IETF
homenet@jabber.ietf.org
Tuesday, March 24, 2015< ^ >
DanYork has set the subject to: Home Networking WG at IETF 91
Room Configuration
Room Occupants

GMT+0
[00:59:21] tadanori.teruya@gmail.com joins the room
[01:00:16] tadanori.teruya@gmail.com leaves the room
[04:02:01] tob cast joins the room
[04:02:14] Simon Romanone joins the room
[04:02:20] John Doe joins the room
[04:02:33] ajeje brazorf joins the room
[04:07:20] John Doe leaves the room
[04:17:22] Simon Romanone leaves the room
[04:18:13] Simoncione Romano joins the room
[04:18:30] ajeje brazorf leaves the room
[04:19:35] cris tosanto joins the room
[04:20:34] cris tosanto leaves the room
[04:23:48] ciccio ciccione joins the room
[04:25:19] tob cast leaves the room
[04:25:31] Simoncione Romano leaves the room
[04:26:12] Simoncino Romanino joins the room
[04:26:12] tobia janus joins the room
[04:27:47] Simoncino Romanino leaves the room
[04:28:06] Simo Roman joins the room
[04:28:15] ciccio ciccione leaves the room
[04:30:46] Simo Roman leaves the room
[04:36:57] tobia janus leaves the room
[04:37:27] tob castlll joins the room
[04:37:35] Simon Romano joins the room
[04:38:23] tob castlll leaves the room
[04:38:29] Simon Romano leaves the room
[04:49:06] Simo ne joins the room
[04:52:50] franco nero joins the room
[04:53:42] franco nero leaves the room
[04:53:49] Simo ne leaves the room
[04:55:26] franco nero joins the room
[04:55:31] homm nett joins the room
[04:55:50] Sim on joins the room
[04:57:30] homm nett leaves the room
[04:57:30] Sim on leaves the room
[04:57:59] Sim one joins the room
[04:58:30] cuoll e cazz joins the room
[05:02:01] cuoll e cazz leaves the room
[05:02:19] no player joins the room
[05:04:20] no player leaves the room
[05:04:26] franco nero leaves the room
[05:04:29] Fire fox joins the room
[05:05:03] tob firefox joins the room
[05:05:16] ora mi senti joins the room
[05:06:01] ora mi senti leaves the room
[05:06:17] cris tosanto joins the room
[05:24:49] <cris tosanto> hummmmm
[05:31:27] Fire fox leaves the room
[05:32:54] Sim one leaves the room
[05:33:23] cris tosanto leaves the room
[05:34:52] tob firefox leaves the room
[13:10:37] jch joins the room
[13:55:36] Ted Lemon joins the room
[13:55:48] Wes George joins the room
[13:56:14] Jason Weil joins the room
[13:58:29] marka joins the room
[13:59:02] Juliusz Chroboczek joins the room
[13:59:34] Ralf Weber joins the room
[14:00:14] Brian Carpenter joins the room
[14:00:34] jch leaves the room
[14:00:36] Markus Stenberg joins the room
[14:00:53] <Brian Carpenter> Yo! Meeting materials link is defective - slides are all loaded at https://www.ietf.org/proceedings/92/homenet.html
[14:01:27] <Wes George> ah thanks brian
[14:01:35] <Wes George> are you or anyone else remote?
[14:01:50] <Ted Lemon> Juliusz is remote.   :(
[14:01:50] <Brian Carpenter> I'm in the room
[14:02:01] <Wes George> audio check?
[14:02:10] <Wes George> currently on WG UPdate slide of https://www.ietf.org/proceedings/92/slides/slides-92-homenet-0.pdf
[14:02:31] <Juliusz Chroboczek> Hi Ted, hi Brian, hello all.
[14:02:55] <Ted Lemon> Hi Juliusz!
[14:02:59] <Wes George> Michael Kloberdans now presenting https://www.ietf.org/proceedings/92/slides/slides-92-homenet-1.pdf
[14:03:14] leslie joins the room
[14:03:15] <Wes George> slide 2
[14:03:33] naoki joins the room
[14:03:40] Meetecho joins the room
[14:03:41] <Wes George> slide 3
[14:03:51] naoki leaves the room
[14:03:56] Ole Troan joins the room
[14:04:10] <Wes George> audio ok Juliusz?
[14:04:14] <Wes George> slide 4
[14:05:00] <Wes George> slide 5
[14:05:12] <Wes George> slide 6
[14:05:26] <Juliusz Chroboczek> Nope.
[14:05:40] <Wes George> no audio? or too low, or what
[14:05:44] <Juliusz Chroboczek> Probably problem on my side.
[14:05:53] Matthieu Boutier joins the room
[14:05:57] <Wes George> have you tried the mp3 stream vs meetecho?
[14:05:58] <Markus Stenberg> did you try both normal mp3 and meetecho?
