17:01:01 #startmeeting anti-censorship weekly checkin 2019-09-19 17:01:01 Meeting started Thu Sep 19 17:01:01 2019 UTC. The chair is phw. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:01 Useful Commands: #action #agreed #help #info #idea #link #topic. 17:01:11 hi everyone 17:01:17 here's our meeting pad: https://pad.riseup.net/p/tor-censorship-2019-keep 17:01:34 hi 17:01:53 Hi 17:01:58 * catalyst is sort of here 17:02:18 hey serna, good to have you around again 17:02:19 i am a tiny bit here but will be running off to medical adventures soon 17:02:42 my main item is that when i am back, i plan to get up to speed on all the angles of race that we (the race team) are doing, so i can explain them to people in the hallways at the PI meeting 17:03:16 sounds good, arma2 17:03:27 phw: I came to ask some help with #25598 17:03:52 looks like gaba has two reminders for today: we should use the BugSmashFund keyword for non-sponsor tickets and we should remember to use 'actual points' when closing tickets 17:04:35 serna: alright, let's talk about it after we're done with our announcement and discussion section, ok? 17:04:43 i am really bad about putting actual points 17:04:45 (in the meeting pad, that is) 17:05:02 Sure 17:05:05 cohosh: to keep track of them or to remember to enter them? 17:05:09 both 17:05:25 o/ 17:06:19 i'll try to remember to do it 17:06:26 i'm working with this "time tracker" gnome app. i use it to keep track of how much time i spend per ticket. it sometimes get messy though, when you get interrupted and forget to update it 17:06:33 yeah 17:06:39 it's the best solution i have for now, though. i'm curious if anyone has found something better 17:06:52 keeping track of my time on this granularity stresses me out 17:07:09 does it have to be very accurate? 17:07:18 it does not have to be totally accurate 17:07:25 if paper and pencil works for you then go for it 17:07:36 ok 17:07:54 i think it would be useful to at least have a sense of "this took much, much longer than anticipated because of X, Y, and Z" 17:08:55 yes, that is mostly the idea 17:09:01 ok, a brief announcement from my side: i took a look at bridgedb's logs the other day and it looks like the bot switched its http headers, to bypass our anti-bot mechanism 17:09:02 and to get better at estimating tasks 17:09:13 :o 17:09:14 so our theory that it's no longer maintained doesn't seem true 17:09:29 that's so interesting. how many days did it take to adapt? 17:09:30 wow interesting 17:09:31 i wouldn't be surprised if whoever made this change is reading this now 17:09:57 dcf1: i don't know, i'm afraid. our bridgedb logs only go back a few hours 17:09:58 that is kind of creepy :P 17:10:26 phw: yeah but upper bound, when did you make the change and when did you notice the headers had changed to compensate? 17:10:50 (hey folks! I'm an occasional volunteer dev who's gonna lurk for a meeting or two and see if anything pops out as a good fit. have done some things like webrtc, p2p, webdev in the past.) 17:11:07 i noticed the header change yesterday. let me check when i first deployed this 17:11:38 hi cjb o/ 17:11:44 welcome cjb! 17:11:47 i think i deployed it almost exactly a month ago 17:11:48 hey cjb! everyone should know cjb, they did the first version of a standalone webrtc client/server app long time ago. 17:11:59 phw: that's great, thanks. 17:12:00 oh nice 17:12:05 hi cjb! o/ 17:12:43 Hi cjb! 17:12:44 http://blog.printf.net/articles/2013/05/17/webrtc-without-a-signaling-server/ https://blog.printf.net/articles/2014/07/01/serverless-webrtc-continued/ 17:13:13 tyty! that was super long ago. 17:13:49 an eternity in internet years 17:14:04 seven eternities in internet dog years 17:15:04 another brief announcement: somebody created https://ntc.party, an internet censorship-focused forum. 17:15:09 lately I'm working at Keybase and trying to think about applying binary transparency to our updater, so happy to chat about things like that too. 17:15:10 ValdikSS, apparently 17:15:52 gaba: do you want to cover your items on our agenda? 17:16:00 yes 17:16:13 i added other announcement about where we are at with gitlab 17:16:28 and then for the discussion I want us to look at what we had planned in the roadmap for july and august 17:16:37 ValdikSS isn't just a somebody, they wrote GoodbyeDPI among other things. 17:16:40 and see what to do with the tickets letf from there 17:17:10 oops, i didn't know that dcf1. thanks for the context 17:18:13 from July we have #30830 17:19:16 gaba: are these on the board in gitlab? 17:19:25 mmm, I'm seeing that is not 17:19:48 I wonder why but it had the keyword 17:19:56 for july 17:19:59 that's what i've been using for choosing what to work on >.