[10:40:34] barryleiba joins the room [10:43:06] /title DKIM working group session at IETF 80 (Prague) [10:43:49] /topic DKIM working group session at IETF 80 (Prague) [10:48:04] Barry Leiba joins the room [10:48:18] Barry Leiba has set the subject to: DKIM working group session at IETF 80 (Prague) [10:48:23] Barry Leiba leaves the room [10:51:00] sm joins the room [10:52:00] SM has arriven. [10:52:09] Barry :-) [10:55:12] tony.l.hansen joins the room [10:58:07] fenton joins the room [10:59:46] Hello, 10 people! [11:00:01] rlbob joins the room [11:00:17] R.E. Sonneveld joins the room [11:00:26] dcrocker joins the room [11:01:28] sftcd joins the room [11:01:47] Are the slides posted somewhere? [11:02:09] Slides are on the meeting materials page. URL forthcoming (copying/pasting now). [11:02:12] https://datatracker.ietf.org/meeting/80/materials.html [11:02:20] http://www.ietf.org/proceedings/80/slides/dkim-1.pdf [11:02:21] Jim Galvin joins the room [11:02:37] https://datatracker.ietf.org/meeting/80/materials.html [11:02:43] Thanks, Barry...I was on the wrong one of the many agenda pages [11:08:31] I'm looking at http://tools.ietf.org/rfcdiff?url2=draft-ietf-dkim-rfc4871bis-04 ... is that the right diff? [11:08:39] john.levine joins the room [11:08:40] yes [11:08:56] -04 is 20 minutes old. [11:09:33] and it's proper to be discussing a draft this new? [11:11:05] Jim, the diffs won't show between -03 and -04 [11:11:33] I have no objection to the discussion...just wanted to check on a process point. [11:12:09] on p46 line 6 [11:12:15] p49 sorry [11:12:47] 03/04 diff version is at: http://tools.ietf.org/rfcdiff?url2=draft-ietf-dkim-rfc4871bis-04 [11:12:56] J.D. joins the room [11:17:18] "strongly advised" sounds like SHOULD [11:17:20] This is not giving normative advice on the handling of the message...just on whether the signature should be considered valid or not, which is in scope. [11:19:31] I do wonder what "suspicious" means in this context, though. "suspicious" was an ADSP concept that was rejected. [11:19:57] jim: is that for the mic? [11:20:41] yes, at the proper time, thanks [11:21:20] Jim, we are going to pick another word to same the same thing. [11:21:26] to say [11:21:51] http://www.youtube.com/watch?v=SBmAPYkPeYU <— Elvis' Suspicious Mind, as requested [11:22:48] Mic: Murray had suggested an informative appendix cautioning users not to put in empty g= tags [11:23:17] Jacky Yao (Health Yao) joins the room [11:23:28] I'd like to see a bit more cautionary text about that because of DK legacy use. [11:24:10] ship text. [11:26:38] not from me [11:26:55] Jacky Yao (Health Yao) leaves the room [11:27:04] Jacky Yao (Health Yao) joins the room [11:27:08] Notes: rewrite para in 3rd para of 8.15 Murray to add para at the end about discussed text Murray to add column to iana registration text update by April 10 [11:27:28] "a bit more cautionary text" is your request, which means you know what you have in mind. it's better that the rest of us not guess. [11:28:10] right, that sounds more like an MUA BCP [11:28:31] @dcrocker: was that directed at me? I can write something if needed. [11:28:39] lost audio [11:28:48] audio back [11:29:31] @jim-yes, please. [11:29:58] a review of what MUAs do today would be a good start [11:30:06] lost audio [11:30:12] audio back [11:30:18] Audio working [11:30:23] haven't been losing audio here, FWIW [11:30:42] Audio issues seem to be coming and going for various people and in various rooms. [11:31:01] (I didn't lose audio when John did) [11:31:24] pure speculation -> experimental where I come from [11:35:33] nodding here also [11:35:53] good luck getting consensus on what we have consensus on, though [11:35:58] Notes on mailing list doc: mark as BCP, let ietf Last Call vet the decision [11:36:28] (nods) [11:36:48] (more