httpbis - Wednesday, July 31, 2013

httpbis

httpbis@jabber.ietf.org

Wednesday, July 31, 2013< ^ >

stpeter has set the subject to: IETF HTTPbis Working Group | https://datatracker.ietf.org/meeting/86/agenda/httpbis/

Room Configuration

Room Occupants

GMT+0
[05:50:35] lear joins the room
[06:01:00] lear leaves the room
[06:53:18] lear joins the room
[06:55:29] yuioku.yj joins the room
[07:04:26] nygren@mit.edu/barnowl leaves the room
[09:27:18] lear leaves the room
[09:48:26] lear joins the room
[09:48:43] lear leaves the room
[11:03:46] g.e.montenegro joins the room
[11:13:51] g.e.montenegro leaves the room
[12:42:45] nygren@mit.edu/barnowl joins the room
[13:00:56] Florian Zeitz joins the room
[13:03:38] fielding joins the room
[13:04:16] g.e.montenegro joins the room
[13:08:03] andrey.uzunov joins the room
[13:09:17] mnot joins the room
[13:09:31] <mnot> Roy, are you there?
[13:09:36] <fielding> mnot, move the microphone further away from your mouth
[13:10:00] <fielding> still a bit to sensitive
[13:10:16] <fielding> darth mnot
[13:10:27] synp joins the room
[13:10:35] <fielding> damn, lost audio feed
[13:12:03] <fielding> okay, back on
[13:12:49] Ted joins the room
[13:14:50] tony.l.hansen joins the room
[13:14:50] <synp> I will jabber-scribe.
[13:14:53] SM joins the room
[13:14:59] redaka joins the room
[13:15:10] <synp> If you want anything to be read into the room, please preface it with "mic:"
[13:15:26] <fielding> none at all ... I have a guest sleeping next to my espresso machine :(
[13:15:26] <synp> Otherwise, it's just us talking...
[13:17:49] Julian joins the room
[13:19:16] <synp> Julian is presenting
[13:19:21] <synp> slide #3
[13:20:39] m&m joins the room
[13:21:00] kmurchison joins the room
[13:21:54] <fielding> slides are in http://trac.tools.ietf.org/wg/httpbis/trac/browser/wg_materials/ietf87
[13:22:35] <synp> slide #4
[13:23:15] <synp> sliede #5
[13:23:40] <Julian> http://trac.tools.ietf.org/wg/httpbis/trac/export/2331/wg_materials/ietf87/ietf-87-httpbis-changes.pdf
[13:23:40] <Julian> http://trac.tools.ietf.org/wg/httpbis/trac/export/2331/wg_materials/ietf87/ietf-87-httpbis-changes.pdf
[13:23:57] <synp> Mark is showing a list of the open issues
[13:24:06] Julian thanks
[13:24:29] <fielding> http://trac.tools.ietf.org/wg/httpbis/trac/report/3
[13:25:37] <fielding> DNT is "on a break"
[13:25:43] <synp> issue 470
[13:25:43] <fielding> no coffee
[13:26:04] <Julian> http://trac.tools.ietf.org/wg/httpbis/trac/ticket/470
[13:27:12] JcK joins the room
[13:27:58] <synp> Any thoughts?
[13:28:04] adb joins the room
[13:28:09] <fielding> mic: insufficient caffeine
[13:28:52] <Ted> @fielding.  He's faithfully going up to say that.
[13:29:00] MikeBishop joins the room
[13:29:20] <fielding> mic: no specific thoughts ... devil is in the details. I prefer to just have cacheability of a response, and then describe the inputs to that decision
[13:30:33] nygren@mit.edu/barnowl leaves the room
[13:30:45] nygren@mit.edu/barnowl joins the room
[13:31:12] <fielding> mnot, why is 470 a design issue instead of editorial?
[13:31:18] Dan Wing joins the room
[13:31:31] <fielding> bah, almost got that out before mnot said it
[13:31:35] <synp> mnot just said he's flipping it to editorial
[13:32:00] <synp> http://trac.tools.ietf.org/wg/httpbis/trac/ticket/475
