[00:23:49] Pete McCann joins the room [00:42:11] Martin Stiemerling joins the room [00:50:37] bortzmeyer joins the room [00:50:53] Ni hao [00:50:55] becarpenter joins the room [00:53:43] alexmcmahon leaves the room [00:57:43] Hyong-Jong Paik joins the room [00:59:11] Lars joins the room [00:59:36] Dave Thaler joins the room [01:01:56] pk [jabber scribe] joins the room [01:02:31] pk [jabber scribe] has set the subject to: NBS BoF in garden ballroom 3 [01:02:45] Javier Ubillos joins the room [01:02:49] exploration BoF, not talking about how to come to WG [01:03:13] presenters: do adhere to the schedule! [01:03:55] anybody on audio? soundcheck? [01:04:09] checking.. [01:04:11] It's coming through fine. [01:04:12] alexmcmahon joins the room [01:04:17] glen joins the room [01:04:18] tnx [01:04:26] Christian is a little too close to the mic. [01:04:29] glen leaves the room [01:04:42] first slide [01:04:43] samante joins the room [01:05:18] http://www.ietf.org/proceedings/79/slides/nbs-6.pdf [01:05:28] http://www.ietf.org/proceedings/79/slides/nbs-6.pdf, slide 2 [01:06:28] slide 3, agenda OK? [01:06:29] wmhaddad joins the room [01:06:31] tsavo_work@jabber.org/Meebo joins the room [01:06:57] hmm these slides are not nbs-6, nbs-6 is the note well and logistics [01:07:15] sorry, checking and downloading [01:07:39] it's actually nbs-7 [01:07:44] http://www.ietf.org/proceedings/79/slides/nbs-7.pdf [01:08:23] tnx, it's "overviwe" [01:08:38] now slide with hourglass [01:08:53] jounkorh joins the room [01:09:06] ꈲ joins the room [01:09:50] slide 4 of 5 in nbs-7.pdf [01:11:58] Colman Ho joins the room [01:12:35] slide 5/5 [01:13:54] jelte joins the room [01:14:44] Dave Thaler nbs-4.pdf, slide 2 [01:15:19] slide 3/8 [01:17:35] slide 4/8, host mobility [01:20:36] slide 5/8, mobility 2 [01:22:38] slide 6/8 [01:22:52] aboba joins the room [01:23:13] unknown@mic: user experience benefit? [01:23:32] Why a cement pillar? Is cement specially hostile to radio waves? [01:24:08] back to slide 6/8 [01:26:07] slide 7/8 [01:27:12] last slide (8/8) [01:27:43] ꈲ leaves the room [01:29:08] end of presentation [01:29:14] no Q/A [01:29:41] Stuart Cheshire next [01:30:43] Dave Thaler leaves the room [01:30:45] Dave Thaler joins the room [01:30:50] cannot find slides, moment [01:31:42] btw the "unknown@mic: user experience benefit?" was David Harrington (AD) [01:32:15] I'm unhappy about the unavailability of slides. [01:32:32] resnick joins the room [01:32:59] chairs will post them now [01:33:11] ꈲ joins the room [01:33:16] slides being uploaded [01:33:26] [some trouble with projector] [01:33:51] they're trying to either increase the brightness (or decrease the illumination in the room) [01:34:37] likely to come up as http://www.ietf.org/proceedings/79/slides/nbs-8.pdf [01:35:23] not reachable... [01:35:28] mstiemerling joins the room [01:35:29] Not yet uploaded... [01:35:39] jounkorh leaves the room [01:35:42] do AAAA and A DNS queries in parallel ... [01:35:51] The short version of the current slides is that both A & AAAA records are resolved in paralell. [01:36:03] Martin Stiemerling leaves the room [01:36:06] mstiemerling leaves the room [01:36:23] and that both responses need to be available to make the decision which way to go [01:36:41] samante leaves the room [01:36:42] one may be cached, the other not (or not exist) [01:36:50] Martin Stiemerling joins the room [01:37:05] jounkorh joins the room [01:37:25] for the slides: they came very,very late and the