[14:06:02] <Meetecho> working on that
[14:06:13] <Juliusz Chroboczek> Solved with by using a Mac :-/
[14:06:20] Andrew Yourtchenko_m joins the room
[14:07:09] <Wes George> erik kline - this is useless
[14:07:17] <Wes George> equivalent of evil/not evil bit - no trust
[14:07:30] <Wes George> i'd turn it off
[14:07:43] <Wes George> security issues
[14:07:53] <Wes George> john b - we're being asked for this
[14:08:01] jlatour joins the room
[14:08:21] <Wes George> ted : who "us" ?
[14:08:29] <Wes George> jjb - other mfr of retail CPE routers are asking
[14:08:46] <Wes George> ted: they want comcast to send CER ID?
[14:08:51] <Wes George> jjb - something that smells like it
[14:09:16] <Wes George> michael - lightweight alternative of homenet
[14:09:53] <Wes George> margaret wasserman
[14:10:32] <Wes George> jjb
[14:10:56] Peter Koch joins the room
[14:10:58] <Wes George> fatal flaw- no notion of >1 edge router
[14:11:05] Ida Leung joins the room
[14:11:05] <Wes George> kerry lynn
[14:11:33] <Wes George> barbara stark
[14:11:48] <Peter Koch> where are the slides? <https://datatracker.ietf.org/meeting/92/materials.html#homenet> is almost empty
[14:12:00] Ida Leung leaves the room
[14:12:04] <Juliusz Chroboczek> https://www.ietf.org/proceedings/92/homenet.html
[14:12:10] Victor  Kuarsingh joins the room
[14:12:18] ida leung joins the room
[14:12:32] <Wes George> hi all
[14:12:46] <Wes George> slides are posted here: https://www.ietf.org/proceedings/92/homenet.html
[14:12:51] <Wes George> not sure why they aren't showing up on materials site
[14:12:56] <Peter Koch> tnx, the nice thing about URLs is that there are so many ...
[14:13:00] <Wes George> peter lothberg at mic
[14:13:32] Mark Baker joins the room
[14:13:32] shinta joins the room
[14:13:53] <Wes George> hum for applicability for CER in homenet with HNCP on table
[14:14:03] ozindfw joins the room
[14:14:13] <Wes George> lorenzo
[14:14:54] <Wes George> barbara stark
[14:15:26] <Wes George> margaret wasserman
[14:15:46] <Wes George> any hums here?
[14:15:59] <Wes George> second hum: useful in other ietf groups?
[14:16:11] <Wes George> ted lemon
[14:17:03] <Wes George> daniel migault now presenting https://www.ietf.org/proceedings/92/slides/slides-92-homenet-2.pdf
[14:17:06] <Wes George> slide 1
[14:17:10] Brian Haberman joins the room
[14:18:29] Mikael Abrahamsson joins the room
[14:19:22] <Wes George> slide 2
[14:19:25] <Wes George> slide 3
[14:20:09] <Wes George> slide 4
[14:21:14] <Wes George> slide 5
[14:21:45] Mark Baker leaves the room
[14:22:23] <Wes George> slide 6
[14:23:57] <Wes George> ray
[14:24:10] Brian Haberman leaves the room
[14:25:00] <Wes George> mark andrews
[14:26:14] Mark Baker joins the room
[14:26:23] Paolo Saviano joins the room
[14:26:42] <Wes George> if there is any comments from jabber, please precede them with mic:
[14:27:30] <Juliusz Chroboczek> mic: does this version address the case where there are multiple CPEs on the Homenet?
[14:27:30] <Wes George> jim geddes
[14:27:35] <Ted Lemon> Gettys
[14:28:00] <Mikael Abrahamsson> david lamparter
[14:28:01] <Brian Carpenter> David Lampeter
[14:28:01] <Wes George> ack
[14:28:06] Paolo Saviano leaves the room
[14:28:21] Miles McCredie joins the room
[14:28:59] each joins the room
[14:29:03] Franck Martin joins the room
[14:29:15] <Juliusz Chroboczek> Yes.
[14:29:19] <Juliusz Chroboczek> ack.
[14:29:22] <Wes George> thanks
[14:29:25] David Lamparter joins the room
[14:29:29] <Wes George> mark andrews
[14:29:40] <Mikael Abrahamsson> markus stenberg
[14:29:53] <Wes George> thanks mikael - I was going to have to ask
[14:30:30] <Mikael Abrahamsson> wes, I sit pretty well to read speaker names, so if you want me to do that when I'm not in line, I'll volunteer for that specific task.
[14:31:04] <Ole Troan> Mikael, appreciated from note taker pov as well
[14:31:14] <Mikael Abrahamsson> oki, will do!
[14:31:22] <Wes George> slide 9, btw
[14:31:31] Peter DeVries joins the room
[14:33:03] <Mikael Abrahamsson> mark andrews
[14:34:18] <Wes George> slide 10
[14:34:23] <Wes George> slide 11
[14:34:43] <Ted Lemon> Peter Koch
[14:34:47] <Mikael Abrahamsson> ok
[14:35:06] leslie leaves the room
[14:35:42] <Wes George> ray belis
[14:36:41] <Mikael Abrahamsson> do we only have one blue sheet? It never passed by front left in the room
[14:37:22] <David Lamparter> Mikael: it did pass by behind you :o
[14:37:33] <Ole Troan> I didn't see it either.
