[02:34:08] fabinader leaves the room [03:21:59] zdshelby joins the room [03:26:32] zdshelby leaves the room [03:31:13] ꈲ joins the room [03:38:09] zdshelby joins the room [03:38:16] Test [03:40:44] <ꈲ> tseT [03:46:17] kerlyn joins the room [03:46:17] kerlyn leaves the room [03:46:25] kerlyn joins the room [03:46:36] hello zach? [03:47:48] Hi [03:50:51] matt.anderson joins the room [03:52:42] Barry Leiba joins the room [03:54:21] tonyhansen joins the room [03:55:41] Simon Perreault joins the room [03:55:55] Lisa joins the room [03:56:40] Cullen Jennings joins the room [03:57:14] Ed J. joins the room [03:57:37] Ed J. leaves the room [03:57:43] Joe Hildebrand joins the room [03:58:04] AdrianFarrel joins the room [03:58:25] Andrew McGregor joins the room [03:58:27] Simon Perreault leaves the room [03:58:40] AdrianFarrel leaves the room [03:59:19] resnick joins the room [03:59:56] fabinader joins the room [04:00:08] shamus joins the room [04:00:19] Robert Cragie joins the room [04:00:24] Ed J. joins the room [04:00:28] Remote people who want me to channel them at the microphone, please prefix your message with "mic: ". Thanks. [04:01:08] Simon Perreault joins the room [04:01:14] kanda joins the room [04:01:39] yuichi igarashi joins the room [04:02:00] fabinader leaves the room [04:02:08] fabinader joins the room [04:02:12] Bernie joins the room [04:02:34] nr2805 joins the room [04:02:55] Ulrich Herberg joins the room [04:02:56] rababy joins the room [04:03:16] stpeter joins the room [04:04:40] stpeter leaves the room [04:04:46] toru.asahina joins the room [04:05:44] stpeter joins the room [04:06:13] Bruno STEVANT joins the room [04:08:21] AdrianFarrel joins the room [04:09:35] Lars joins the room [04:09:49] magnus joins the room [04:10:35] ohnolab.t-takashima joins the room [04:11:09] stpeter leaves the room: Disconnected: conflict [04:11:59] sftcd joins the room [04:12:38] mic: are we assuming "hidden nodes"? [04:13:33] stpeter joins the room [04:14:32] MANAMANA joins the room [04:14:58] momose joins the room [04:15:32] "No real IP multicast" is an odd thing to say? Ah, he clarifies... IP multicast *an* exist but may be "too expensive" [04:15:44] s/an/can/ [04:16:05] In terms of power, maybe [04:18:28] "Smart Energy(Smart Grid)" seems too abstract. [04:18:34] These sort of networks always have these cross-layer issues [04:19:10] Smart Energy 2.0 is a specific standardization effort, the charter references that. [04:19:51] Ah, OK. [04:20:16] tom herbst joins the room [04:21:04] the point of the battery devices for smart energy is the refrigerator magnet type displays, not meters or lightswitches [04:21:27] mwtr joins the room [04:21:30] mwtr leaves the room [04:21:31] Or remote sensor switches; think outdoor security lighting [04:21:50] Right, and thermostats, temp sensors, HVAC, gas and water meters etc. [04:22:48] mwtr joins the room [04:22:52] mwtr leaves the room [04:23:10] Constrained Operations Protocol for Applications - COPA, COPA-CABANA [04:23:23] lol [04:23:28] lqtm [04:23:59] dlohman joins the room [04:25:17] that was Dave Oran at the mic [04:25:28] ("are you going to describe the trust model") [04:25:39] dlohman leaves the room [04:26:12] markus.isomaki joins the room [04:26:28] If there's one thing that concerns me in the charter, it's having CoGII some arbitrary number of hops away [04:26:30] dlohman joins the room [04:26:46] I don't understand that requirement [04:26:57] That would be OK, so long as it's on the default route (presuming there is one) [04:27:23] It is IP, so there is no requirement for where a CoGII is placed. [04:27:43] but seems to preclude 802.1x, for example [04:28:12] I'm not convinced that all of the protocols he mentioned for CoGII can be reduced to REST. For example, SNMP