[08:53:37] roberto_peon joins the room [08:56:45] ilari.liusvaara joins the room [09:04:42] mcmanus joins the room [09:06:11] o/ [11:10:08] Bill Hulley joins the room [11:11:17] plh joins the room [11:18:16] plh leaves the room [11:34:47] mcmanus leaves the room [12:01:32] wseltzer joins the room [12:41:47] Dirk-Willem van Gulik joins the room [12:44:01] jimsch joins the room [12:55:30] Rhys joins the room [12:56:30] bartosz.balazinski joins the room [13:00:31] And it has begun... [13:00:44] tap tap tap [13:01:21] We hear ya fine Mark. [13:01:33] Its tonight for me :) [13:01:41] ted.h joins the room [13:01:49] jjrh70 joins the room [13:01:56] SM joins the room [13:01:58] o/ Ted. [13:02:15] ted.h feels underdressed [13:02:22] Leif Hedstrom joins the room [13:02:24] jjrh70 leaves the room [13:02:24] roberto_peon feels invisible [13:02:24] jjrh joins the room [13:02:48] Barry Leiba joins the room [13:02:50] Remote participant here. [13:02:54] o/ Barry [13:02:57] tony.l.hansen@gmail.com joins the room [13:03:03] loreto.salvatore joins the room [13:03:08] Hiya, Roberto. [13:03:19] Julian joins the room [13:03:19] Julian leaves the room [13:03:25] Yo. I get my chance to see what remote participation is like today :) [13:03:35] Julian joins the room [13:03:36] Feedback welcome. [13:03:50] we should have asked for a meetecho support [13:03:50] Adam Bishop joins the room [13:04:09] I'm sure it'll work out. [13:04:58] teole05 joins the room [13:05:27] Eliot Lear joins the room [13:05:42] mcmanus joins the room [13:05:46] hi- if you have comments for the microphone, please preface with MIC: [13:05:57] Wolfgang Beck joins the room [13:06:05] now on http/1.1- julian speaking [13:06:18] mdonnelly joins the room [13:06:37] Eliot Lear leaves the room [13:06:40] Eliot Lear joins the room [13:06:59] nakakura joins the room [13:07:07] wseltzer joins the room [13:07:09] frodek joins the room [13:07:48] wseltzer leaves the room [13:07:52] thomas.mangin joins the room [13:08:28] hiro1017 joins the room [13:09:35] now on slide 9 [13:09:38] m&m joins the room [13:10:16] Joe Hildebrand joins the room [13:10:23] Joe Hildebrand is now known as hildjj [13:13:07] Yeay Julian, Roy, Mark, et al! [13:13:11] npdoty joins the room [13:14:24] now on zurich interim [13:14:34] Are there slides? [13:14:43] (checking) [13:16:40] not that i see: https://datatracker.ietf.org/meeting/89/materials.html#wg-httpbis [13:17:13] now on http://www.ietf.org/proceedings/89/slides/slides-89-httpbis-2.pdf [13:17:24] kmurchison joins the room [13:17:27] Hirotaka speaking [13:18:03] Dirk-Willem van Gulik is struggling to find a URL to these slides [13:18:07] correction- not hirotaka but his co-author [13:18:19] Dirk: http://www.ietf.org/proceedings/89/slides/slides-89-httpbis-2.pdf [13:18:24] Adam Bishop leaves the room [13:18:38] Wolfgang Beck leaves the room [13:18:48] Eliot: thanks - it showed up empty - but a cmd-reload seem to fare better. [13:18:56] slide 7 noiw [13:19:04] 8 [13:19:09] thanks! [13:19:25] "HPACK Test Case" [13:21:03] m&m leaves the room [13:21:06] m&m joins the room [13:21:19] lovely [13:21:26] MIC: Very cool guys! [13:22:23] now on martin [13:22:28] http/2 [13:23:25] Jxck joins the room [13:23:34] hiro1017 leaves the room [13:23:38] design issues remaining [13:25:31] hpack security [13:25:34] priority [13:26:05] 28 issues, 18 design [13:27:02] alpn update, Andrei [13:27:37] is peter lepeska in the room? [13:27:53] - now on "Changes since IETF88" on TLS [13:28:30] adamb joins the room [13:29:19] now on "ALPN Deployment" [13:30:10] hiro1017 joins the room [13:31:57] Eliot Lear leaves the room [13:32:04] Eliot Lear joins the room [13:32:12] hmm - a fire alarm ? [13:32:13] npdoty leaves the room [13:32:13] ted.h leaves the room [13:32:14] m&m leaves the room [13:32:14] Leif Hedstrom leaves the room [13:32:15] loreto.salvatore leaves the room [13:32:15] Barry Leiba