[04:30:23] Frederico Neves joins the room [04:31:39] PHB joins the room [04:44:44] yone joins the room [04:50:22] sukmoonlee joins the room [04:54:16] tlyu joins the room [04:55:11] matthijs joins the room [04:56:34] huguei joins the room [04:56:42] huguei leaves the room [04:57:05] Pete McCann joins the room [04:57:12] HugoSalgado joins the room [04:59:27] wouter joins the room [04:59:37] fdupont joins the room [05:01:18] fujiwara joins the room [05:02:28] wow, a distorted version of andrew [05:02:29] aleez joins the room [05:02:50] ogud joins the room [05:03:02] Suz joins the room [05:03:27] Isn't_that_nice joins the room [05:03:36] محسن joins the room [05:03:36] naptee joins the room [05:04:14] sm joins the room [05:04:24] jinmei joins the room [05:05:32] administrivia.... [05:05:33] ywang830 joins the room [05:05:53] mailing list to change to dnsext@ietforg, AD's decision [05:06:06] Is anyone getting sound? [05:06:16] Raman joins the room [05:06:22] Yes it's coming through for me but clipping. [05:06:27] would you please ask speaker to push a little back from the MIC [05:06:44] koji joins the room [05:06:47] RFC 5625(?) to be bis'ed to update mailing list name [05:07:00] wang.xincindy joins the room [05:07:12] mailing list archive to be moved, final landing place tbd [05:07:15] wang.xincindy leaves the room [05:07:42] wang.xincindy joins the room [05:07:59] subscription list also to be moved as of mid-October [05:08:00] zhang likun joins the room [05:08:13] wang.xincindy leaves the room [05:08:37] wang.xincindy joins the room [05:09:07] wang.xincindy leaves the room [05:09:28] wang.xincindy joins the room [05:09:29] rfc 5395 (bis) [05:09:58] wang.xincindy leaves the room [05:10:24] Peter Koch: pushback on including the mailing address in the bis doc given that there's some hope of someday shutting down the working group [05:10:29] wang.xincindy joins the room [05:10:43] suggests dedicated mailing address for RRtype questions [05:10:50] Splitting URI and uri-review actually worked reasonably well [05:10:59] wang.xincindy leaves the room [05:11:12] uri-review being for uri scheme reviews [05:11:13] AJS: the existing address is unusable, we're trying to fix that [05:11:23] Sam Weiler: why the hurry? [05:11:28] shinta joins the room [05:11:29] ray joins the room [05:11:30] Olafur: we lost control of the mailing lsit [05:11:33] wang.xincindy joins the room [05:11:38] AJS: any other comments on chairs' report? [05:12:00] Suzanne - the URL in the agenda pointing at your site doesn't work [05:12:01] Ted Hardie to talk about draft-faltstrom-uri [05:12:03] wang.xincindy leaves the room [05:12:06] Ray: i'm busy [05:12:11] :) [05:12:11] Chris Griffiths joins the room [05:12:50] Ted summarizes the draft: what's the point? [05:13:04] URI provides an additional abstraction layer [05:13:24] samante joins the room [05:13:54] matthijs leaves the room [05:13:58] wang.xincindy joins the room [05:14:05] ogud leaves the room [05:14:25] what's the issue? [05:14:35] first: right mechanism to bind serfvices to URIs where they're dleivered [05:14:35] matthijs joins the room [05:14:44] Wes joins the room [05:14:55] aleez leaves the room [05:15:06] limiting to IANA registered services may be limiting [05:15:31] ogud joins the room [05:15:46] Anyone else having periodic/frequent network disconnect issues ? [05:15:50] this re-uses srv/enumservices registries, may be a problem when different services use the same registered service definition [05:15:51] aleez joins the room [05:16:02] We need a second jabber scribe any volunteers ? [05:16:14] Suz is doing such a good job though ;_0 [05:16:23] unlike me and my shift key [05:16:34] jelte joins the room [05:16:42] Hi [05:16:50] (juggling scribes) [05:17:05] ogud leaves the room [05:17:08] sorry i'm late, network problems, i'll take over from now [05:17:38] we're at slide 4 [05:18:15] (alternatives) [05:18:50] ogud joins the room [05:19:08] ((btw if someone has something for the mic, please put [mic] in front of your