:19:52] <Dave Crocker> As noted, DNS resolution can be complicated.  That's ok, with the current language.  Saying more puts this at risk.
[08:20:09] <Seth Blank> Perhaps clarification is needed here in the A/S without a change in 5321?
[08:21:02] <Fraser Tweedale> "Resolvable" - by whom?  A client could introduce a name that is not resolvable by the server (or vice-versa)
[08:23:24] <Ned Freed> I would like to drop resolvable.
[08:23:33] <Fraser Tweedale> +1
[08:24:03] <Ned Freed> Exactly.
[08:24:12] <Dave Crocker> Dropping 'resolvable' means it's ok to be an entirely arbitrary string in domain name format.  Saying resolvable constrains that.
[08:24:31] <Bron Gondwana> +1 to dropping resolvable
[08:24:38] <Fraser Tweedale> @Dave but there is no guarantee that what is resolvable to the client, is resolvable to the server.
[08:24:40] <Murray Kucherawy> agree
[08:25:03] <Bron Gondwana> @Dave yes, that seems true - with the side effect of "the other end might decide you're a jerk and treat you accordingly"
[08:25:20] <Pete Resnick> Did someone just say that "resolvable" appears somewhere else in the document?
[08:25:27] <Bron Gondwana> there's never a guarantee that the other end won't consider you to be a jerk
[08:25:29] <Alexey Melnikov> Pete: yes
[08:26:07] dkg joins the room
[08:28:59] <Bron Gondwana> I was going to pop into voice to say "even if it's resolvable right now, it may not be resolvable when you go to reply"
[08:29:07] <Bron Gondwana> resolvability is a whack-a-mole game
[08:29:49] Daniel Gillmor joins the room
[08:30:44] Alwin de Bruin leaves the room
[08:30:57] Julian Reschke joins the room
[08:31:01] <Pete Resnick> Dave's in the queue.
[08:31:12] <Kurt A> We've seen how well "clarifying things out of existence" works - I'm against it
[08:32:20] <Ned Freed> Exept that it isn't. This just isn't how things work.
[08:33:05] <Ned Freed> I posted about this on the list.
[08:34:13] Ned Freed leaves the room
[08:34:58] <Jim Fenton> we're going to get cut off.
[08:35:06] Bron Gondwana leaves the room
[08:35:13] <Alexey Melnikov> LASER routing is another case
[08:35:38] Peter Koch leaves the room
[08:35:44] <Pete Resnick> I am mostly sympathetic with Dave's position, but when an implementation simply changes it's RCPT TO commands and still uses its SMTP queue, it sure seems like something different than delivery-and-resend.
[08:35:57] Trent Adams leaves the room
[08:36:04] <Fraser Tweedale> Time is ideal for some of us :)
[08:36:08] Bernie Hoeneisen leaves the room
[08:36:08] <Fraser Tweedale> Thanks!
[08:36:09] Barry Leiba_258 leaves the room
[08:36:10] Robert Stepanek leaves the room
[08:36:10] Marco Davids leaves the room
[08:36:11] Murray Kucherawy leaves the room
[08:36:12] Kenneth Murchison leaves the room
[08:36:12] Peter Yee leaves the room
[08:36:13] Kurt Andersen leaves the room
[08:36:13] <Harald Alvestrand> See you all!
[08:36:15] Neil Jenkins leaves the room
[08:36:17] Masaki Kase leaves the room
[08:36:19] Harald Alvestrand leaves the room
[08:36:20] <Alexey Melnikov> Thank you all
[08:36:22] Kazunori Fujiwara leaves the room
[08:36:22] Jim Fenton leaves the room
[08:36:26] Fraser Tweedale leaves the room
[08:36:27] Jacob Rideout leaves the room
[08:36:29] Julian Reschke leaves the room
[08:36:30] jtrentadams leaves the room
[08:36:31] Pete Resnick leaves the room
[08:36:31] <Dave Crocker> @pete, that confuses networking architecture with software implementation.
[08:36:33] Barry Leiba leaves the room
[08:36:34] Dave Crocker leaves the room
[08:36:40] Lorenzo Miniero leaves the room
[08:36:42] Daniel Gillmor leaves the room
[08:36:42] John Klensin leaves the room
[08:36:42] Hernâni Marques leaves the room
[08:36:42] Paolo Saviano leaves the room
[08:36:42] Alexey Melnikov leaves the room
[08:36:42] Seth Blank leaves the room
[08:36:42] Jiankang Yao leaves the room
[08:37:20] Meetecho leaves the room
[08:38:14] Kurt A leaves the room
[08:38:35] Kurt A joins the room
[08:45:38] JcK leaves the room
[11:27:54] dkg leaves the room
[15:32:46] Matthew joins the room
[16:19:52] halfshot joins the room
[16:19:58] halfshot leaves the room
[16:31:53] undefined joins the room
[18:41:34] jtrentadams joins the room
[18:46:41] jtrentadams leaves the room