[14:37:33] <Wes George> now on https://www.ietf.org/proceedings/92/slides/slides-92-homenet-3.pdf
[14:37:38] <Ted Lemon> It did actually—it just passed before you sat down.   That's why Ray was asking people to stop passing it until the meeting started. :)
[14:38:06] <Wes George> now onto Markus Stenberg presenting https://www.ietf.org/proceedings/92/slides/slides-92-homenet-4.pdf
[14:38:18] <Peter Koch> clarifying question: the slides mention draft-mglt-homenet-front-end-naming-delegation-05.txt, but there is draft-ietf-homenet-front-end-naming-delegation-01.txt?
[14:39:01] <Wes George> slide 2
[14:39:09] Markus Stenberg leaves the room
[14:39:58] <Wes George> slide 3
[14:40:39] <Wes George> slide 4
[14:41:19] <Wes George> slide 5
[14:41:52] <Wes George> questions
[14:42:16] <Wes George> mikael abrahamsson
[14:42:39] <Juliusz Chroboczek> because of the whole what?
[14:43:10] <Ted Lemon> I don't remember the name, but there was a petition from a bunch of universities asking IETF to fix the multiple-subnet service discovery problem.
[14:43:16] <Wes George> stuart cheshire
[14:44:30] <Wes George> steven barth now presenting https://www.ietf.org/proceedings/92/slides/slides-92-homenet-5.pdf
[14:44:32] <Wes George> slide 2
[14:45:25] Markus Stenberg joins the room
[14:45:38] Lorenzo Miniero joins the room
[14:46:45] <Wes George> slide hncp-prefix assignment
[14:47:16] <Wes George> slide latest updates and todos
[14:47:29] leslie joins the room
[14:47:50] Peter DeVries leaves the room: Replaced by new connection
[14:47:52] Peter DeVries joins the room
[14:49:59] leslie leaves the room
[14:50:20] Simon Romano joins the room
[14:50:51] <Wes George> mikael abrahamsson
[14:52:32] Simon Romano leaves the room
[14:52:49] <Wes George> slide reference implementation updates
[14:53:41] <Mikael Abrahamsson> markus stenberg
[14:53:52] <Mikael Abrahamsson> david lamparter
[14:54:01] Lorenzo Miniero leaves the room
[14:54:59] Paul Selkirk joins the room
[14:55:56] vi nn joins the room
[14:56:10] vi nn leaves the room
[14:56:44] <Wes George> slide ref implementation updates #2
[14:57:02] suprita s joins the room
[14:57:45] LeeHoward@lightwitch.org joins the room
[14:58:00] m.behringer joins the room
[14:58:12] <Mikael Abrahamsson> brian carpenter
[15:00:47] <Mikael Abrahamsson> markus stenberg
[15:00:54] <Mikael Abrahamsson> richard
[15:01:06] C Hopps joins the room
[15:01:06] <Mikael Abrahamsson> michael richardson next, sorry.
[15:01:46] suprita s leaves the room
[15:04:22] leslie joins the room
[15:04:44] leslie leaves the room
[15:05:00] <Wes George> Lorenzi Collitti
[15:05:00] m.behringer leaves the room
[15:05:08] <Wes George> Brian Carpenter
[15:06:31] <Juliusz Chroboczek> Who said "I'm going to think about it and send a message to the list"?
[15:06:57] <Markus Stenberg> Brian C.
[15:07:02] <Wes George> mikael Abrahamsson speaking curently
[15:07:06] <Wes George> lorenzo
[15:07:40] <Wes George> mikael
[15:07:53] <Wes George> alex petrescu
[15:09:29] Victor  Kuarsingh leaves the room
[15:09:34] <Ole Troan> *minute taker overload*
[15:09:42] <Wes George> barbara stark
[15:10:42] <Juliusz Chroboczek> Barbara has a point, though.
[15:11:05] <Wes George> Margaret Wasserman presenting https://www.ietf.org/proceedings/92/slides/slides-92-homenet-6.pdf
[15:11:30] <Mikael Abrahamsson> it's hard to delegate multiple prefixes over DHCPv6-PD to get temporal overlap, but at least homenet needs to support this if it's there.
[15:11:31] <Wes George> slide 2
[15:12:11] <Ole Troan> Mikael, what do you mean? DHCPv6 PD has always supported that.
[15:12:32] <Wes George> slide 3
[15:12:35] <Mikael Abrahamsson> ole, let's talk about that later then. I want to know.
