18:00:31 #startmeeting race bi-weekly checkin 2019-06-27 18:00:31 Meeting started Thu Jun 27 18:00:31 2019 UTC. The chair is phw. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:31 Useful Commands: #action #agreed #help #info #idea #link #topic. 18:00:41 here's our pad: https://pad.riseup.net/p/tor-censorship-S28-keep 18:01:12 ok. I'm reading your update there phw 18:01:29 right, so we made a bit of progress on the obfs5 and the PT spec front 18:02:06 obfs5 seems to be progressing well and we have lots of help from micah and the nrl, which is great 18:02:49 i dropped the ball a bit on the PT spec front 18:03:19 we've been in touch with a race collaborator who sees some android-related issues with our PT spec 18:03:39 sysrqb provided some insight and i also intend to talk to nathan in stockholm 18:04:35 so that's my priority over the next few days 18:05:08 nice! 18:05:09 ok, should that conversation on PT spec being public during stockholm? 18:05:57 my meeting with nathan is mostly about me understanding how they solved challenges that other race performers anticipate 18:06:00 if the spec concern is the "subprocess handling" then talking to mike on iOS might be useful too 18:06:09 ack 18:06:39 right, ios is another issue. probably not for race (because the goal is an android app) but we should keep this in mind too 18:07:21 i think the two issues can be solved at the same time if it is the subprocess issue 18:07:34 ahf: fwiw, the concern was "Our concerns on Android were being able to setup and run a socks proxy on Android without root. The RACE app must work on un-rooted phones so there is no undue burden on RACE users and accessible to as many people as possible." 18:08:31 cohosh: do you want to talk about how the brainstorming and integration meetings went and what it came out of it for us? 18:08:34 aiui, the v2 PT spec may solve this because it also defines how to use PTs as a library 18:08:51 yeah 18:09:41 ios people does this right now by a hack mike made with statically linking in the PT's. if we can handle dynamic (or static) linked PT's on android that would be great too 18:10:10 ah, gotcha 18:11:29 gaba: yep 18:11:59 right now the integration meetings are a bit confusing but what they're aiming for is to provide everyone working on obfuscated channels wiht a list of properties 18:12:27 and we're going to have to be able to report to the applications using our channels about which properties our "link" has 18:12:46 so for example if we're using tcp as the underlying transport then our link is "reliable" 18:13:02 if we send information thorugh Tor then it is "anonymous", etc 18:13:23 we've already given them the dependencies we think we'll need 18:13:46 hmm, the PT spec has a method that would allow a PT to communicate its properties but we're not doing that 18:13:53 so we're really just waiting on them to make more progress with the API (which right now their focusing on the link properties part of that) 18:14:02 phw: oh interesting 18:14:50 cohosh: i think it's just the PT writing to stdout, which the invoking process can read from. not sure what the v2 spec does in that regard. 18:15:22 ok, might be interesting to point that out to them 18:15:49 i'll do a bit more reading, so i'll have a better answer for you before the next integration meeting 18:15:59 sounds good 18:16:40 ok 18:17:01 we did a STATUS API in v1 PT for sponsor 8 that can be used to emit any key/value sets to the parent process that we can process 18:17:28 oh, thanks ahf 18:19:31 cohosh: wasn't the TA1 group supposed to provide anonymity in race? 18:19:39 that's what I thought 18:19:46 i guess it comes down to what you define as anonymity 18:19:49 we're trying to figure out what's going on 18:19:54 gotcha :) 18:20:35 either way, it would be good for us to have an easy answer for "how do we communicate properties?" 18:20:44 yes 18:21:03 gaba: our monthly report is due soon, right? how can we help? 18:21:08 yes 18:21:13 i just put a link in the pad 18:21:24 if you have any ticket that you want to add as things the team did in june 18:21:26 please add it there 18:21:33 wasn't there a nextcloud document somewhere, too? 18:21:44 * gaba checking 18:21:45 i remember adding some tickets there 18:21:59 sorry, maybe it was and I forgot. Let me check. 18:22:35 https://nc.riseup.net/apps/onlyoffice/s/MdNwgWk8CL8FBis 18:23:15 yes. let's do it there. I have that doc in the s28 folder in nc 18:23:44 so this is all sponsor 28 tickets we've done? 18:23:47 and it is any work since we started until june 18:23:50 yes 18:24:06 ah cool 18:24:28 does our report go directly to darpa? or do we send it to georgetown? 18:24:45 georgetown 18:24:50 they will add it to their own report 18:25:07 oh, i see. so all we need is some prose that micah can then turn into a more comprehensive report 18:25:12 yes 18:26:09 anything else that we should be discussing? i have nothing else from my side 18:26:10 this is an example on how we do it for other sponsors https://pad.riseup.net/p/QI0pdcp0_mR0TPRu-tT7 18:26:30 ah, thanks gaba 18:27:19 i think that is all 18:27:38 we can have next update during Tor meeting 18:27:54 yes, let's do that 18:28:03 #endmeeting