datatracker is now taking 5 mins + to upload them. SORRY for this [01:38:53] query for AAAA only if machine has routable v6 address (solution only for !ipv6) [01:39:00] samante joins the room [01:39:39] here are the slides: http://www.ietf.org/proceedings/79/slides/nbs-8.pdf [01:39:45] decouple data paths [01:39:47] Got them, thanks. [01:40:58] I don't know the slide number, but the title of the current slide is "Apple Core Foundation APIs". [01:42:00] Slide: "summary" [01:42:17] probably last slide, download not yet complete ... [01:42:29] 0ver 10 mins for me (on the IETF network :) [01:42:56] the wireless is not really good right now... [01:43:02] end of presentations, Q/A [01:43:16] next speaker: Javier Ubillos [01:43:19] samante leaves the room [01:43:20] nbs-3.pdf [01:44:13] title: slide 1/22 [01:44:32] slide 2/22 [01:44:47] link is http://www.ietf.org/proceedings/79/slides/nbs-3.pdf [01:44:59] samante joins the room [01:45:04] jelte leaves the room: Replaced by new connection [01:45:05] jelte joins the room [01:45:34] slide 3/22 [01:45:37] goals [01:46:45] slide 4/22: constraints [01:47:28] slide 5/22: approach [01:48:39] slide 6/22: how is this different [01:49:38] slide 7/22: Uni-/Bi-lateral [01:50:02] {audio experience in room is poor due to air condition, what's the outside quality?} [01:50:15] I'm getting a lot of clipping. [01:50:19] wolfgang.beck01 joins the room [01:51:07] slide 8/22: time line example [01:51:13] Colman Ho leaves the room [01:51:29] slide 9/22 overview; slide 10/22 [01:52:10] slide 11/22, api status [01:52:20] Martin Stiemerling leaves the room: Replaced by new connection [01:52:23] martin.stiemerling joins the room [01:52:52] slide 12 [01:53:54] slide 13/22 [01:54:16] slide 14/22 [01:55:05] So this proposal proposes using IP extension headers? Will those really get through the middleboxes (firewalls)? [01:55:06] becarpenter leaves the room [01:55:09] slide 15/22, name shim6 viz [01:55:14] slide 16/22 [01:56:00] {for relay requests pls use "mic:" prefix or similar} [01:56:05] Colman Ho joins the room [01:57:43] Thanks! [01:58:08] slide 17/22 [01:58:17] slide 18/22 [01:59:37] slide 19 [02:00:01] slide 20 [02:01:22] samante leaves the room [02:01:41] teemu @mic: [02:01:57] back to slide 10 [02:02:13] Shane Amante joins the room [02:02:13] Shane Amante is now known as samante [02:02:13] samante is now known as Shane Amante [02:02:40] Shane Amante is now known as samante [02:03:34] samante is now known as Shane Amante [02:03:42] slide 20 again [02:04:01] {sorry, audio too poor} [02:04:54] slide 21 [02:05:01] slide 22, end of presentation, Q/A [02:05:06] dave thaler @mic [02:05:22] why ip addresses in ip6.arpa format? [02:05:29] JU: easy way to do it [02:05:56] JU: no problem changing that [02:06:13] Erik Nordmark: when combining with shim6, uid variable length string [02:06:13] wolfgang.beck01 leaves the room [02:06:38] JU: md5 hash of name used as an IP address [02:07:23] next: zhongxing ming [02:07:40] http://www.ietf.org/proceedings/79/slides/nbs-2.pdf [02:07:52] Javier Ubillos leaves the room [02:07:56] nobobo joins the room [02:08:23] slide 2/3 [02:08:35] Shane Amante is now known as samante [02:09:33] nobobo leaves the room [02:09:35] slide 3/3 [02:10:14] Javier Ubillos joins the room [02:10:51] bhoeneis joins the room [02:12:25] EoP, Q/A [02:13:12] next: Saleem Bhatti, DNS Caching: runing on zero [02:13:36] samante is now known as Shane Amante [02:13:58] slides? [02:15:27] This is a short