[15:12:53] <Markus Stenberg> PD has always supported multiple prefixes (with different lifetimes)
[15:12:55] <Ole Troan> I'll do anything for beer. ;-)
[15:13:06] <Markus Stenberg> I have implemented DHCPv6 PD too, can I have beer? :-)
[15:13:13] <Wes George> slide 4
[15:13:25] <Markus Stenberg> of course, Ole's name is on 'the stuff'
[15:13:26] <Wes George> you need a shirt "will deploy IPv6 for beer"
[15:13:32] <Wes George> slide 5
[15:13:52] <Ole Troan> I like Lorenzo's "I will not stand in the way of people who want to do useless work"
[15:13:56] <Ole Troan> for shirts
[15:14:00] <Wes George> +1
[15:14:03] <Wes George> slide 6
[15:14:07] Simon Romano joins the room
[15:14:22] <Mikael Abrahamsson> I see Lorenzo having opinions on useless proposals all the time :P
[15:14:35] <Markus Stenberg> I think it is perhaps 'do not as I do, but as I say'
[15:15:22] <Wes George> slide 7
[15:15:24] Simon Romano leaves the room
[15:15:55] <Wes George> slide 8
[15:16:52] <Wes George> slide 9
[15:17:25] <Mikael Abrahamsson> peter lothberg
[15:17:40] shwetha bhandari joins the room
[15:18:45] leslie joins the room
[15:18:47] <Mikael Abrahamsson> michael richardson
[15:19:33] Peter DeVries leaves the room
[15:19:33] Brian Haberman joins the room
[15:21:00] <Juliusz Chroboczek> The newest, but yes.
[15:22:17] <Mikael Abrahamsson> peter lothberg
[15:23:27] <Mikael Abrahamsson> pascal thubert
[15:23:37] <Mikael Abrahamsson> janos farkas
[15:23:48] <Wes George> slide 10
[15:26:09] J Hadi Salim joins the room
[15:26:20] Miles McCredie leaves the room
[15:26:23] Ralf Weber leaves the room
[15:27:06] <Wes George> slide 11
[15:27:17] LeeHoward@lightwitch.org leaves the room
[15:27:19] <Mikael Abrahamsson> alex patrescu
[15:27:37] <Juliusz Chroboczek> mic: independent
[15:27:38] suprita s joins the room
[15:27:44] leslie leaves the room
[15:27:51] <Juliusz Chroboczek> cancel mic
[15:27:58] <Wes George> ack and ac
[15:28:00] <Ted Lemon> ?
[15:28:03] <Ted Lemon> ah
[15:29:13] <Juliusz Chroboczek> mic: I can provide peer-reviewed references.
[15:29:27] <Mikael Abrahamsson> alex patrescu was at the mic
[15:29:44] <Mikael Abrahamsson> alia atlas at mic now
[15:29:49] <Ole Troan> Isn't adhoc networks outside of homenet scope?
[15:30:13] <Juliusz Chroboczek> Ole, yes.
[15:30:17] <Mikael Abrahamsson> ole, we have no concensus on what a homenet is :P
[15:30:46] <Markus Stenberg> blink, I have yet to see (readable) IS-IS (core) RFC.
[15:30:48] <Markus Stenberg> does one exist?
[15:30:48] <Ted Lemon> We do have a consensus.   We published a document that says what a homenet is.
[15:31:37] <Juliusz Chroboczek> Who is that?
[15:31:41] <Markus Stenberg> routing AD
[15:31:41] <Wes George> alia atlas
[15:31:42] <Mikael Abrahamsson> alia atlas
[15:31:42] <Ted Lemon> Alia Atlas
[15:32:38] Miles McCredie joins the room
[15:32:40] <Markus Stenberg> what the fuck.. 'I have not read it but it is not ready'
[15:32:46] Miles McCredie leaves the room
[15:33:10] <Wes George> well, it's not. it's not a standards doc
[15:33:13] <Mikael Abrahamsson> lorenzo colitto now at mic
[15:33:15] Miles McCredie joins the room
[15:33:19] <Markus Stenberg> that isn't definition of ready
[15:33:20] marka leaves the room
[15:33:33] <Markus Stenberg> she was judging document quality by hearsay
[15:33:48] <Mikael Abrahamsson> chris hopps
[15:33:48] marka joins the room
[15:33:50] <Markus Stenberg> and it has e.g.  academically peer reviewed paper
[15:33:50] <Wes George> if it's ready to become a standard, then it won't take a long time to become one
[15:34:06] <Wes George> academically reviewed != implementation experience and IETF review
[15:34:27] <Mikael Abrahamsson> david lamparter now at mic
[15:34:37] <Markus Stenberg> point me at IS-IS RFC that is readable
[15:34:42] <Mikael Abrahamsson> chris hopps again
[15:34:49] <Markus Stenberg> and you can claim _it_ is standard
[15:34:54] <Markus Stenberg> (I don't count other std org documents)
[15:35:03] <Mikael Abrahamsson> jari arrko
[15:35:07] John Scudder joins the room
[15:35:28] <Wes George> I'm not going to debate readability with you. IS-IS, regardless of the density of the docs is widely implemented and a product of ietf consensus
[15:35:30] <Mikael Abrahamsson> arkko, sorry.
[15:35:30] <ida leung> but ISIS do has a lot more implementation especially in ISP
[15:35:31] <John Scudder> https://www.ietf.org/rfc/rfc1142.txt is the RFC-ification of ISO 10589.
[15:35:48] <Mikael Abrahamsson> peter lothberg
[15:35:49] <Markus Stenberg> unreadable one, but yes.