leaves the room [13:32:16] thomas.mangin leaves the room [13:32:16] adamb leaves the room [13:32:17] nakakura leaves the room [13:32:21] oh god a fire alarm. [13:32:21] frodek leaves the room [13:32:21] mcmanus leaves the room [13:32:23] we have been asked to evacuate [13:32:23] Julian leaves the room: Computer went to sleep [13:32:25] Eliot Lear leaves the room [13:32:26] hildjj leaves the room [13:32:27] bartosz.balazinski leaves the room [13:32:29] tony.l.hansen@gmail.com leaves the room [13:32:30] very polite brittisch :) [13:32:39] that was fast. [13:32:46] Jxck leaves the room [13:32:55] That is a first:) [13:32:59] teole05 leaves the room [13:33:06] MIC: This effort is going so fast, its almost, but not quite on fire. [13:34:00] Joe Hildebrand joins the room [13:34:33] loreto.salvatore joins the room [13:34:35] ted.h joins the room [13:34:43] thomas.mangin joins the room [13:34:45] npdoty joins the room [13:34:45] nakakura joins the room [13:35:08] wseltzer joins the room [13:35:16] Barry Leiba joins the room [13:35:18] teole05 joins the room [13:35:18] wseltzer leaves the room [13:35:20] frodek joins the room [13:35:42] Dirk-Willem van Gulik joins the room [13:35:48] Dirk-Willem van Gulik leaves the room [13:36:10] m&m joins the room [13:36:48] adamb joins the room [13:36:53] Leif Hedstrom joins the room [13:37:31] mdonnelly leaves the room: Disconnected: Replaced by new connection [13:37:34] Julian joins the room [13:37:40] mdonnelly joins the room [13:37:46] ted.h leaves the room: Replaced by new connection [13:37:47] ted.h joins the room [13:38:37] MICL: This is an issue for the registry. [13:38:48] jimsch leaves the room [13:38:55] tony.l.hansen@gmail.com joins the room [13:39:09] MIC; Thus not for us? [13:39:23] jimsch1 joins the room [13:39:40] mic: correct. [13:40:29] g.e.montenegro joins the room [13:40:53] mic: I'm concetrned that this will be used for more thyan just http... [13:40:54] bartosz.balazinski joins the room [13:41:28] micL: I don't want structured stuff... [13:41:47] mic: but I am worried about what happens when it is used for more than what we ahve now. [13:41:56] hiro1017 leaves the room [13:42:42] Jxck joins the room [13:43:02] Joe Hall joins the room [13:43:46] dougt joins the room [13:44:21] mcmanus joins the room [13:44:54] loreto.salvatore leaves the room [13:45:34] loreto.salvatore joins the room [13:45:43] mic: We can do it in a new codepoint if necessary. [13:45:59] mic: or request such from the tls group in the future. [13:46:08] Joe Hildebrand leaves the room [13:46:30] m&m leaves the room [13:46:34] m&m joins the room [13:47:28] mic: no objection [13:48:02] sgtm [13:48:10] So relayed; no need to have it on mic. [13:48:18] *nod* new to this :) [13:48:38] hiro1017 joins the room [13:48:58] we're missing the part that says: don't send this scheme to me again at the http2 layer [13:49:01] ekr@ecotroph.net joins the room [13:50:26] very difficult to hear this speaker in the back of the room, not matter which mic he chooses [13:50:53] You're in the room and can't hear him? [13:51:20] Did that help? [13:51:22] Oh god, yuck [13:52:13] Martin's suggestion seems good. [13:52:35] yes [13:52:39] thank you, Barry [13:52:51] does it make it easier for a passive attacker to correlate the traffic if a fuller set of settings/headers is sent every time? [13:53:40] MIC: I like Martin's suggestion [13:54:20] dcollange joins the room [13:54:21] Join [13:55:33] wmills_92105 joins the room [13:55:35] MIC: If we find that ALPN doesn't work well enough, we can define a new codepoint. [13:55:43] sound is not loud at the back of the room, for any speaker [13:57:23] We need an error code on RST that says that the scheme is denied and the remote endpoint shouldn't send it. [13:58:56] This is likely to take a lot of time to talk about, and this should probably be tabled until the end of the issues. [14:00:27] loreto.salvatore