[13:34:34] Dan Wing leaves the room
[13:36:05] nemo joins the room
[13:37:10] <synp> http://trac.tools.ietf.org/wg/httpbis/trac/ticket/458
[13:37:17] Dan Wing joins the room
[13:39:20] <synp> http://trac.tools.ietf.org/wg/httpbis/trac/ticket/345
[13:39:39] <fielding> mic: the 1.0 requirement in 458 comes from a race condition report in some paper
[13:40:54] <synp> http://trac.tools.ietf.org/wg/httpbis/trac/ticket/455
[13:42:28] <fielding> mic: I can't keep up ... 345 is intentional
[13:43:29] <synp> mnot says we'll have to discuss on the list to find out why
[13:43:35] <synp> http://trac.tools.ietf.org/wg/httpbis/trac/ticket/486
[13:44:49] <fielding> mic: still haven't fixed 455; for 486, my concern is that nobody implements this *very* expensive requirement
[13:47:52] <fielding> mic: I can live with mnot's solution, but it has to more targeted than "recipient" (i.e., it needs to say a recipient that makes use of the Warning header field)
[13:47:59] nemo leaves the room
[13:48:38] sal joins the room
[13:49:55] <fielding> mic: more time does not produce more comments -- just later comments.
[13:52:11] <JcK> Could speakers _please_ identify themselves.
[13:52:26] <synp> This has been Barry Leiba
[13:52:28] <Ted> That was Barry
[13:52:37] Julian before Barry
[13:52:44] sal leaves the room
[13:52:51] <Ted> before that was Martin
[13:53:07] <synp> Elliot Lear
[13:53:34] <Julian> Barry
[13:53:53] g.e.montenegro leaves the room
[13:53:57] g.e.montenegro joins the room
[13:54:05] <SM> mic:  This document set will need several reviewers and at least a month for review.
[13:54:56] <fielding> mic: DNT is "on a break" until Sep 4, so Roy intends to finish HTTP/1.1 drafts before then, but tickets like 484 make it very hard to estimate how much time will actually be needed.
[13:55:52] <fielding> http://trac.tools.ietf.org/wg/httpbis/trac/ticket/484
[13:56:41] <fielding> I suggest next week would be better
[13:56:56] <fielding> ... time to start reviewing
[13:57:44] <fielding> and assuming we generate a new draft for that purpose
[13:58:13] Julian nods
[13:58:56] <synp> On to 2.0
[14:02:26] adb leaves the room
[14:02:33] <synp> http://www.ietf.org/proceedings/87/slides/slides-87-httpbis-0.pdf
[14:02:50] <synp> slide #2
[14:03:53] <synp> "Methods" slide
[14:04:00] adb joins the room
[14:04:10] adb leaves the room
[14:04:32] <synp> "Header Cleaning" slide
[14:06:39] <synp> Roberto at the mike
[14:07:08] <synp> "Length Distribution for the traces" slide
[14:07:51] <synp> "Number of bytes per header" slide
[14:08:19] <synp> "Number of key-value pairs in the headers" slide
[14:09:17] <synp> "Compression (public benchmark)" slide
[14:09:30] nygren@mit.edu/barnowl leaves the room
[14:09:32] jessian joins the room
[14:09:39] nygren@mit.edu/barnowl joins the room
[14:09:43] <synp> "Compression (raw data)" slide
[14:09:59] <synp> "Publication" slide
[14:10:27] <synp> "Do you have any questions?" slide
[14:10:43] <synp> Mark talking
[14:11:43] <synp> Roberto Peon
[14:11:57] andrey.uzunov leaves the room
[14:12:01] <Julian> this was https://gnunet.org/httpbenchmark/
[14:12:54] <synp> EKR has entered the room. He's supposed to give us an update about TLS.
[14:13:15] PasiS joins the room
[14:13:32] andrey.uzunov joins the room