comment) [05:19:30] Antoin joins the room [05:20:07] Jacky Yao (Health Yao) joins the room [05:20:10] Antoin has set the subject to: IETF-79 [05:20:33] .example.org, .example.cn, etc, if all of these all alternatives, and you only serve on .example.cz, you put in these records in all the other zones [05:20:50] but people will have to play along, and use this then [05:21:28] mic: I have spent a lot of time looking at the prefix problem. It is certainly important but I would not address it for URI alone, I would do it as a general approach that works for all DNS schemes that use prefixes [05:21:49] Suzanne Woolf: since i was asked, not really familiar, always interested in things that play well with other things [05:22:05] johani joins the room [05:22:14] mohsen: i still have problems understanding how this would be compatible with NAPTR [05:22:47] briand joins the room [05:22:52] .....and not sure how to make this play well with others, trusting apps folks to tell us [05:23:00] ted: document has some info on this [05:23:15] weiler joins the room [05:23:18] peter koch: i'm not sure about what problem this is trying to solve [05:24:25] peter: problem comes in with UNAPTR and SNAPTR [05:24:25] NAPTR does not work because it is based on regular expression matching which is a terrible hack. REs are tantamount to being Turing complete and that has no place in a discovery scheme. [05:24:43] peter: if the attempt is to let people put uri in a record, but not sure if that addresses the real problem, which is 'oh there is a service in some domain'. A domain is a part of a namespace, it is not something that hosts a service [05:25:06] peter: so what is the actual scenario here [05:25:21] (wes channeling PHB) [05:26:37] mohsen: i would go further than peter, i expect this document to be clearer, maybe 3 scenarios, one being that this brings something completely new, two: something that exists, but do it better, or 3: complements other schemes. If this wording is clear enough, we can catch the real use-cases [05:26:52] people are overdriving the mics [05:27:13] ted: given those comments, going back to peter's: there are times when you have an activity you want to perform with an address [05:27:15] jinjian joins the room [05:27:42] ted: so for instance you fetch an MX record which returns an address for a mail server. This is a generic system for such an action [05:28:07] immelman@jabber.org joins the room [05:28:46] there are two things going on: one is somethign that is currently there for a limited set of services, and the second is a simplification of the mechanism that was created uearlier for this purpose [05:29:26] hta joins the room [05:30:05] matthijs leaves the room [05:30:21] It is an expert review thing [05:30:26] peterkoch: so we do have a general type here, we learned that subtyping is not the way to go, what this provides is resolution context as a tool, it is a framework, and does not provide meaning. Could the chairs tell what the status is? Is this an expert review thing or should it go through the full review? [05:30:57] matthijs joins the room [05:31:14] andrew: as an action item: one of the things to do is to send specific text to the authors addressing the particular concerns from the mic here [05:31:56] ted: so one thing is a good point; are these registered services granular enough? are they too granular? [05:32:48] peter koch: re expert review; this is more complex than what the rfc on expert review specifies as what is suitable for expert review [05:33:19] peter koch: and reusing registry services migh be appealling, but it might clash, so it may be good to open a separate registry for this [05:33:31] Next item [05:34:00] Suzanne Woolf on Identical Resolution [05:34:02] Actually we don't have a proper registry at all for prefixes at the moment. [05:34:43] Current status slide [05:35:28] new charter makes this wg item [05:35:52] is this ready to go (suzanne herself thinks not) [05:36:04] Slide: Open Questions for Review [05:37:01] what, is she going to dig through the web server logs for that nonexistent URL? [05:37:05] PHB: was that a relay request? [05:37:23] ok, reminder to everyone: please prefix with [mic] if you want it relayed. [05:37:27] Slide: Open questions for review 2 [05:37:32] (otherwise I'll assume not) [05:38:19] Andrew: If this document is still on our charter in six months, we are not going to do it. (should go to IRTF) [05:38:57] slide: proposed next steps [05:39:57] proposed deadline: mid-january. [05:40:06] (for WGLC) [05:40:46] andrew: we are going to push hard on this [05:41:05] ogud leaves the room [05:41:32] [mic] and brian on jabber [05:41:34] Andrew: are there people who want to hang around after the meeting? [05:41:39] I don't want to work on this [05:41:46] One response (stephen), a reluctant Jelte [05:41:49] I think it is a rat hole [05:42:03] Then we again will only have DNS people again [05:42:19] vincent-afnic joins the room [05:42:30] Ted Hardie: you can ask apps review team to look at it [05:42:56] I think I'm already on the hook for it, so might as well volunteer. [05:43:23] briand leaves the room [05:43:28] briand: was that an offer to work on it or review? [05:43:30] missed the timing. [05:43:50] Ted Hardie: (sorry missed first thing he missed), second thing is that there is no example of a use-case that is not character-variant based [05:44:14] ogud joins the room [05:44:38] Harald Alvestrand: committed some things to the list yesterday [05:45:13] briand: I'll make sure Suz sees your note. [05:45:14] non-character-variant use case — TLD rename transition? [05:45:24] utoronto.ca and toronto.edu is his non-character variant use case [05:45:26] briand joins the room [05:45:32] ah, thanks [05:45:35] Next item [05:45:48] Simplified DNS Query under Ipv4/IPv6 [05:45:57] matthijs leaves the room [05:45:59] Slide 2; Index [05:46:26] matthijs joins the room [05:46:31] Slide 3: goalds [05:46:33] -d [05:46:43] Skipping to slide 6 :) [05:46:51] thanks briand [05:46:55] 7 [05:47:07] the audio is clipping really badly — almost unintelligible [05:47:20] slide 8 [05:47:28] it's bad in the room, too. [05:47:36] he's going through slides faster than i can keep up with, let alone content :/ [05:47:44] Antoin leaves the room: Replaced by new connection [05:47:58] slide 11 [05:48:01] slide 11: why two queries method? [05:48:50] most dns servers are ipv6 ready now, so two queries method is not practical anymore [05:48:54] slide 12 [05:49:21] slide 13 [05:49:41] this will become biggest problem for ipv6 deployment [05:49:46] Rickard Bellgrim joins the room [05:49:49] bad audio may be the mobile mic [05:50:03] slide 15 [05:50:17] to someone on the remote, are Olafur and Andrew sounding as bad as the presenters? [05:50:39] slide 17 [05:50:53] 18 [05:50:54] yes [05:50:58] (to Suz) [05:51:22] oh well....was a nice theory [05:51:25] 19 [05:51:37] :) [05:51:47] Olafur is asking the guy at the soundboard to fix [05:52:06] any better? [05:52:13] 20 [05:52:32] 21 [05:52:35] still clipping but not quite as painful [05:52:55] Mark Andrews: i suspect we need to do more with AAAA or v6 addressing anyway [05:53:10] i'm not noticing any differences, but I manage to follow [05:53:38] Mark: we need to do something, not sure if this gets us anywhere near where we need to be [05:53:40] olafur and andrew sound better though [05:53:49] Andrew: are you saying this is a problem and that we can solve this? [05:54:06] Mark: yes this is a problem, and we don't know when to stop doing A queries [05:55:06] Ondrej Sury: I would like to ask if you did measurements, and see some hard numbers on how inefficient the current system is [05:55:44] Ondrej: in kidns we have another problem that might be similar: multiple types in one query. Can we make this proposal more general? [05:55:49] Rickard Bellgrim leaves the room [05:56:03] Rickard Bellgrim joins the room [05:56:14] Peter koch: I see that suggestions like this come up over and over again, and either we fail to see the