[05:59:27] David Bryan joins the room [06:05:10] yinxing joins the room [06:05:19] kpfleming joins the room [06:12:29] spencerdawkins joins the room [06:12:56] FYI Spencer, audio feed is up and running and good (and it is a sensitive mic ;) ) [06:13:21] spencerdawkins: nice email address [06:13:59] redaka joins the room [06:14:04] tomkri has set the subject to: http://tools.ietf.org/wg/p2psip/agenda?item=agenda76.html [06:22:02] Adam Roach joins the room [06:23:31] shamus joins the room [06:23:38] hirotaka.sato joins the room [06:24:04] wolfgang.beck01 joins the room [06:24:30] victor joins the room [06:24:37] Audio at http://videolab.uoregon.edu/events/ietf/ietf767.m3u [06:24:51] Ted joins the room [06:25:23] enrico joins the room [06:25:33] SoftgearKo joins the room [06:25:45] cullen presenting reload [06:27:01] ..some data structure changes [06:27:09] no one disagree [06:27:23] Masashi joins the room [06:28:53] RjS joins the room [06:29:22] issues with turn servers density discovery [06:30:20] cullen's proposal: leave base draft as is, work on a good service discovery mechanism [06:30:30] There were questions/concerns early about if this really works. Glad to see that data is out there that supports it. I might have missed it, but can you share (or reshare if I missed it) the links/data on the list that you are talking about (both the sensitivity analysis you did and the production data you mentioned?) so we can make a better decision? [06:30:55] Emil Ivov joins the room [06:31:07] (spencer's relaying davi) [06:31:07] suzukisn joins the room [06:31:15] (david) [06:31:15] mlepinski joins the room [06:31:32] jon-ietf joins the room [06:31:35] ben@xmpp.estacado.net joins the room [06:32:21] cullen: too much a complex issue, if anyone has data from simulations or whatever please share [06:32:26] brtech joins the room [06:32:47] enrico, could you make sure to interrupt us when david tries to ask like a chair :D [06:33:29] i'll try, but can't promise anything :9 [06:33:51] missed last two questions.. sorry :( [06:33:57] thanks! [06:34:08] i know it's not easy... [06:34:11] current question is why is service discovery only used for TURN and not for all services [06:34:46] stephanwenger joins the room [06:34:50] answer is that when the WG was chartered, it was known that something would be needed for TURN right away, and service discovery for other services could wait until later when more mature discovery protocols exist [06:35:09] (cullen's answer) [06:37:45] side about proposing to integrate s. 5.6.1 from self tuning draft in base specs [06:38:17] s/side/slide (i'm typing on an eeepc :() [06:39:45] gonzalo supports [06:40:15] supports as a should [06:40:42] open mic for other issues [06:40:53] people in the room: shoot! [06:41:31] no issues, cullen thinks the draft is ready for wglc [06:41:35] Need really for real this time serious we will read it reviewers [06:42:14] that should always coincide with a wglc [06:43:15] calvin joins the room [06:44:26] hum for wglc, sortof [06:44:34] consensus [06:44:48] I'll review it [06:45:08] mr sparks [06:45:21] john buford [06:45:35] and david from the jabber room [06:45:41] volunteer for review [06:47:02] more volunteers: jouni [06:47:21] Share the list of reviewers on list please [06:48:04] next presentation: security [06:48:17] haibin presents [06:48:47] david, do you need to say anything about adoption as a WG item? [06:49:20] consensus to adopt as wg item in stockholm, discussion on the list [06:49:31] Spencer: Not right now. Thx. [06:49:38] thanks! [06:49:44] Bernie joins the room [06:49:48] BTW FYI there is a lag of several seconds in audio [06:49:48] presenter says 2 comments have been addressed [06:50:03] "only 2 substantial comments" [06:50:33] show of hands for who has read the draft: 4-5 [06:52:00] There were objections made on list (EKR, and he said Cullen had them), then an iteration, but I didn't hear if they were happy with iteration [06:52:26] Sorry for lag, I think Cullen is now saying his objection... [06:53:25] my lag is worse, cullen just basically said that he's not sure about the scope of this doc. Is it a tutorial? [06:53:26] One purpose of this was to make sure that people who are not familiar with P2P can understand the unique problems [06:54:03] (spencer relays david again, tnx spencer!) [06:54:06] and why the protocol (and other work in here) looks like it does and the decisions we made [06:54:57] no other comments [06:55:52] next presentation: relay [06:56:04] Emil Ivov leaves the room [06:58:32] comparison of routing approaches, DRR/PRR vs SRR [06:58:36] see slides [06:58:45] cullen at the mic [06:59:13] stephanwenger leaves the room [06:59:22] ... warms up... [06:59:50] (i'm too stupid to understand this slide) [07:00:13] cullen: is this using attach, attach-lite? [07:00:25] what are the assumptions? [07:01:09] roni: it is going to go on a tls connection [07:02:47] Roni please use mic.... [07:03:15] haibinsong@jabber.org joins the room [07:03:24] roni seems to have convince cullen.. [07:03:35] no, still confused [07:03:48] yinxing leaves the room: Replaced by new connection [07:03:48] yinxing joins the room [07:03:51] Masashi leaves the room [07:04:31] can