notice presentation, I'll try to get them for you. [02:15:33] may try http://www.nanog.org/meetings/nanog50/presentations/Tuesday/NANOG50.Talk64.bhatti-nanog50_dns.pdf [02:15:34] being uploaded [02:15:48] http://www.ietf.org/proceedings/79/slides/nbs-9.pdf [02:15:54] same here: late. [02:16:14] future URL http://www.ietf.org/proceedings/79/slides/nbs-9.pdf [02:16:22] oops, sorry for dup [02:16:24] :) I was too slow :) [02:16:44] this was only because the datatracker and the net are fast again :) [02:17:48] download not yet complete ... basic dataset meta-data [02:18:36] Shane Amante is now known as samante [02:18:36] samante is now known as Shane Amante [02:18:49] that's slide 9/16 [02:18:54] jelte leaves the room [02:19:25] slide 7 [02:20:41] slide 11 [02:21:05] slide 12 [02:21:26] slide 13 [02:21:49] slide 14 [02:22:28] slide 15 skipprd, 16 [02:22:38] slide 17 [02:23:03] slide 18 [02:23:37] slide 19/19, EoP, Q/A [02:23:37] Shane Amante is now known as samante [02:23:37] samante is now known as Shane Amante [02:24:01] toerless@mic: how can you conclude frequent access is practical? [02:25:08] aaron falk: server centric analysis [02:25:21] client may have different experience, any analysis on that? [02:25:33] 1s TTL vs 0s TTL [02:26:14] dave thaler: did you change the RDATA? [02:26:53] no caching in recursive resolvers larger than ttl? [02:27:50] what workload did the clients perform? [02:28:18] how much was intra site destinations? [02:28:38] Shane Amante is now known as samante [02:28:51] SB: do not recommend everybody move to 0 TTL [02:28:57] chairs cut discussion [02:29:13] next speaker: Brian Carpenter [02:29:29] http://www.ietf.org/proceedings/79/slides/nbs-1.pdf [02:29:49] slide 1/15, title [02:30:09] slide 2/15 [02:30:40] slide 3/15 [02:31:44] slide 4/15 [02:32:23] slide 5/154 [02:32:31] s/154/15/ [02:33:03] slide 6 [02:33:39] samante is now known as Shane Amante [02:33:39] Shane Amante is now known as samante [02:33:45] Lars Eggert (AD) @mic [02:33:52] wmhaddad leaves the room [02:33:56] address or address+port? [02:34:06] BC: yes! [02:34:19] slide 7/15 [02:35:12] slide 8/15: real scopes [02:35:44] bhoeneis leaves the room [02:35:49] bhoeneis joins the room [02:36:25] slide 9 [02:36:54] slide 10 [02:38:19] slide 11 [02:38:40] samante is now known as Shane Amante [02:38:40] Shane Amante is now known as samante [02:39:18] JLCjohn joins the room [02:39:57] slide 12 [02:40:43] slide 13: conclusion [02:41:17] EoP, Q/A [02:41:48] no Q/A; last Speaker: Matthias Wählisch [02:42:14] alexmcmahon leaves the room [02:42:14] http://www.ietf.org/proceedings/79/slides/nbs-5.pdf [02:42:29] slide 2/3: problem statement [02:43:42] samante is now known as Shane Amante [02:44:04] slide 3/3 [02:46:07] EoP, Q/A [02:46:24] Erik Nordmark: how do you resolve the "cnn" part in your example URI? [02:46:37] DNS or something else [02:47:09] Toerless Eckert: SSM vs global resolvability [02:47:23] end of presentation session [02:48:27] discussion opened, questions to be answered on slide 5 in nbs-7.pdf [02:48:43] Shane Amante is now known as samante [02:48:43] samante is now known as Shane Amante [02:48:47] Toni Li: to Dave and Stuart: portable API? [02:49:25] Erik Nordmark: also important for middleware providers [02:49:56] Dave Thaler: W3C has platform independent APIs in the browser [02:50:01] (as an example) [02:50:47] Austin Group "owns" C APIs, so not to be done by IETF - many different frameworks [02:51:38] Toerless: Austin group potentially