traps. [04:28:22] mic: is it possible to enumerate all nodes? [04:28:33] This seems to need a 'binding' process, a la bluetooth or some hobby RC systems, so security associations are long-lived. [04:28:49] .1x doesn't fit that model [04:29:05] Are those resources (wrt. REST) on the constrained nodes themselves or do the constrained nodes just act as clients toward some other servers? [04:30:10] IOW: bias toward SLP [04:30:56] fujiwara joins the room [04:31:59] is that stuart cheshire? [04:32:02] yes [04:32:27] Describing a couple of gloriously hideous hacks... [04:34:10] The resources are on constrained nodes. [04:34:21] Moore's law will save us ;-) [04:34:22] Or at least can be [04:35:18] Unfortunately, we can't revoke thermodynamics or Shannon capacity, so there will still be energy constraints on these things. [04:36:07] Ted Hardie at mic [04:36:27] For small values of "scales". [04:37:07] i can't imagine a device that is small enough that it can't do all of HTTP, but large enough that it can parse ASN.1 and do biginteger operations. [04:37:59] They exist, they do the bigints in hardware, and sometimes even ASN parsing [04:38:28] [Peter van der Stok about to present] [04:38:32] just put HTTP in hardware, then. [04:38:36] AdrianFarrel leaves the room [04:38:38] Right, there are existing application protocols - we'd like to make sure they can migrate to IP [04:40:05] Cullen Jennings leaves the room [04:41:23] Cullen Jennings joins the room [04:41:29] yuichi igarashi leaves the room [04:43:23] yuichi igarashi joins the room [04:47:20] rababy leaves the room: Computer went to sleep [04:49:16] [Dave Oran at mic] [04:50:10] rababy joins the room [04:51:31] [Don Sturek about to present] [04:54:19] That's hardly a clarifying question [04:55:10] I suspect he questions their security expertise [04:55:31] I'd like to see someone announce a rhethorical question, see how that goes... [04:56:13] So please, approve our RFC today [04:59:30] [Pete Resnick] [04:59:31] Pete Resnick speaking [05:00:12] RoHC? [05:00:21] markus.isomaki leaves the room [05:00:57] Sigcomp[ [05:01:14] AdrianFarrel joins the room [05:01:17] alexey.melnikov joins the room [05:01:19] s-takuo joins the room [05:01:29] [Lars Eggert] [05:02:37] 200ms over radios isn't going to work end to end [05:02:59] [Dave Oran] [05:03:00] Dave Oran at Mic [05:03:04] right, hence my concern about putting CoGII far away [05:03:06] That is an example for local communications in building automation and SE. They are not talking across the Internet. [05:03:54] The IMAP RFC Pete was talking about is RFC 4978 - IMAP COMPRESS command [05:03:59] even in se, you get a few hops involved and there are issues [05:07:02] [Zach Shelby] [05:08:06] [Dave Oran] [05:08:33] Who was before Zach? [05:08:48] mumble mumble from UC Berkeley I think? [05:09:00] David Culler [05:09:56] Link layer security is used like WPA in WiFi [05:10:02] [Zach Shelby presenting now] [05:10:25] yes, but this isn't really the place to have that conversation. [05:11:53] gregdavies joins the room [05:18:32] gregdavies leaves the room [05:19:01] ohnolab.t-takashima leaves the room: Replaced by new connection [05:19:05] gregdavies joins the room [05:19:10] Nicolas joins the room [05:20:30] Lars leaves the room [05:24:02] redaka joins the room [05:24:52] Thinking about compression again: TLS also has Deflate compression, this might be more applicable to this work [05:25:21] TLS over UDP? [05:26:17] Maybe in DTLS, I haven't looked at DTLS in details [05:26:24] Is the REST API on the wire or in the box? [05:26:40] deflate will de difficult to make work in DTLS, since i don't think it can recover from lost packets. [05:26:41] [Pete Resnick] [05:27:10] [didn't get name of speaker] [05:27:26] I get the sense the goal is to make for