13:36:45 #startmeeting 13:36:45 Meeting started Wed Jan 20 13:36:45 2016 UTC. The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:36:45 Useful Commands: #action #agreed #help #info #idea #link #topic. 13:37:02 it's ok, you rolled in at 1337 UTC. s'all good 13:37:11 hi all, i'll start with a status update 13:37:34 I've been under a dark spell, doing administrative and whatnot stuff. I hope to crawl out. 13:37:46 been merging patches, reviewing code. It would be nice to write some soon 13:37:49 Hope to get there. 13:37:58 first must at least get pre-Jan patches out of my queue. 13:38:34 also today I'm hanging out with Ben Schwartz from Google, who is interested in a "v2" pt spec that would do more things more pt-integrators want. 13:39:17 I'm unclear whether this is actually a good idea, but it's nice to spend afternoons with engineers. And I've had a chance to talk with Str4d about i2p's needs a bit here too, so at least I'll have something else to go on 13:39:31 ah nice 13:39:53 Boston will be hit by a snowstorm over the weekend; Shari is visiting next week; I'm still optimistic I can Get Stuff Done. :) 13:39:56 Who's next? 13:39:59 (hi all!) 13:40:47 i can go. 13:40:55 go! 13:41:00 i did some post-review prop250 stuff this week. 13:41:18 prepared the spec, and an illustration of the protocol: https://people.torproject.org/~asn/hs_notes/shared_rand.jpg 13:41:23 wrote some emails about it 13:41:33 and then I got back into the guard stuff preparing for the reading group 13:41:40 i re-read prop247 and a bit of prop259 13:41:55 did the reading group eysterday which went reasonably well. mainly it boosted me into remembering things again. 13:42:03 i'm planning to work on prop247 and prop259 on the short-term future 13:42:15 I also wrote an email to tor-dev with the summary of yesterday's meeting 13:42:35 and also i talked with ola bini who wants to help (along with a team of 4 devs) in hidden services development startign in early february 13:42:42 (there is a short emial about this in tor-assistants) 13:42:45 EOF 13:42:47 peace out 13:43:33 Neat! Who else is around? 13:43:35 * dgoulet can go 13:43:40 zoom! 13:44:24 (go) 13:44:32 Worked on #13802 (currently waiting on Rob,s testing/feedback), also started #18098 (for prop#224) 13:45:02 proposal session yesterday was great, I had the chance to study quite in depth the guard proposals and looking forward to comment more on the next steps done by asn and mikeperry 13:45:07 (and isis and all! :P) 13:45:42 tomorrow I'll be presenting Tor at Université de Limoges to students and professors so I'm fixing some slides 13:46:19 cool. English or French? 13:46:20 so I expect to continue 224 work this week and next week along with some OnionPerf graphing 13:46:22 --- 13:46:23 nickm: french :) 13:46:27 woo 13:46:50 We have canonical translations for our key glossary terms, right? :) 13:47:05 oh my difficult! I kept the english haha :P 13:47:23 who else is here? mikeperry ? athena ? 13:47:24 "Threat Model" is suprisingly a difficult one to translate :P 13:48:09 dgoulet: maybe "théorie d'attaque" ? 13:48:41 i'd say menace au lieu d'attaque 13:48:48 modèle de menace 13:48:51 this is the on I use 13:49:00 jvoisin: that! great :) 13:49:05 I have been working on Prop#247 and Prop#265, and looked briefly at prototyping Prop#247 in txtorcon.. that looks like it will be ugly and require either crazy hacks and/or patches to tor. still might just barely be doable 13:49:16 feel free to add a `s` to menace 13:49:26 then I went down a rabbit hole thinking about path lengths and other path constraints for prop#247, which is where I am now 13:49:32 about to brain dump that on the mailinglist 13:50:27 neat. 13:50:50 mikeperry: path lengths? (i guess i'll wait for the mailinglist post. :) 13:50:53 yeah, hopefully I can get that mail out before we hit discussion. I'm done 13:50:54 I'm hoping we can come up with something working in the guardsim-type world so we can reduce it to code an know we agree about what it's supposed to do 13:50:59 isis ? athena ? 13:51:09 nickm: you missed mine i think 13:51:23 05:36 < athena> anyway, i'm working on rebasing and revising #4581 per review, and reading up on proposals for forthcoming proposal review sessions 13:51:29 ah 13:51:36 how's that 4581 going? 13:52:56 oh, rebase went smooth modulo a few new unit tests failing to initialize something in the config it wants and throwing an assert, tweaks per code review seem simple - mostly just need to write new unit tests for it and toss v2 in for review 13:54:46 i've been reading the proposals for the upcoming sessions, revising the guardsim code and triple checking that it exhibits the best behaviour i can get it to, and determining what (if anything) needs to change in prop#259. i will work on that a bit more today, then go to a reception at amnesty international, then possibly tomorrow start on #7144 revisions. 13:55:40 other than that i should probably pick a thing from my list of stuff i said i'd do for OTF for bridgedb and do it so that i can pay rent and stuff. 13:56:33 woo 13:56:52 how are you testing the guardsim behavior? Like, running it? 13:57:07 ah! and i am going to raboud universiteit in nijmegen in a week or so to visit with peter schwabe and some of his students who want to start working on tor stuff :) 13:57:14 whoa 13:57:22 say hi to peter 13:57:34 what kind of stuff are they interested in? 13:57:43 I mainly know him from his high-performance implementation stuff 13:57:44 nickm: yeah, by just running it a bunch of times with different combinations of flags 13:58:08 isis: Do these combinations sorta map to the cases in doc/stuff-to-test.txt ? 13:58:34 i actually did not know that that document existed. i will go read it after the meeting. 13:59:19 If you can edit it to note which cases have implementations and which cases don't, we can start writing the ones that don't. 13:59:38 Extra kudos if you say which combinations of flags to use for which case :) 14:00:11 sure, sounds like a good idea 14:00:26 and yeah, i'll say hi to peter :) 14:01:04 his students mainly seems to want help knowing what part of tor to get involved in, one is an undergrad and the other a masters 14:01:12 neat 14:01:24 oh, i have a related thing we should think about over the next week or so: 14:02:10 I'm maybe going to be mentoring a small group of MIT undergrads as they hack on an open-source project for credit in a software lab this term 14:02:19 We should think about projects to suggest 14:02:44 selfishly: bridge bandwidth measurements :) 14:03:12 nickm: that would be code here not only research? 14:03:33 yup, code 14:03:52 and I think it's gotta be on an existing project, not making something new _de novo_ 14:04:43 ok, prop#247 path length brain dump heaved to tor-dev 14:05:54 so have we got any discussion stuff today, or should we call it a meeting? 14:06:05 asn_: one thing that thoughtworks mighe be able to help with is make hidden services controllable via the control port. I am just going to hack it as a research prototype for prop#247, but they could make it nicer.. they could also help with some of the path selection code refactoring/work 14:06:17 I'd like to ask everybody to take a look at the TorCoreTeam201601 tag and move stuff out if it's not going to happen :) 14:08:41 hmm, #endmeeting? 14:09:35 #endmeeting then