leaves the room [14:00:41] roberto_peon: that would be totally satisfactory to me [14:01:50] ricgold joins the room [14:02:34] ya [14:02:53] roberto_peon bows [14:03:50] loreto.salvatore joins the room [14:03:59] npdoty is not quite willing to just starting twiddling with the knobs in the sound system, for fear of creating a feedback loop [14:04:54] Joe Hall well, it's just gain and you can dial it back… prolly will work better than people trying to remember not to speak so softly [14:05:24] npdoty will blame Joe Hall if I screw something up [14:05:53] mic: no.. I was an advocate for having options open later. I do NOT want to change what we're doing right now. [14:06:30] yes, ok with having alpn identifiers righ tnow. [14:06:47] and. sorry, the audio is delayed, and I have to type my answers, so some delay.. :/ [14:06:48] Joe Hall that's amazingly better! [14:07:07] npdoty success! [14:07:08] Joe Hall will buy you your weight in beer [14:07:12] yes, sound is much better in the room now [14:07:18] yes [14:07:18] reminder: the front mic is CH01 [14:07:23] wseltzer thanks, Nick! [14:08:31] geir joins the room [14:09:59] This follows the security discussion. [14:10:06] defer, please. [14:11:10] http://greenbytes.de/tech/webdav/draft-ietf-httpbis-p1-messaging-26.html#establishing.authority [14:12:37] Eliot Lear joins the room [14:13:39] dougt leaves the room [14:14:14] There was good information posted about deflate having problems, and thus preferring gzip. [14:14:18] which sgtm [14:15:26] > client has to support; server not — consequence on the client only [14:15:43] good luck with that :) [14:18:48] Eliot Lear leaves the room [14:19:07] MIC: This requirement for static data is so fundamental to performance that HTTP/1 would likely be faster if we denied its use in HTTP/2 [14:20:18] Dan York joins the room [14:21:36] yes. [14:21:46] This was originally intended to be symmetric. [14:21:58] wondering if EKR is worried about implementers thinking GZIP is a security tool when it's not [14:22:20] No, I'm worried that compression of dynamic content is a security threat [14:22:25] ah [14:22:28] because it is :) [14:22:40] > Joe — no the issue is that compression itself is a security issue. [14:22:56] no-- compression of content with mixed authors is a security issue. [14:23:04] authors/origins/blah blah [14:23:45] > we know deflate can be an issue (beast, crime) - we do not (yet) know for gzip. [14:23:50] is the risk to confidentiality of the content, or some other security risk? [14:23:52] roberto: I don't disagree with your characterization [14:23:53] they are the same [14:23:59] Thanks for clarifying [14:24:06] gzip/deflate are the same risk. NP ekr. :) [14:24:23] > roberto - violently agree - any compression really. [14:24:37] > dynamic traffic over tls is the crux. [14:24:55] mostly-- LZ77 based compressors are better at leaking info than others. [14:24:56] > that is where the extra risk originates imho. so the sec-considerations should highlight this. [14:25:02] yes [14:25:08] gzip: yes [14:26:04] which issue? [14:26:14] I think we are back on the same issue [14:26:47] decopmpression is often cheaper than IO. [14:27:34] the nuance for body vs. header+body compression as it interacts with encryption is interesting, but I'd much rather see the info leakage just get solved. [14:27:59] michael.r.sweet joins the room [14:28:26] Joe Hall leaves the room [14:28:29] Joe Hall joins the room [14:28:33] Agree with roberto that multiple authors is the issue, but info reflected off the server by the attacker can also be a problem. [14:29:43] MIC: WS and other protcols, and at least as interestingly, loadbalance them all. We'd need the RST for unsupported scheme thing. [14:31:03] maybe julian could do the editorial work now that he's got time on his hands :) [14:31:44] dcollange leaves the room [14:32:28] RST error code at the HTTP2 layer [14:32:35] NOT an http error code. [14:32:45] you got it right. [14:33:23] Julian leaves the room [14:35:06] Julian joins the room [14:35:32] correct. [14:35:56] You don't want the client to waste time on sending the server a scheme it will not support (ostensibly because it doesn't have code to support it) [14:38:45] WS doesnt' use HTTP status codes. [14:40:22] MIC: no no.. careful Will, when we talkeda bout this we required structure! [14:40:26] sftcd joins the room [14:40:33] mic: And we don't want to require structure now. [14:40:42] (for ALPN I mean) [14:43:19] I hope I got that to the mic for you roughly in time. [14:45:50] mic: In all cases on out-of-band knowledge, the browser or client MUST be willing to learn that it now fails. DNS is no different from having learnt from a prior connection in-band. [14:45:56] tony.l.hansen@gmail.com leaves the room [14:45:57] yes, you got it in time, thanks :) [14:46:01] +1 to roberto [14:46:53] I don't see any reason it should block. [14:47:13] however... [14:47:39] dude.... you need a robo-telepresence [14:47:40] .. we do need to suggest at least how http/2 should fail if it is not successful. [14:47:42] :) [14:47:49] cyrus joins the room [14:47:56] I'd have been there, as usual, but had family obligations :/ [14:47:58] want me to take that to the mic? [14:48:02] yea [14:48:03] thanks :) [14:50:13] If you believe you can connect via HTTP/2, and you fail to negotiate http/2 [14:50:37] .. then what does the client do? That is the question to which we should suggest an answer. I.e. the client should either be prepared to fall-back or not. [14:52:13] or the server reboots... [14:52:19] yes. [14:52:29] jsut discuss the failure case more. this is independent of dns. [14:52:42] I don't think this question is DNS-only; isn't it true for any other source of prior knowledge? [14:53:43] yes. that is what I was trying to say waaaaay above. [14:55:28] hiro1017 leaves the room [14:57:07] Julian leaves the room [14:58:22] m&m leaves the room: Disconnected: closed [14:58:30] not right now [14:58:38] tony.l.hansen@gmail.com joins the room [14:59:03] ambivalence. [14:59:43] Hey, I imlpement it :) [15:00:49] sftcd leaves the room [15:01:10] Joe Hall leaves the room [15:01:49] frodek leaves the room [15:01:57] loreto.salvatore leaves the room [15:01:58] ted.h leaves the room [15:02:03] adamb leaves the room [15:02:04] ted.h joins the room [15:02:05] bye all! [15:02:06] thomas.mangin leaves the room [15:02:09] geir leaves the room [15:02:13] jimsch1 leaves the room [15:02:13] cyrus leaves the room [15:02:16] roberto_peon leaves the room: offline [15:02:16] Barry Leiba leaves the room [15:02:25] SM leaves the room [15:02:25] Jxck leaves the room [15:02:32] Leif Hedstrom leaves the room [15:02:36] bartosz.balazinski leaves the room [15:02:41] tony.l.hansen@gmail.com leaves the room [15:02:47] nakakura leaves the room [15:02:47] Rhys leaves the room [15:02:49] mcmanus leaves the room [15:04:32] ricgold leaves the room [15:04:35] nakakura joins the room [15:06:25] Dan York leaves the room [15:06:33] ekr@ecotroph.net leaves the room [15:08:25] ted.h leaves the room [15:09:10] michael.r.sweet leaves the room [15:10:18] kmurchison leaves the room [15:10:53] nakakura leaves the room [15:14:37] wmills_92105 leaves the room [15:14:49] npdoty leaves the room [15:15:46] thomas.mangin joins the room [15:17:51] metricamerica joins the room [15:18:27] Dan York joins the room [15:18:47] ekr@ecotroph.net joins the room [15:19:14] thomas.mangin leaves the room [15:19:38] mdonnelly leaves the room: Disconnected: Replaced by new connection [15:19:41] mdonnelly joins the room [15:20:15] Joe Hall joins the room [15:20:31] Dan York leaves the room [15:20:34] adamb joins the room [15:20:48] teole05 leaves the room [15:21:03] Joe Hall leaves the room [15:21:19] wseltzer leaves the room [15:22:15] ted.h joins the room [15:24:28] geir