[14:13:42] <JcK> @synp: while your identifications are helpful, when there is a discussion going on, there is a lag of one or two speakers, maybe more, between Jabber and audio.  In addition, someone listening to the recordings later won't have the even loosely-synchronized Jabber process.   We really need speakers to identify themsevles at the Mic.  Could you r(or someone else) please remind the room?
[14:13:51] <JcK> thanks in advance
[14:15:08] Ted leaves the room
[14:15:16] <synp> I do. But people tend to say their names in EKR speeds. Similarly, if I had said my name, you would have a hard time matching it to the letter string "Yoav"
[14:17:34] Dan Wing leaves the room
[14:19:09] <synp> EKR up next
[14:19:48] Ted joins the room
[14:20:47] <JcK> Just lost audio
[14:21:21] <Ted> There was a short update on the status of the TLS working group:  going okay; more work on the issue on Friday.
[14:21:28] <jessian> had to reconnect again to the audio stream
[14:21:29] <synp> Are you sure? Mark was talking to EKR rather than the mic. Can you hear now?
[14:21:40] <Ted> (in re: ALPN)
[14:21:41] <synp> Gabriel presenting
[14:21:45] <JcK> I suspect that no one told the audio gods that you were not taking a break now. :-(
[14:22:23] <Ted> We are at "review: Unknown STartup State" slide now
[14:23:14] <Ted> Has it come back for you jck?
[14:23:16] <MikeBishop> Are the slides posted somewhere?
[14:23:22] <fielding> http://trac.tools.ietf.org/wg/httpbis/trac/browser/wg_materials/ietf87/Known%20Startup%20State%20for%20HTTPS.pdf
[14:23:31] <MikeBishop> Thanks.
[14:23:36] <Ted> Now on "HTTP-TLS negotiation case"
[14:24:12] Barry Leiba joins the room
[14:24:27] <synp> "Alternatives"
[14:24:32] Dan Wing joins the room
[14:24:40] danyork joins the room
[14:32:24] <MikeBishop> mic:  There's an assumption here that client always speaks first, but the framing layer has been trending toward being slightly more generic.  Do we need to consider two-way exchange at the framing layer, in case a different use involves the server speaking first?
[14:36:16] nemo joins the room
[14:36:46] m&m leaves the room: Disconnected: connection closed
[14:38:09] <JcK> Worked again after restarting (sorry for not saying so earlier), but it definitely did disconnect.
[14:39:00] Wendy Seltzer joins the room
[14:41:11] yuioku.yj leaves the room
[14:43:29] m&m joins the room
[14:43:35] m&m leaves the room: Disconnected: connection closed
[14:43:42] m&m joins the room
[14:44:43] <tony.l.hansen> so, was there a decision on startup state after all that conversation?
[14:44:52] <synp> HTTP/2.0 Issues
[14:44:55] m&m leaves the room
[14:45:01] <synp> Conversation will be next week
[14:46:16] <synp> Issue #57
[14:46:36] adb joins the room
[14:47:14] <synp> https://github.com/http2/http2-spec/issues/57
[14:47:32] <synp> https://github.com/http2/http2-spec/issues/47
[14:48:21] Dan Wing leaves the room
[14:48:27] <synp> https://github.com/http2/http2-spec/issues/22
[14:48:49] <synp> https://github.com/http2/http2-spec/issues/21
[14:49:39] <synp> https://github.com/http2/http2-spec/issues/104
[14:50:19] <synp> https://github.com/http2/http2-spec/issues/183
[14:51:29] Dan Wing joins the room
[14:52:20] <synp> https://github.com/http2/http2-spec/issues/67
[14:53:30] <synp> https://github.com/http2/http2-spec/issues/174
[14:53:59] danyork leaves the room
[14:55:33] nemo leaves the room