[15:36:09] <David Lamparter> John: i don't think 1142 is sufficiently complete to implement IS-IS
[15:36:28] <Markus Stenberg> if we compare 1142 and Babel RFC quality
[15:36:30] <Markus Stenberg> I think my vote's on Babel
[15:36:40] <Ted Lemon> We don't vote.
[15:37:02] Peter DeVries joins the room
[15:37:04] <Wes George> then babel won't have any trouble becoming a full standard and this argument is moot
[15:37:18] <John Scudder> I admit to not being up to date on my ISO/IETF techno-politics. 10589 is a high-quality protocol spec IMO. Assuming 1149 is a faithful copy of it, well then.
[15:37:28] <Markus Stenberg> 1149 is unreadable ASCII dump of it
[15:37:30] <John Scudder> s/1149/1142/
[15:37:31] <Wes George> requiring babel to follow the process is not necessarily advocating for ISIS
[15:37:34] <Markus Stenberg> or whatever the # was
[15:38:09] <Mikael Abrahamsson> david lamparter at mic
[15:38:30] <Mikael Abrahamsson> peter lothberg
[15:38:43] <John Scudder> The quality of the ASCII dump is to some extent techno-politics, insofar as fully-formatted 10589 is readily available. Of course, you might say that about insisting Babel be in the IETF process too.
[15:38:48] <Wes George> slide 13
[15:39:37] <each> 1142 is 25 years old. surely the spec has drifted a bit since then?
[15:39:37] <Mikael Abrahamsson> chris hopps
[15:39:46] shinta leaves the room: Replaced by new connection
[15:39:46] shinta joins the room
[15:39:56] LeeHoward@lightwitch.org joins the room
[15:40:13] <John Scudder> AFAIK all the meaningful revs to IS-IS have been done in the IETF and are published in the ISIS WG.
[15:40:15] <Markus Stenberg> sadly it has not much (the OSI-side spec)
[15:40:21] <Markus Stenberg> there's shitload of other documents needed to understand it
[15:40:35] <John Scudder> There may have been minor updates to the ISO spec, but the 25 year old one is good enough.
[15:40:48] <Wes George> but ultimately that's a problem in lots of areas of IETF. one doc gets updated by 5 others, all of which are updated, etc
[15:40:54] <John Scudder> Yep
[15:40:58] <Wes George> there's no document that represents the union of docs
[15:40:58] <Markus Stenberg> but not particularly applicable to implementing one from scratch; you have to essentially avoid 2/3 of the document
[15:40:58] Ralf Weber joins the room
[15:41:02] <Wes George> slide 14
[15:41:26] suprita s leaves the room
[15:41:51] <Wes George> slide 15
[15:42:06] <Juliusz Chroboczek> 700 mails, ugh.
[15:42:46] <Mikael Abrahamsson> pascal thubert
[15:45:29] <Mikael Abrahamsson> david lamparter
[15:45:31] Brian Haberman leaves the room
[15:46:07] <Wes George> ted lemon
[15:46:26] <Mikael Abrahamsson> michael richardsson
[15:46:37] <Mikael Abrahamsson> david lamparter again
[15:47:09] <Juliusz Chroboczek> Thanks, Margaret.
[15:47:20] <Wes George> Ted opening discussion
[15:49:12] <Mikael Abrahamsson> lorenzo colitti
[15:49:32] <Mikael Abrahamsson> steven barth
[15:49:56] <Juliusz Chroboczek> mic: Re "optimisation for lossy links", more generally optimisation for networks that have links of varying quality (not only loss, bandwidth is very variable in homenets).
[15:50:21] <Juliusz Chroboczek> s/bandwidth/throughput/
[15:50:52] <Wes George> ross callon
[15:50:52] <Mikael Abrahamsson> I can't read the name
[15:51:05] jlatour leaves the room
[15:51:09] <Mikael Abrahamsson> thanks
[15:51:19] leslie joins the room
[15:51:27] shinta leaves the room
[15:51:27] shinta joins the room
[15:51:34] <Mikael Abrahamsson> margeret wasserman
[15:51:57] <Mikael Abrahamsson> chris hopps
[15:52:35] shinta leaves the room
[15:53:18] <Mikael Abrahamsson> so, personally, I believe there are two main questions that we need to decide on: To what extent is the home going to use "crappy wifi", and how important is small footprint running code going to influence the decision. Since we do not have consensus on these parameters, this is the main problem for me and why I say we don't have consensus on what a homenet is.
[15:53:54] <Mikael Abrahamsson> david lamparter
[15:54:07] <Mikael Abrahamsson> michael richardson
[15:54:13] <Mikael Abrahamsson> lorenzo colitti
[15:54:54] leslie leaves the room
[15:55:04] leslie joins the room
[15:55:10] <Juliusz Chroboczek> Yeah, I wish I had taken the time to compile the Quagga IS-IS version, but ran out of time.