< 17:20:04 * gaba is going to look at the stuff from sweden 17:20:06 yes 17:20:09 ok 17:20:09 that makes sense 17:20:36 i can take care of that ticket, it should be pretty quick since we have metrics for measurements now 17:20:49 I will add it to the roadmap in dip 17:20:53 ok. thanks 17:21:49 any other orphaned tickets, gaba? 17:22:06 that is the only one for July 17:22:21 For August there are a few 17:22:33 https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=new&status=reopened&keywords=~anti-censorship-roadmap-august&col=id&col=summary&col=keywords&col=status&col=owner&col=type&col=priority&col=milestone&col=component&col=severity&col=resolution&col=actualpoints&col=points&col=reviewer&col=sponsor&order=status 17:22:52 #31153 is on ice 17:23:04 and they are in 'Next' column in dip roadmap 17:23:14 ok 17:23:16 #30986 is done by our race project partners, with our assistance 17:23:30 #29205 and #19001 are solved by #25483 which is in review now 17:23:49 #30716 is a ticket that's still ongoing and will likely require many more weeks of work 17:24:27 #29285 will also require much more work 17:25:24 #30368 we can probably close for now. we have a lot of measurements and found some interesting blocking behaviours 17:25:41 if we want to do more or different measurements we can re-open another ticket? 17:26:02 can we close #30986 ? 17:26:03 now that we have so many snowflakes we'd probably want to change how we do the measurements anyway 17:26:15 i'd say so 17:26:29 i'l; put it in review and let someone else second that 17:26:32 i don't want to close #30986 because we'll end up with some insight that we should share in the ticket 17:27:09 I will move it to backlog and have it to check again in October, ok? 17:27:12 okay we can leave it open indefinitely 17:27:23 i'd like to make another snowflake network health ticket which is somewhat related 17:27:24 ok, let's do that 17:27:37 and involves measurements but isn't quite the same 17:27:52 cool 17:29:39 i think that's all of them? 17:29:47 yes 17:29:48 thanks! 17:30:08 ok, shall we move on to reviews? 17:30:22 * phw wonders if hiro is here and needs some reviews 17:30:25 one more question, cohosh: you are closing #30368 ? 17:30:58 (sorry i do not understand what you said before about this one) 17:31:07 gaba: sorry, no we're leaving that one open 17:31:14 but we can put it on ice for now 17:31:23 it seems like a convenient place to log censorship events 17:31:36 and plan how to measure them 17:31:49 it will continue on your plate for now then 17:31:55 ok 17:32:14 ty 17:32:57 review tiiime 17:33:29 i just have #28942 17:33:30 hmm, sina mentioned that he already updated meek server but hasn't found the time to update #31455 17:34:10 cohosh: sorry for losing track of the pion progress, I didn't have any time last week and won't have much time next week either 17:34:21 dcf1: no worries 17:34:30 sorry I am here my screen was probably frozen 17:34:50 hi hiro! is there anything we can review for you? 17:35:03 This week I finished the monitoring part for gettor emails 17:35:17 since it also checks if links are delivered it can be also usesd for bridgedb 17:36:00 oh, that's very cool. did you do this in nagios? 17:36:05 yes 17:36:34 we even have an ancient ticket for this: 12802 17:36:39 #12802 17:37:01 well we will probably close this acient relict this week or the next 17:37:06 \o/ 17:37:14 that's great, thanks 17:37:54 serna: you wanted to discuss #25598, right? 17:38:08 Yeah 17:39:09 I'm having some trouble with the js proxies 17:40:05 Bc of the js event loop and I don't want to make a big change 17:40:15 if you have specific questions about the code, that might be a better discussion over in #tor-dev after the meeting 17:41:01 i will for sure be around and can take a look with you 17:42:02 Sure, as I said, the main problem is whit the js event loop and its asynchronicity 17:42:47 yeah i found it difficult to deal with that as well with the polling behaviour 17:42:54 cohosh: do you mind taking a look at my revisions for #31780? 17:42:56 i am still not sure if what we are doing there is good 17:43:14 phw: yes, added it to my todos for today 17:43:18 thanks 17:43:58 it looks like these were all the reviews. did we forget anyone? 17:44:10 * phw wonders if anything popped up as a good fit for cjb 17:44:52 cjb: since you have a background in webrtc, i definitely wouldn't mind a second look at the library switch we are contemplating :) 17:44:59 in ticket #28942 17:45:04 but no pressure 17:45:11 sure thing, will take a look 17:45:31 anything that doesn't ship half of Chromium is good :) 17:46:08 heh 17:46:39 that is part of the reason for the switch, we could not build and link the chromium library on windows 17:48:05 I see you already switched to pion-webrtc, is there another move after that being considered? 17:48:21 cjb: we haven't merged the branch yet 17:48:26 ah! ok 17:48:27 it's in review now 17:50:18 looks like a good idea, no specific experience with pion-webrtc. did you manage to get golang reproducible builds? 17:50:56 golang reproducibility has never been a problem, except in some cases with Cgo. 17:51:07 great 17:54:16 anything else to talk about? if not, let's wrap it up for today 17:54:59 #endmeeting