nods) [11:37:52] To what address ;-) [11:37:55] Notes: keep notes on dealing with broken implementations [11:43:42] Agree with the "all in/all out" approach: either the MLM takes responsibility for the message or not. (doesn't need to be echoed to mic) [11:44:05] I worry this is a proxy for the MUA highlighting bad advice [11:45:59] Notes: MLM signing headers should go to the list [11:46:04] eliot.lear joins the room [11:47:18] Do ESPs consider themselves to be mailing lists? (I don't know since I'm not one of them) [11:47:34] If not, they're not going to consider this applicable anyway. [11:48:12] to put it another way: ESPs blur the line between author and originator [11:48:55] (and perform a lot of other services not described in email-arch) [11:49:37] To dcrocker's comment, this assumes something about verification of multiple signatures. [11:49:49] eliot.lear leaves the room [11:49:59] Notes: ESPs are no different from other entities that add 2nd sigs [11:51:20] identifying subdomains is problematic, so don't go there. [11:52:00] that's a real issue, but it's not the only one; list mail vs individual is probably useful with or without ADSP [11:52:22] streams [11:52:38] Notes: message streams advice should be left in [11:52:43] different signing for list mail vs individual may fit into the MUA BCP, should we ever write such a thing [11:53:04] JD, why MUA? [11:53:07] You can't in general tell a subdomain from a "top level" domain (one that was delegated by a registrar) [11:53:22] (not for mic) sm: because it's related to user actions [11:53:31] Ah [11:53:58] or maybe more accurate to call 'em author actions [11:59:03] I'm intrigued enough to work on a review of what MUAs do with authentication results today, to determine if there are best current practices ready to document — but that'll look beyond DKIM, so it may not fit in this WG anyway [11:59:08] We do have a lot more ADSP checking than people realize, since spamassassin checks it. [11:59:11] Patrik Halfar joins the room [11:59:41] But typical spamassassin config is synthetic ADSP [11:59:42] do we have consenus for ambivalency? [12:00:00] dormant is a possibility [12:00:06] @john.levine: spamassassin does both synthetic and real ADSP. [12:00:06] I think we're done, JD's stuff sounds like a different WG [12:03:45] (singing) we can't go on together [12:03:50] Does Barry really want to spend his birthday reading lots of drafts? [12:04:41] (claps), and thanks the Chairs for their leadership. [12:04:41] Thanks all! [12:05:11] Thanks to the Chairs [12:05:34] Thanks especially to the West-coasters for getting up so early. [12:05:51] any other exciting working groups to listen in on, now that I'm awake? *grin* [12:06:01] thanks for not having the meeting at 9:00 [12:06:06] AbFab is next, at 3:10, same channel. [12:06:12] sm leaves the room [12:06:41] 3:10 meaning 9:10 EDT, 6:10 PDT. [12:07:33] After that it's the technical plenary, where you can follow the "browser as an application protocol engine" food fight. [12:07:47] ooh! [12:08:00] chat you all soon... [12:08:22] J.D. leaves the room [12:08:25] john.levine leaves the room [12:08:46] barryleiba leaves the room [12:08:50] Have a pilsner for me...bye [12:09:18] fenton leaves the room [12:09:40] sftcd leaves the room [12:12:11] Jacky Yao (Health Yao) leaves the room [12:13:41] dcrocker leaves the room [12:14:29] Patrik Halfar leaves the room [12:18:39] barryleiba joins the room [12:18:50] barryleiba leaves the room [12:28:11] Jim Galvin leaves the room [13:00:37] rlbob leaves the room [13:05:26] Jacky Yao (Health Yao) joins the room [13:05:36] Jacky Yao (Health Yao) leaves the room [13:12:07] R.E. Sonneveld leaves the room [13:20:52] tony.l.hansen leaves the room [13:29:21] john.levine joins the room [13:29:28] john.levine leaves the room