[15:55:36] <Markus Stenberg> I have some incomplete #'s, but this measurement is mostly meaningless
[15:55:38] <Mikael Abrahamsson> ted lemon speaking
[15:55:43] <Markus Stenberg> you can make small IS-IS (even real one) if you want to
[15:56:04] jlatour joins the room
[15:56:05] <Juliusz Chroboczek> Uh-huh.
[15:56:07] <Markus Stenberg> that is not the question; from my point of view, though, I cannot use it this year, or probably even next, if I want it to work
[15:56:14] jlatour leaves the room
[15:56:15] <Markus Stenberg> e.g. BIRD OSPF is not much bigger than Babel
[15:56:19] <Markus Stenberg> and it's link state routing protocol
[15:56:48] <Mikael Abrahamsson> lorenzo colitti at mic
[15:57:07] <Markus Stenberg> but I find beauty contest with 'some' implementations and incomplete specs mostly sad
[15:58:23] leslie leaves the room
[15:58:43] <Mikael Abrahamsson> margeret wasserman
[15:59:05] <each> IMHO, the "crappy wifi" consideration will get bigger and bigger over time, and the "code size" consideration will get smaller and smaller. routers will get more memory; airwaves will get more crowded. so I wouldn't choose babel over is-is because of code size but I might choose it because it was optimized from the start for variable-quality links
[15:59:15] jlatour joins the room
[15:59:17] iffy50@jabber.org joins the room
[15:59:41] <Mikael Abrahamsson> if 802.11 fixes their multicast problem, then most of the problems of using ISIS for the home goes away
[15:59:47] <Mikael Abrahamsson> Pierre Pfister
[16:00:08] <Juliusz Chroboczek> Mikael: avoiding marginal links?
[16:00:11] <David Lamparter> MIPS: 672kB RSS
-rwxr-xr-x    1 root     root      278.3K Feb 17 05:30 isisd
AMD64: 1496kB RSS
-rwxr-xr-x  1 equinox equinox 303K Mar 24 10:59 isisd
[16:00:51] <Markus Stenberg> David Lamparter: does it need quagga too?
[16:00:54] <Juliusz Chroboczek> Thanks, David.  Does this include the zebra daemon?
[16:01:01] <Markus Stenberg> or zebra, ya
[16:01:17] <David Lamparter> it doesn't - but I'd claim that shouldn't be part of the comparison ;)
[16:01:23] <Mikael Abrahamsson> Ted Lemon
[16:01:26] <Markus Stenberg> why? it has parts of the infra
[16:01:33] <Markus Stenberg> (although admittedly more than is needed for minimal case)
[16:01:38] <David Lamparter> Markus: the infra is needed for redistribution between protocols
[16:01:47] <David Lamparter> the Auto-ISIS implementation can run with or without zebra
[16:01:50] <Mikael Abrahamsson> alia atlas
[16:01:57] <Juliusz Chroboczek> Who was the guy before?
[16:02:01] <David Lamparter> without zebra, it loses redistribution and management capabilities
[16:02:19] <Markus Stenberg> Pierre Pfister
[16:02:22] <Markus Stenberg> was the previous one
[16:02:25] <Mikael Abrahamsson> pascal thubert
[16:02:57] <Mikael Abrahamsson> ted lemon
[16:04:46] Olafur joins the room
[16:05:47] leslie joins the room
[16:06:07] <Mikael Abrahamsson> pascal thubert
[16:06:28] tob cast joins the room
[16:06:30] Simone Romano joins the room
[16:07:17] <Mikael Abrahamsson> ted, we do not have consensus on how important it is to handle crappy links. This means we will never have consensus on what routing protocol to use.
[16:07:29] <Mikael Abrahamsson> peter lothberg
[16:08:19] <Markus Stenberg> sigh.. marketing.
[16:08:50] tob cast leaves the room
[16:09:07] <Juliusz Chroboczek> "It could be done, we just haven't done it yet".
[16:09:17] <Markus Stenberg> "it can be specified, and implemented". yay.
[16:09:28] <Markus Stenberg> I think we will wind up running Steven's incremental stuff and just leave routing out in _practise_
[16:09:31] <Markus Stenberg> regardless of what IETF does
[16:09:36] <Juliusz Chroboczek> Grr.
[16:09:40] <Mikael Abrahamsson> jim gettys
[16:09:48] Simon Romano joins the room
[16:09:49] <Markus Stenberg> because I do not have faith in this converging in finite time
[16:11:01] <Mikael Abrahamsson> wes george, jabber scribe at mic
[16:11:10] <Mikael Abrahamsson> oh, not in jabber scribe role
[16:11:30] Dan York joins the room
[16:11:49] Simon Romano leaves the room
[16:11:53] <Markus Stenberg> we can run Babel alongside that, and FIB in kernel is bit fatter, and someone can run IS-IS as well if they want to
[16:12:08] <Markus Stenberg> => from my point of view, it is sufficient, and allows for migration between RPs if someone really wants to, without dropping connectivity
[16:13:25] <Juliusz Chroboczek> Still George?
[16:13:29] <LeeHoward@lightwitch.org> Amen, Wes!