joins the room [15:24:40] thomas.mangin joins the room [15:25:01] geir leaves the room [15:25:05] nakakura joins the room [15:25:06] loreto.salvatore joins the room [15:25:40] wmills_92105 joins the room [15:25:49] wmills_92105 leaves the room [15:25:55] ted.h leaves the room: Replaced by new connection [15:25:56] ted.h joins the room [15:26:03] sftcd joins the room [15:26:09] sftcd leaves the room [15:26:56] mdonnelly leaves the room: Disconnected: Replaced by new connection [15:27:00] mdonnelly joins the room [15:27:00] thomas.mangin leaves the room: Disconnected: Replaced by new connection [15:27:01] thomas.mangin joins the room [15:31:04] thomas.mangin leaves the room: Disconnected: closed [15:31:05] thomas.mangin joins the room [15:31:18] loreto.salvatore leaves the room [15:31:22] mcmanus joins the room [15:31:58] mcmanus leaves the room [15:32:04] adamb leaves the room [15:32:38] frodek joins the room [15:33:49] nakakura leaves the room [15:34:15] nakakura joins the room [15:34:24] loreto.salvatore joins the room [15:34:57] tony.l.hansen@gmail.com joins the room [15:35:02] wseltzer joins the room [15:35:35] frodek leaves the room [15:37:06] loreto.salvatore leaves the room [15:40:34] ted.h leaves the room [15:45:11] nakakura leaves the room [15:46:49] mdonnelly leaves the room: Disconnected: Replaced by new connection [15:46:52] mdonnelly joins the room [16:00:09] npdoty joins the room [16:00:38] thomas.mangin leaves the room: Disconnected: Replaced by new connection [16:00:38] thomas.mangin joins the room [16:00:47] npdoty leaves the room [16:02:55] loreto.salvatore joins the room [16:04:08] thomas.mangin leaves the room: Disconnected: Replaced by new connection [16:04:55] Bill Hulley leaves the room [16:05:16] jjrh leaves the room [16:05:54] thomas.mangin joins the room [16:13:06] m&m joins the room [16:13:19] m&m leaves the room [16:13:23] m&m joins the room [16:13:59] Dirk-Willem van Gulik leaves the room [16:15:45] m&m leaves the room [16:18:26] loreto.salvatore leaves the room [16:19:05] thomas.mangin leaves the room: Disconnected: closed [16:19:07] thomas.mangin joins the room [16:20:57] thomas.mangin leaves the room: Disconnected: Replaced by new connection [16:20:58] thomas.mangin joins the room [16:29:34] tony.l.hansen@gmail.com leaves the room [16:31:20] ekr@ecotroph.net leaves the room [16:31:49] thomas.mangin leaves the room [16:40:16] tony.l.hansen@gmail.com joins the room [16:42:20] wseltzer leaves the room [16:42:47] tony.l.hansen@gmail.com leaves the room [16:43:08] tony.l.hansen@gmail.com joins the room [16:47:40] mdonnelly leaves the room: Disconnected: Replaced by new connection [16:47:43] mdonnelly joins the room [16:54:36] tony.l.hansen@gmail.com leaves the room [16:54:49] ilari.liusvaara leaves the room [16:56:50] metricamerica leaves the room [17:08:23] Eliot Lear joins the room [17:19:36] nakakura joins the room [17:23:24] Eliot Lear leaves the room [17:34:33] mdonnelly leaves the room [17:34:59] Eliot Lear joins the room [17:54:29] teole05 joins the room [18:04:02] tony.l.hansen@gmail.com joins the room [18:10:00] loreto.salvatore joins the room [18:10:17] tony.l.hansen@gmail.com leaves the room [18:33:51] wseltzer joins the room [18:33:56] Eliot Lear leaves the room [18:33:58] Eliot Lear joins the room [18:51:09] loreto.salvatore leaves the room [18:56:17] nakakura leaves the room [19:24:18] metricamerica joins the room [19:33:10] Eliot Lear leaves the room [19:33:19] Eliot Lear joins the room [19:33:31] metricamerica leaves the room [19:33:33] metricamerica joins the room [19:52:49] teole05 leaves the room [20:14:43] Eliot Lear leaves the room [20:16:02] wseltzer leaves the room [20:21:48] metricamerica leaves the room [22:29:43] nakakura joins the room [22:34:40] wseltzer joins the room [23:35:24] teole05 joins the room [23:46:05] Eliot Lear joins the room [23:53:47] Eliot Lear leaves the room