not experts on topic at hand [02:51:59] Dave: job of the API may be definition of an "abstract API", cf. GSS API [02:52:26] s/job of the API/job of the IETF/ [02:52:27] finlayson joins the room [02:53:43] Shane Amante is now known as samante [02:54:02] finlayson leaves the room [02:54:11] toerless: primary objective of presenters was how to add their extensions to the ip stack [02:55:15] Jari Arkko: to what extent if we delevlop abstract API with names i/o addresses would other orgs be able to use this? [02:55:57] Dave Thaler: {refers to presentation} people are moving to name based apis already [02:57:56] Erik Nordmark: disagree wrt benefits of concrete API [02:58:44] samante is now known as Shane Amante [02:59:30] Christian: Erik, can you elaborate on IETF relation with other orgs re: IPv6 API? [03:00:36] Dave Thaler: basic sockets API did not go to Austin Group before being reviewed @IETF, multicast src filter api was different [03:02:28] DT: agree with Erik on value of the concrete API; doesn't need to be done in the IETF, but same time (in other og) [03:02:56] Liam Casey: on question 3 (feasible to solve problem)? [03:03:28] Javier Ubillos: name exchange w/o authentication [03:03:45] Shane Amante is now known as samante [03:03:45] samante is now known as Shane Amante [03:05:17] LiamC: wonder whether this will be acceptable to the IETF security people [03:05:49] Christian summarizes: could be name verification along the communication i/o at the beginning [03:07:06] Andrew McGregor: doing s/th hip like [03:07:30] jounkorh leaves the room [03:07:51] Erik Nordmark: some unknowns here: how early to do validation depends on how to deal with TCP conn [03:08:46] Shane Amante is now known as samante [03:09:44] Christian: initiators "real" name necessary? [03:09:44] Colman Ho leaves the room [03:11:40] Dave Thaler: 3 cases; ...; the other two: ip address reachability, may be true for names (name doesn't have to be resolvable, client can have multiple names) [03:11:49] Colman Ho joins the room [03:11:55] Zhen Tsao joins the room [03:11:59] resovable by the recipient or verify that two are the same [03:13:19] Javier Ubillos @mic [03:13:47] samante is now known as Shane Amante [03:14:02] Lars Eggert (AD): change track of discussion -> 2nd bullet (demand by apps, os vendeors)? [03:14:31] problem is developing protocols that see adoption rather than protocols that work [03:15:07] Erik Nordmark: benefit of having API (provided this turns into concrete API) is worth it [03:15:17] alexmcmahon joins the room [03:16:25] Jari Arkko: lars was spot on [03:17:19] sceptical about protocol stuff, api is of course interesting, but no clear answer how much is to be done [03:17:33] Toerless Eckert: not sure how to separate API from protocol [03:18:38] Yuri Ismailov: question of API is matter of taste like which editor to use [03:18:48] Shane Amante is now known as samante [03:18:49] Yuri Ismalov @ mic [03:20:33] Dave Thaler: three different questions to ask: 1) do we need to address APIs, 2) need to address behaviour underneath API, 3) protocol [03:21:21] Erik Nordmark [03:23:06] Lars Eggert: if you can structure API so it can be offered to local user base, might give you partial benefit for partial deployment [03:23:23] "why would the first guy implement it?" [03:23:50] samante is now known as Shane Amante [03:24:16] Christian: two feature sets: initial, local vs. what happens if all the peers kick in in a later stage of deployment [03:24:24] Randall