[16:13:41] Simone Romano leaves the room
[16:14:29] <Markus Stenberg> Alia
[16:15:10] <Wes George> for those on jabber: the issues are:
[16:15:13] <Wes George> transitivity
[16:15:23] <Wes George> optimization for variability of links
[16:15:37] <Wes George> tradeoff in terms of how much work is required to adopt babel vs is-is
[16:15:45] <Wes George> consensus progress for use of babel if it is chosen
[16:16:12] <John Scudder> s/progress/process/
[16:16:24] <Juliusz Chroboczek> mic: Babel has all the usual optimisations for wired networks, unlike some other mesh protocols.
[16:16:46] <Wes George> fair warning, the mic line stands at 8+ people
[16:16:55] <Juliusz Chroboczek> ok, cancel.
[16:18:22] <Markus Stenberg> most home links will be wireless at any rate; question is, how many of them will be homenet
[16:18:35] <Markus Stenberg> and how many some other meshy things or whatever
[16:19:22] <Wes George> margaret wasserman
[16:19:30] <Markus Stenberg> Ted Lemon: amusing idea (having a race that he wrote on the presentation screen)
[16:19:31] <Wes George> and before her david lampeter
[16:19:38] <Markus Stenberg> unfortunately 'standards process races' take years
[16:19:56] <Wes George> fastest lame horse wins!
[16:20:20] <John Scudder> that is not correct (about the two implementations). AFAIK.
[16:20:25] <Markus Stenberg> it's like watching dead snails racing; sometimes wind may blow one forward
[16:20:42] <Juliusz Chroboczek> Fair point Margaret is making.
[16:20:43] <Wes George> lol
[16:20:47] <John Scudder> I wish it were correct, but it's not.
[16:21:00] <Wes George> indeed, but it's work to be done (or if it's done, documented)
[16:21:00] iffy50@jabber.org leaves the room
[16:22:22] <each> so to summarize: "is-is isn't good enough but maybe it can be if someone fixes it" vs "babel is good enough but ugh ietf standardization"?
[16:22:31] <Markus Stenberg> yes
[16:22:31] <Juliusz Chroboczek> Hmm, "just a feature"?  Using variable link metrics in link-state without stability issues?
[16:22:47] <Wes George> alex petrescu
[16:22:57] <Markus Stenberg> Juliusz Chroboczek: also lack of even static default metrics for IS-IS and how the hell to get them working given platforms without L2 visibility + switches + etc.
[16:23:06] <Markus Stenberg> and IS-IS L2 dependence
[16:23:12] <Markus Stenberg> guess the IS-IS over IPv6 fixes that
[16:23:15] <Markus Stenberg> if I find that someday ;)
[16:23:43] <Wes George> alia atlas
[16:24:41] <Juliusz Chroboczek> Competent and not biased?
[16:24:50] <Markus Stenberg> good luck with that
[16:25:01] leslie leaves the room
[16:25:25] Peter DeVries leaves the room
[16:25:26] <Markus Stenberg> wonder if she expects us to be hard of hearing
[16:25:59] <LeeHoward@lightwitch.org> [drops mic]
[16:26:37] <Wes George> mikael abrahamsson
[16:26:43] <Markus Stenberg> john brzowski
[16:26:44] <Wes George> john brzowski
[16:27:37] jlatour leaves the room
[16:27:42] LeeHoward@lightwitch.org leaves the room
[16:27:44] <Mikael Abrahamsson> chris grundeberg
[16:27:50] <Wes George> grundemann
[16:28:01] <Mikael Abrahamsson> ack.
[16:28:02] Brian Carpenter leaves the room
[16:28:16] Brian Carpenter joins the room
[16:28:57] Olafur leaves the room
[16:29:01] <Juliusz Chroboczek> HOw many people, roughly?
[16:29:13] <Mikael Abrahamsson> ted lemon is speaking
[16:29:30] Peter DeVries joins the room
[16:29:40] <John Scudder> there were a very small number of hands each time, but the questions are so confusingly asked I don't think the counts are meaningful
[16:29:41] <Markus Stenberg> dozen or two were against IS-IS
[16:29:44] Ralf Weber leaves the room
[16:29:48] <Markus Stenberg> yep
[16:29:52] <Mikael Abrahamsson> chris hopps
[16:29:52] <Markus Stenberg> (out of probably 150-ish)
[16:30:08] Jason Weil leaves the room
[16:30:18] <Mikael Abrahamsson> tony przgenda
[16:30:23] <Wes George> "..too painful" should be the theme for this particular discussion
[16:30:45] C Hopps leaves the room
[16:31:26] Miles McCredie leaves the room
[16:31:31] <Ole Troan> ask the question if there is support for Atlas' proposal?
[16:31:36] Peter Koch leaves the room
[16:33:16] Ted Lemon leaves the room: Disconnected: closed
[16:33:17] <Wes George> session ends
[16:33:17] John Scudder leaves the room
[16:33:18] <each> I want to see running code that makes is-is meet the needs of poor-quality wireless networks as well as babel. if there's running code, then is-is wins on expedience.