someone who's not totally lost summarize this discussion? [07:05:16] cullen is expressing skepticism about the number of messages exchanged to establish connections on these slides [07:05:33] this is largely a question about what layers of messages you include (e.g. tcp, tls, ice, etc) [07:05:50] so cullen is asking what's being counted on these slides and what isn't [07:06:04] they're taking it offline, anyway [07:06:12] thanks jon! [07:07:33] Masashi joins the room [07:07:34] enrico leaves the room [07:08:07] enrico joins the room [07:08:47] the authors think they've addressed all previous comments, no more comments from the room [07:09:04] next presentation: load balancing models for dht based p2p networks [07:10:57] 3 phases: measure, distribute info, balance [07:13:33] going through different load balancing approaches [07:14:57] If a node holds multiple virtual nodes, will it overload this node? [07:15:24] JXW181977BC3B2203 joins the room [07:16:58] haibinsong@jabber.org leaves the room [07:17:22] haibin joins the room [07:20:36] haibin, do you want that question to get relayed? Have you left the room?? [07:21:09] I think the presenter has answered during the presentation [07:21:42] presentation's over, comments? [07:22:19] 2 people read the draft [07:22:44] 3... gonzalo, cullen and the current speaker, who did not swipe his card [07:23:06] jouni [07:23:35] whose question i missed btw :-/ [07:24:26] next presentation: should be self tuning or something [07:24:41] yeah, self-tuning dht for reload [07:25:31] hirotaka.sato leaves the room [07:26:16] the draft defines a mechanism for self-tuning of standard dhts [07:27:09] Ted leaves the room [07:30:18] 5-6 people read the draft [07:30:52] ben@xmpp.estacado.net leaves the room [07:30:54] Do the authors feel that there is sufficient support in base draft to do the extension they want to do to implement this? [07:30:54] enrico and gonzalo think this is needed [07:31:30] cullen supports [07:32:43] i actually got about 4-5 people who had read the draft, but... [07:32:44] the authors think there is sufficent support [07:33:02] I also wanted to add that I read it and support the work [07:33:23] Emil Ivov joins the room [07:35:05] haibin mentions that this could overlap with diagnostic [07:35:21] Bernie leaves the room [07:35:32] i missed a few calls [07:35:41] last was an hum for adoption [07:35:41] suzukisn leaves the room [07:35:42] suzukisn joins the room [07:35:46] consensus [07:36:55] next presentation, same presenter: service discovery [07:38:07] (in today's meetings this is actually only the 2nd presentation about service discovery i've seen, what's happening?) [07:38:17] heh [07:38:55] Emil Ivov leaves the room [07:41:00] hirotaka.sato joins the room [07:42:34] there is a service discovery mechanism in reload [07:43:57] questions or comments? [07:44:12] 4 people read the draft [07:44:28] I really prefer one general solution, we don't need to have several service discovery mechanisms [07:45:13] gonzalo suggests to ask the group if we want a general service discovery [07:45:44] cullen really wants to adapt zeroconf on reload [07:45:51] err, maybe was sarcastic ;) [07:46:20] hum for a general service discovery [07:46:22] did anyone hum that they didn't want to do service discovery? [07:46:24] weak consensus [07:46:40] no one against, few support [07:47:06] (Unlikely to go out at the end of the meeting...it's 3 am here ;) ) [07:47:14] we were going to get cookies early [07:47:24] jan is keeping us in the room [07:47:43] asks whether the security draft is wg doc [07:48:17] no jan, it's not [07:48:53] suzukisn leaves the room [07:48:56] shamus leaves the room [07:48:58] wolfgang.beck01 leaves the room [07:49:09] haibin leaves the room [07:49:12] RjS leaves the room [07:49:21] hirotaka.sato leaves the room [07:49:46] SoftgearKo leaves the room [07:49:54] spencerdawkins leaves the room [07:50:24] Adam Roach leaves the room: Computer went to sleep [07:50:44] victor leaves the room [07:50:57] kpfleming leaves the room: Disconnected [07:51:30] Masashi leaves the room [07:51:41] enrico leaves the room [07:53:25] brtech leaves the room [07:53:29] David Bryan leaves the room [07:53:32] jon-ietf leaves the room [07:54:30] haibin joins the room [07:54:35] haibin leaves the room [07:55:44] David Bryan joins the room [07:55:53] David Bryan leaves the room [07:57:43] calvin leaves the room [07:57:51] redaka is now known as Takehito Akagiri [07:58:19] yinxing leaves the room: offline [08:00:38] calvin joins the room [08:00:49] calvin leaves the room [08:06:30] mlepinski leaves the room [08:09:07] RjS joins the room [08:09:54] RjS leaves the room [08:22:03] JXW181977BC3B2203 leaves the room [08:22:09] JXW181977BC3B2203 joins the room [08:25:08] tomkri leaves the room [08:26:09] Takehito Akagiri leaves the room [08:44:55] calvin joins the room [08:45:28] RjS joins the room [08:46:37] calvin leaves the room [08:50:34] RjS leaves the room [08:50:51] Emil Ivov joins the room [08:51:07] Emil Ivov leaves the room [09:25:15] JXW181977BC3B2203 leaves the room [11:10:12] Adam Roach joins the room [11:11:37] Adam Roach leaves the room [13:59:31] tomkri joins the room [14:01:10] tomkri leaves the room