Stewart [03:24:35] {mic line closed} [03:25:19] Who's talking now? [03:25:34] Randall Stewar[dt] @mic [03:25:34] randy stewart [03:25:37] Oh, randal, sorry. [03:26:41] Lars Eggert: BSD socket API had sendfile extension that avoided app level copy function; adoption took years [03:27:00] Martin: sum up [03:27:38] do people think it's useful to have enhanced socket API with FQDN defined by the IETF? [03:28:05] 15 people hand - bad idea? 2 hands [03:28:21] who thinks IETF should define behaviour of API? 25 [03:28:30] bad idea? 0 [03:28:48] who believes it's a good idea to work on protocoll below API [03:28:51] Shane Amante is now known as samante [03:28:51] samante is now known as Shane Amante [03:29:01] lars: rephrase question: not "bad idea" [03:29:18] again: protocl work in IETF? ~25 [03:29:31] [numbers non auth] [03:29:42] 1) 21: 2 [03:29:45] 2) 17:0 [03:30:07] resnick leaves the room [03:30:10] Toerless: 1st question had no distinction between conrete and abstract API [03:30:19] not work on protocol? [03:30:26] so 3) 15:2 [03:30:39] Yuri: how should last question be interpreted? [03:31:35] Martin: whatever requires coordination between hosts [03:31:52] this is the end of this BOF session [03:32:04] pk [jabber scribe] has set the subject to: NBS BoF over [03:32:27] pk [jabber scribe] leaves the room: Computer went to sleep [03:32:39] Hyong-Jong Paik leaves the room [03:32:52] Pete McCann leaves the room [03:33:03] Shane Amante leaves the room [03:33:14] tsavo_work@jabber.org/Meebo leaves the room [03:33:14] ꈲ leaves the room [03:33:24] Lars leaves the room [03:33:44] bhoeneis leaves the room [03:34:44] Colman Ho leaves the room [03:35:29] bortzmeyer leaves the room [03:39:14] alexmcmahon leaves the room [03:42:44] Dave Thaler leaves the room [03:45:14] Colman Ho joins the room [03:53:37] ꈲ joins the room [04:04:37] Zhen Tsao leaves the room [04:05:36] Javier Ubillos leaves the room [04:17:01] alexmcmahon joins the room [04:21:55] JLCjohn leaves the room [04:45:09] alexmcmahon leaves the room [04:45:23] Colman Ho leaves the room [04:49:36] aboba leaves the room [04:52:45] ꈲ leaves the room [04:55:25] Lars joins the room [05:02:32] samante joins the room [05:04:15] bhoeneis joins the room [05:06:23] samante leaves the room [05:08:34] Lars leaves the room [05:09:33] samante joins the room [05:34:04] martin.stiemerling leaves the room: Replaced by new connection [05:34:07] martin.stiemerling joins the room [05:37:07] martin.stiemerling leaves the room [05:37:08] ꈲ joins the room [05:54:49] nobobo joins the room [05:57:45] ꈲ leaves the room [06:06:03] ꈲ joins the room [06:09:36] bortzmeyer joins the room [06:10:27] bortzmeyer leaves the room [06:15:27] Jacky Yao (Health Yao) joins the room [06:49:46] ꈲ leaves the room [07:02:35] ꈲ joins the room [07:03:16] bhoeneis leaves the room [07:03:16] Jacky Yao (Health Yao) leaves the room [07:11:19] nobobo leaves the room [07:14:55] bhoeneis joins the room [07:16:17] nobobo joins the room [07:25:28] Jacky Yao (Health Yao) joins the room [07:25:53] Jacky Yao (Health Yao) leaves the room [07:31:39] nobobo leaves the room [08:01:15] samante leaves the room [08:12:16] bhoeneis leaves the room [08:29:46] ꈲ leaves the room [08:47:13] bhoeneis joins the room [08:55:30] ꈲ joins the room [09:34:46] bhoeneis leaves the room [10:38:31] pk [jabber scribe] joins the room [10:39:27] pk [jabber scribe] leaves the room [11:34:17] ꈲ leaves the room [12:00:37] ꈲ joins the room [22:58:20] ꈲ leaves the room