[16:33:22] Markus Stenberg leaves the room
[16:33:26] marka leaves the room
[16:33:35] Wes George leaves the room
[16:33:39] <Juliusz Chroboczek> each, it's an open research problem.  An interesting one.
[16:33:48] <each> right
[16:34:04] Ole Troan leaves the room
[16:34:10] <Juliusz Chroboczek> The OSPF crowd tried, I don't have the RFC number handy.  Search for MPR optimization.
[16:34:16] Paul Selkirk leaves the room
[16:34:18] Mikael Abrahamsson leaves the room
[16:34:34] <each> basically, I'm more sure that babel can be standardized than I am that is-is can be fixed
[16:34:53] shwetha bhandari leaves the room
[16:35:11] Meetecho leaves the room
[16:35:39] <Juliusz Chroboczek> each: feel free to drop on #babel at freenode if you want to chat some more.
[16:36:09] Andrew Yourtchenko_m leaves the room
[16:36:56] each leaves the room
[16:38:00] Mark Baker leaves the room
[16:40:11] Franck Martin leaves the room
[16:42:47] David Lamparter leaves the room: offline
[16:45:11] Brian Carpenter leaves the room
[16:45:22] Matthieu Boutier leaves the room
[16:50:32] ida leung leaves the room
[16:55:09] Peter DeVries leaves the room
[16:59:56] iffy50@jabber.org joins the room
[17:00:41] iffy50@jabber.org leaves the room
[17:00:57] iffy50@jabber.org joins the room
[17:03:44] Brian Carpenter joins the room
[17:06:20] Markus Stenberg joins the room
[17:16:13] Markus Stenberg leaves the room
[17:17:00] Markus Stenberg joins the room
[17:17:21] Brian Carpenter leaves the room: offline
[17:18:20] Markus Stenberg leaves the room
[17:33:24] Markus Stenberg joins the room
[17:33:50] marka joins the room
[17:37:12] iffy50@jabber.org leaves the room
[17:52:13] Mikael Abrahamsson joins the room
[17:57:39] Ralf Weber joins the room
[17:57:59] Wes George joins the room
[18:08:10] Ted Lemon joins the room
[18:08:24] Ted Lemon leaves the room
[18:09:09] John Scudder joins the room
[18:09:21] John Scudder leaves the room
[18:12:30] iffy50@jabber.org joins the room
[18:13:12] iffy50@jabber.org leaves the room
[18:13:31] iffy50@jabber.org joins the room
[18:13:54] iffy50@jabber.org leaves the room
[18:21:02] LeeHoward@lightwitch.org joins the room
[18:21:29] Markus Stenberg leaves the room
[18:22:05] Markus Stenberg joins the room
[18:22:06] Markus Stenberg leaves the room
[18:22:11] Markus Stenberg joins the room
[18:23:41] Markus Stenberg joins the room
[18:24:37] Wes George leaves the room
[18:24:41] Markus Stenberg leaves the room
[18:25:30] Markus Stenberg leaves the room
[18:25:35] Markus Stenberg joins the room
[18:30:26] Markus Stenberg leaves the room
[18:35:38] jlatour joins the room
[18:37:26] Brian Haberman joins the room
[18:37:47] Brian Haberman leaves the room
[18:44:01] marka leaves the room
[18:44:01] jlatour leaves the room
[18:44:02] marka joins the room
[18:51:04] jlatour joins the room
[18:56:14] marka leaves the room
[18:56:45] Peter Koch joins the room
[18:56:51] Peter Koch leaves the room
[18:59:12] jlatour leaves the room
[19:00:29] jlatour joins the room
[19:00:36] jlatour leaves the room
[19:04:37] LeeHoward@lightwitch.org leaves the room
[19:11:46] marka joins the room
[19:18:40] Mikael Abrahamsson leaves the room
[19:35:48] Olafur joins the room
[19:40:33] Jason Weil joins the room
[19:41:20] Ralf Weber leaves the room
[19:53:59] Jason Weil leaves the room
[19:55:43] Olafur leaves the room
[19:56:35] Olafur joins the room
[20:05:00] marka leaves the room
[20:07:43] Olafur leaves the room
[20:10:07] marka joins the room
[20:20:06] Olafur joins the room
[20:23:20] Ralf Weber joins the room
[20:23:30] leslie joins the room
[20:23:53] Ralf Weber joins the room
[20:23:54] Ralf Weber leaves the room
[20:25:38] Olafur leaves the room
[20:34:19] Jason Weil joins the room
[20:39:20] leslie leaves the room
[20:40:01] Jason Weil leaves the room
[20:59:06] leslie joins the room
[21:01:35] leslie leaves the room
[21:14:32] <marka> REFUSED + SOA + AA=1
[21:15:01] <marka> for zone level refused
[21:36:37] Ralf Weber leaves the room
[22:05:02] leslie joins the room
[22:06:00] leslie leaves the room
[22:22:26] marka leaves the room
[22:35:22] marka joins the room
[22:44:42] leslie joins the room
[23:08:36] ozindfw leaves the room
[23:14:32] leslie leaves the room
[23:27:14] marka leaves the room
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!