[14:57:38] nygren@mit.edu/barnowl leaves the room
[14:58:55] <Julian> I like how Bill said "my browser"
[14:59:11] <Ted> Will Chan was speaking.
[14:59:11] nygren@mit.edu/barnowl joins the room
[14:59:18] <Ted> Now it is Roberto Peon
[14:59:32] <synp> Barry next
[15:00:07] <MikeBishop> mic:  Seems like the main reason a receiver wouldn't notice a duplicate is because it's a setting value they aren't using anyway.  Implementations need to validate things they'll be using, but they don't have to validate a setting they intend to ignore.
[15:00:29] <synp> in queue
[15:00:50] <MikeBishop> n/m - same thing Martin said.
[15:01:05] <synp> OK. out of queue
[15:01:13] <synp> dequeued?
[15:01:19] <synp> sat down?
[15:03:19] sa10kan3@gmail.com joins the room
[15:05:59] <fielding> link?
[15:06:09] <synp> https://github.com/http2/http2-spec/issues/171
[15:06:18] <synp> https://github.com/http2/http2-spec/issues/118
[15:08:26] <synp> https://github.com/http2/http2-spec/issues/17
[15:09:06] <synp> https://github.com/http2/http2-spec/issues/112
[15:09:24] <synp> https://github.com/http2/http2-spec/issues/118
[15:09:53] <synp> https://github.com/http2/http2-spec/issues/177
[15:10:49] Ted leaves the room
[15:12:39] Barry Leiba leaves the room
[15:13:28] <synp> https://github.com/http2/http2-spec/issues/158
[15:17:17] Julian leaves the room: Computer went to sleep
[15:19:27] synp leaves the room
[15:19:29] redaka leaves the room
[15:19:29] adb leaves the room
[15:19:43] SM leaves the room
[15:20:01] MikeBishop leaves the room: offline
[15:20:06] andrey.uzunov leaves the room
[15:20:32] jessian leaves the room
[15:20:36] <JcK> @synp: many thanks. This would have been ocmpletely impossible for me to follow without you.
[15:20:39] mnot leaves the room
[15:20:50] JcK leaves the room
[15:20:54] kmurchison leaves the room
[15:21:06] sa10kan3@gmail.com leaves the room
[15:22:35] fielding leaves the room
[15:24:44] PasiS leaves the room
[15:25:28] Florian Zeitz leaves the room: offline
[15:28:34] Wendy Seltzer leaves the room
[15:29:46] Dan Wing leaves the room
[15:31:41] adb joins the room
[15:36:48] nemo joins the room
[15:37:18] g.e.montenegro leaves the room
[15:37:23] adb leaves the room
[15:45:34] nemo leaves the room
[15:55:07] Wendy Seltzer joins the room
[15:56:13] tony.l.hansen leaves the room
[15:58:49] andrey.uzunov joins the room
[15:59:20] sa10kan3@gmail.com joins the room
[16:00:02] sa10kan3@gmail.com leaves the room
[16:00:47] andrey.uzunov leaves the room
[16:12:00] PasiS joins the room
[16:39:03] g.e.montenegro joins the room
[16:41:03] g.e.montenegro leaves the room
[16:42:43] mnot joins the room
[16:47:37] Wendy Seltzer leaves the room
[16:48:45] Julian joins the room
[16:48:45] Julian leaves the room
[16:49:51] Wendy Seltzer joins the room
[17:10:38] Wendy Seltzer leaves the room
[17:21:25] Wendy Seltzer joins the room
[17:28:09] Wendy Seltzer leaves the room
[17:29:34] mnot leaves the room
[17:46:16] PasiS leaves the room
[17:52:24] mnot joins the room
[17:52:40] mnot leaves the room
[17:52:44] mnot joins the room
[17:58:39] mnot leaves the room
[18:11:56] Wendy Seltzer joins the room
[18:14:32] Wendy Seltzer leaves the room
[22:16:17] nemo joins the room
[22:24:08] mnot joins the room
[22:27:52] nemo leaves the room
[22:35:59] mnot leaves the room