operational problem, or we fail to send the message over [05:56:45] Pter koch: I would also like to see statistics, noting that we have caching of both negative and positive responses [05:56:48] How about this, an EDNS extension that tells clients it supports an A5 record super-query. When the A5 query is presented, both A and AAAA records are returned. Job done. [05:57:57] Andrew: as action for the wg, there are two positions: one is we want multiple type queries. the other side could be we think this is not a DNS problem. If we go for the first we need drafts [05:58:36] mohsen: read the ??? draft presented at v6ops this morning, about the user problem with this [05:59:10] (Ah, now that the sound is better I find that is what is proposed...) [05:59:31] [mic] general problem of multiple things, would help get things through ignorant middle-boxes. I-D, agree [05:59:50] andrew: the next action is to figure out if the more general problem is or is not a problem, and we need list discussion and or drafts to decide that [06:00:02] briand: we've moved to a different topic at this point. [06:00:02] next item: frederico on dnsext-resimprove [06:00:06] write the list... [06:00:11] Slide 2: Motivation [06:00:11] <محسن> the draft in question is "Happy Eyeballs: Trending Towards Success with Dual-Stack Hosts", draft-wing-v6ops-happy-eyeballs-ipv6-01 // in v6ops, out of scope of dnsext [06:00:13] Sorry, mooted anyway [06:00:20] thank you [06:00:47] slide 3 [06:01:15] slide 4: resiliency [06:01:39] slide 5: robustness [06:02:04] Rickard Bellgrim leaves the room [06:02:15] slide6 [06:02:18] Rickard Bellgrim joins the room [06:02:38] i counted about 7 people [06:03:32] mohsen: just one comment. if it is easy to implement, just do it [06:03:39] shriek from the room [06:04:01] suzanne: maybe we should get some more people to read it first, then decide [06:04:16] peter koch: i think some of the ideas are extremely dangerous [06:04:32] peter koch: you are changing some fundamental ways of how dns works [06:05:00] andrew: so that sounds like it is in charter, and you think it's a bad idea [06:05:39] peter koch: the risk i see here is the accumulative swarm effect [06:06:00] frederico: actually some of these proposals are to prevent swarms [06:06:30] andrew: we will pose these questions to the list after items from today are addressed in a new version [06:06:50] next: steve crocker on dnssec history wiki [06:07:07] slide 2: The first 20 years [06:08:13] this has all taken a very very long time [06:08:19] matthijs leaves the room [06:08:21] Rickard Bellgrim leaves the room [06:08:35] Rickard Bellgrim joins the room [06:08:36] slide 4 [06:09:07] matthijs joins the room [06:09:23] there is a lot of history that is not captured and documented, and might be lost [06:09:58] slide 6 [06:10:16] https://wiki.tools.isoc.org/DNSSEC_History_Project [06:10:49] let the people who were there fill in the history [06:10:50] Isn't_that_nice leaves the room [06:11:23] Invitiation to anybody who had anything to do with DNSSEC to weigh in, the purpose is to gather lost and lots of raw material [06:11:27] slide 7 [06:12:11] slide 9 [06:12:14] 10 [06:12:20] same url again [06:13:06] Alissa Cooper joins the room [06:13:33] net time will be a show of hands to find out how many people have contributed [06:13:55] Olafur: this concludes the formal part of the meeting, we will now open up for feedback to editors of documents [06:14:27] Andrew: can someone collect stuff on DNAME draft? [06:15:09] Rickard Bellgrim leaves the room [06:15:12] matthijs leaves the room [06:15:21] matthijs joins the room [06:15:23] Rickard Bellgrim joins the room [06:15:47] Sam: couple of words on what feedback we need [06:15:54] hta leaves the room [06:16:02] ray leaves the room [06:16:11] Sam: 1. rollover and die. We have no text on why the SHOULD on BAD cache [06:16:42] Sam: 2. cd bit. What to do when CD bit is not set? should always? should never? [06:17:13] if anyone remotely has anything to comment i can relay [06:18:16] shinta leaves the room [06:18:32] Alissa Cooper leaves the room [06:18:39] PHB leaves the room [06:18:43] sukmoonlee leaves the room [06:18:45] HugoSalgado leaves the room [06:18:46] Will anyone be available to relay to/from Suzanne's group? [06:18:48] Bernie joins the room [06:18:52] jinmei leaves the room [06:19:01] matthijs leaves the room [06:19:18] aleez leaves the room [06:19:18] white noise on mic... [06:19:28] immelman@jabber.org leaves the room [06:19:30] Chris Griffiths leaves the room [06:19:41] sm leaves the room [06:20:04] koji leaves the room [06:20:25] Pete McCann leaves the room [06:20:32] fdupont leaves the room [06:20:50] yone leaves the room [06:21:08] Use cases: big-company.example.com and product.example.com [06:21:27] Rickard Bellgrim leaves the room [06:21:38] jinjian leaves the room [06:21:42] Rickard Bellgrim joins the room [06:24:24] hta joins the room [06:25:02] white noise is probably from the 15 discussions going on now :) [06:25:44] It's all good now... I can make out the aliasing conversation, too, bonus [06:25:56] heh [06:26:21] Rickard Bellgrim leaves the room [06:26:36] Bernie leaves the room [06:26:43] Bernie joins the room [06:27:34] Pass along to Mark A and Suz, please? [06:28:10] If equivalence is done on server side ($INCLUDE and friends), can check and prevent problems, vs *NAME client-side impossible [06:28:51] briand: will do [06:29:00] Thanks much. [06:30:01] i showed my jabber to mark but he didn't really get what you meant [06:30:52] Okay. I'll try to take it offline with him. [06:31:06] Bernie leaves the room [06:31:17] wes is going to try suz :) [06:32:15] Suz has seen it. [06:32:25] Kind of difficult to participate in a conversation 20,000 km away or thereabouts :-) [06:32:59] samante leaves the room [06:33:13] yep :-/ [06:35:06] Jacky Yao (Health Yao) leaves the room [06:35:38] Jacky Yao (Health Yao) joins the room [06:36:13] wang.xincindy leaves the room [06:36:15] naptee leaves the room [06:38:01] wang.xincindy joins the room [06:38:42] あっと joins the room [06:38:47] あっと leaves the room [06:38:50] wang.xincindy leaves the room [06:38:58] briand leaves the room [06:41:34] Wes leaves the room [06:42:05] Raman leaves the room [06:42:36] ogud leaves the room [06:48:15] ogud joins the room [06:50:01] weiler leaves the room [06:54:26] wouter leaves the room [06:55:06] ogud leaves the room [06:55:55] Bernie joins the room [06:58:42] Suz leaves the room [06:59:02] hta leaves the room [07:01:44] jelte leaves the room [07:04:40] محسن leaves the room: Computer went to sleep [07:04:43] samante joins the room [07:06:59] Frederico Neves leaves the room [07:07:31] samante leaves the room [07:07:36] Bernie leaves the room [07:07:52] samante joins the room [07:08:11] johani leaves the room [07:09:06] Jacky Yao (Health Yao) leaves the room [07:30:25] Chris Griffiths joins the room [07:30:53] Frederico Neves joins the room [07:32:53] tlyu leaves the room [07:37:49] Frederico Neves leaves the room [07:44:06] Chris Griffiths leaves the room [07:53:10] Frederico Neves joins the room [07:55:35] vincent-afnic leaves the room [08:00:38] Frederico Neves leaves the room [08:05:20] samante leaves the room [08:06:41] ywang830 leaves the room [08:07:03] hta joins the room [08:10:51] hta leaves the room [08:11:56] Chris Griffiths joins the room [08:13:16] zhang likun leaves the room [08:19:41] Frederico Neves joins the room [08:19:55] Frederico Neves leaves the room [08:27:10] ogud joins the room [08:27:21] ogud leaves the room [08:35:53] johani joins the room [08:39:56] hta joins the room [08:40:40] hta leaves the room [08:40:44] hta joins the room [08:42:48] hta leaves the room [08:51:29] fujiwara leaves the room [08:54:26] Chris Griffiths leaves the room [11:09:23] johani leaves the room [13:15:02] Chris Griffiths joins the room [13:15:07] Chris Griffiths leaves the room [15:43:45] johani joins the room [16:22:47] johani leaves the room [18:20:15] johani joins the room [19:43:22] johani leaves the room