easy translation from CoAP to REST/HTTP?TCP [05:27:34] Speaker: Peny Yang, Hitachi [05:28:06] Carsten mentioned "HTTP semantics" for CoAP [05:28:28] Plus eventing [05:28:33] right [05:28:59] robert: that's a pretty key point [05:29:24] which? notification semantics? [05:29:31] yes. [05:29:35] disappointing that the assumption is that the WG is assuming this needs to be strange w/o looking at doing something more normal first [05:30:11] Good point [05:30:39] tom herbst: agreed [05:31:19] +1 [05:31:20] given the time constraints of the smart energy world, does it make sense for this working group to spend a lot of time working on this when other standards orgs in the smart energy world are also working on these things? [05:31:43] yeah, the timeline seems crazily optimistic [05:31:47] I suspect they need the IETF blessing [05:31:56] Well, this is likely to be useful anyway too. [05:31:58] matt.anderson: good question, are you in the room to say that at the mic, or shall I channel? [05:32:06] they don't need the ietf blessing so much as the ietf clue (real or perceived) [05:32:07] channel please [05:32:11] matt.anderson: ok [05:32:36] most of the other groups are tied to a mac/phy and have little IP experience [05:32:59] Bernie leaves the room: Replaced by new connection [05:33:03] understand and agree...but they also probably aren't willing to wait [05:33:04] Anyone know how much effort has gone into analysis of existing IETF Apps protocols? [05:33:18] we know that building automation also wants to go to native IP [05:33:27] Adrian? Lots, but not sure how to formalize that. [05:33:28] I'm having trouble buying that we need to constrain all devices because some constrained devices exist. We should at least define rational constraints and not 10kBytes ram, for example [05:33:47] building automation isn't chasing u.s. stimulus money :-) [05:33:52] Tom: that's not the intent - gatewaying to HTTP is specifically to allow communication with unconstrained devices. [05:34:11] it would be nice if "gatewaying" was just routing [05:34:14] Lisa - write an I-D! Doesn't have to make it to RFC [05:34:42] but don't forget that CoAP devices should be able to interoperate w/o CoGII [05:34:48] tom - that same problem has been fought in roll [05:34:54] redaka leaves the room [05:35:01] Will CoAP run on the "normal" Internet as well? [05:35:03] deciding an acceptable level of constrained [05:35:13] exactly [05:35:34] Lars joins the room [05:35:34] toru.asahina leaves the room [05:35:37] As Einstein said "as simple as possible, but no simpler." [05:35:40] Adrian: Yes, they think it would run over normal intenernet [05:36:27] who's speaking? [05:36:36] ted hardie [05:37:06] Adrian: That's *the* question I think - it would have to be applicable widely as well as in constrained environments. That's where the argument for anything too different gets tricky [05:37:49] PINK SQUIRREL TSHIRT! [05:37:52] if it doesn't work over the normal internet it becomes a lot less useful [05:37:58] (want) [05:38:13] Is he saying REST/HTTP/TCP and call it a day? [05:38:47] He's saying, do that first as a matter of priority, and then think about something else if it proves necessary by then [05:39:12] [Stuart Cheshire] [05:39:34] yuichi igarashi leaves the room: Replaced by new connection [05:39:35] yuichi igarashi joins the room [05:41:07] http://tools.ietf.org/html/draft-cheshire-dnsext-dns-sd-05 [05:41:19] Ulrich Herberg leaves the room [05:42:11] It sure sounds like one [05:43:21] [Rene Struik] [05:44:10] don't misunderstand me - i think the ietf probably would do the best job of addressing this...i just wonder about the time it will take [05:44:49] Lars leaves the room [05:45:01] I think considerable momentum came in with this... so that might help [05:45:23] [Peter St Andre] [05:45:40] so it seems strange that there's no draft-foo-coap protocol spec but yet they want to move this quickly; such a timeline would normally imply that someone had code working already, or am I missing something? [05:45:46] FWIW, Roll has been similarly driven on milestones and has sort of gotten close-ish. [05:46:10] [Patil] [05:46:40] I hope this doesn't become a rubber stamp of SE at the expense of other problems [05:47:04] Ulrich Herberg joins the room [05:47:05] MANAMANA leaves the room [05:47:07] [Lisa packs up knitting - things must be getting serious] [05:47:24] we should at least make sure it covers BA as well [05:47:35] [David Culler}] [05:47:41] Ulrich Herberg leaves the room [05:47:55] alexey.melnikov leaves the room: Computer went to sleep [05:48:00] Bernie joins the room [05:48:06] shamus leaves the room [05:48:28] Ed J. leaves the room [05:48:48] But CoAP is about defining the lower bound [05:49:01] or least upper bound [05:49:21] Hmm, does http://www.ietf.org/id/draft-ietf-hip-hiccups-00.txt constitute such a protocol? (HIP itself can be made small and light enough with the right profile, I'm sure) [05:49:25] [Lars Eggert] [05:49:51] toru.asahina joins the room [05:50:24] [Lisa Dusseault] [05:50:48] gregdavies leaves the room [05:52:09] Bruno STEVANT leaves the room [05:52:20] http://tools.ietf.org/html/draft-cheshire-dnsext-dns-sd-05 [05:55:24] fabinader leaves the room [05:57:42] Bernie leaves the room [05:58:15] AdrianFarrel leaves the room [05:59:37] Barry Leiba leaves the room [06:01:05] mic: so are we throwing CoGII under the bus? [06:01:40] I don't hear that, is the channeler around? [06:01:44] yuichi igarashi leaves the room: Computer went to sleep [06:03:11] Simon Perreault leaves the room [06:03:14] http binding not, everything else yes [06:03:21] Lisa leaves the room [06:03:22] Andrew McGregor leaves the room [06:03:23] Joe Hildebrand leaves the room: Disconnected. [06:03:25] resnick leaves the room [06:03:28] kerlyn leaves the room [06:03:30] dlohman leaves the room [06:03:33] matt.anderson leaves the room [06:03:34] zdshelby leaves the room [06:03:41] Robert Cragie leaves the room [06:03:52] rababy leaves the room [06:04:08] nr2805 leaves the room [06:04:26] Nicolas leaves the room [06:05:06] stpeter leaves the room [06:05:23] tonyhansen leaves the room [06:06:50] Cullen Jennings leaves the room [06:07:20] magnus leaves the room [06:07:25] ꈲ leaves the room [06:08:15] dlohman joins the room [06:08:31] Bruno STEVANT joins the room [06:09:07] dlohman leaves the room [06:10:02] momose leaves the room [06:11:06] s-takuo leaves the room: Replaced by new connection [06:12:17] tom herbst leaves the room [06:13:51] toru.asahina leaves the room [06:17:01] fujiwara leaves the room [06:20:17] Simon Perreault joins the room [06:21:43] Simon Perreault leaves the room [06:23:19] Lars joins the room [06:23:34] Lars leaves the room [06:23:39] ꈲ joins the room [06:23:52] fabinader joins the room [07:03:06] ꈲ leaves the room [07:08:21] Bruno STEVANT leaves the room [07:20:21] sftcd leaves the room [07:23:02] Lisa joins the room [07:26:56] Lisa leaves the room [07:57:11] alexey.melnikov joins the room [08:02:44] Joe Hildebrand joins the room [08:04:15] Joe Hildebrand leaves the room [08:14:48] fabinader leaves the room [08:16:22] alexey.melnikov leaves the room: Computer went to sleep [08:33:03] Andrew McGregor joins the room [08:33:32] Andrew McGregor leaves the room [08:38:37] fabinader joins the room [10:51:04] fabinader leaves the room [11:13:59] kanda leaves the room [11:31:50] MANAMANA joins the room [11:32:06] MANAMANA leaves the room [22:45:36] Juergen Schoenwaelder joins the room [22:47:12] Juergen Schoenwaelder leaves the room [22:52:32] alexey.melnikov joins the room