15:57:57 <meskio> #startmeeting tor anti-censorship meeting
15:57:57 <MeetBot> Meeting started Thu Jul 27 15:57:57 2023 UTC.  The chair is meskio. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:57:57 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:58:01 <meskio> hello everybody!!!
15:58:04 <meskio> here is our meeting pad: https://pad.riseup.net/p/tor-anti-censorship-keep
15:58:08 <shelikhoow> Hi~
15:58:11 <meskio> feel free to add what you've been working on and put items on the agenda
16:00:04 <meskio> I have deleted all the old points, I think they don't need talking, but let me know if I should bring them back
16:00:05 <onyinyang[m]> hello!
16:02:09 <meskio> I like the announcements, many winners around :D
16:02:27 <meskio> congrats shelikhoow and onyinyang[m]
16:02:57 <dcf1> I hope I did not forget any
16:03:22 <meskio> those are the ones I remember
16:04:23 <meskio> I see two points for discussion
16:04:35 <shelikhoo> yes, thanks!
16:04:48 <meskio> I hope they are not from a previous meeting I missed :S
16:04:57 <meskio> "does this issue need attention"
16:04:57 <dcf1> they are new
16:05:03 <meskio> https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snowflake/-/issues/40278
16:05:08 <meskio> dcf1: thanks
16:05:23 <dcf1> on the first point about Android, cohosh left some comments saying that there was a similar problem when integrating with Shadow
16:05:49 <dcf1> I don't know the details, but it looks like we may need an upstream change in Pion
16:07:00 <meskio> should we check with the guardian project to see their thoughts? AFAIK they are our main android user currently
16:07:19 <dcf1> I don't know if this is a problem for us yet? The issue says Android 11, I think Android 13 is current? Is there a reason we haven't received many more such reports?
16:08:16 <meskio> good question, could it be that the guardian project is getting them and we don't hear about it? or that is not affecting much users?
16:08:34 <shelikhoo> I think the reason is that unless you target a platform higher than 11, this restriction won't apply
16:09:24 <shelikhoo> but google is going to force a target of 11 or higher soon
16:09:40 <shelikhoo> so that is the reason this have been given more attention lately
16:10:02 <dcf1> shelikhoo: do you know when google is going to start enforcing that change? we could make it a deadline in the issue.
16:11:30 <shelikhoo> https://support.google.com/googleplay/android-developer/answer/11926878
16:11:50 <dcf1> I found https://github.com/pion/transport/issues/228 in pion that links to the same go issue, closed as WONTFIX
16:11:56 <shelikhoo> If it is going to be a problem, then it will be end of Aug
16:13:06 <dcf1> Then maybe we should apply cohosh's workaround sonner than later? https://gitlab.torproject.org/cohosh/snowflake/-/commit/5bafffd47292327e39dff958b47952c8e300dbed
16:13:32 <dcf1> Or maybe this is something that will magically be fixed by updating some dependencies?
16:14:43 <shelikhoo> I think cohosh's patch might just work. There are other patch that try to workaround this: https://github.com/golang/go/issues/40569#issuecomment-1602644926
16:15:22 <meskio> in theory we have the latest version fo each dependency, I did connect renovate to the repo
16:15:26 <shelikhoo> we could reproduce error, and try all the solutions and see if any of fixes works
16:15:33 <meskio> but I wonder if it worked as I haven't seen any mr by it
16:16:34 <dcf1> okay so rough plan of action:
16:16:48 <dcf1> 1. cc guardian project to see if they have any input
16:17:04 <dcf1> 2. apply cohosh's workaround patch in lieu of any better solution
16:17:05 <dcf1> EOL
16:17:07 <dcf1> ?
16:17:38 <shelikhoo> 1.5 is to reproduce the error, yes
16:19:15 <shelikhoo> there are other patches we could try in the go issue like: https://github.com/golang/go/issues/40569#issuecomment-1605814421
16:19:39 <shelikhoo> but we could just try them until find one works
16:19:53 <meskio> sounds good to me
16:20:56 <meskio> I'll poke the guardian project directly in the issue
16:21:14 <meskio> shelikhoo: can you take care of 2 or 1.5?
16:21:29 <shelikhoo> yes! i will
16:22:08 <meskio> great, anything else on this topic?
16:23:02 <meskio> ok, lets move ot the next one
16:23:12 <meskio> Publish advisory for accidentally collected unhashed proxy churn measurements?
16:23:19 <meskio> https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snowflake/-/issues/40278
16:23:25 <dcf1> I added this one
16:24:08 <dcf1> You can read more about it on the issue (which is currently confidential), but the short summary is that we accidentally collected a small number of proxy churn measurements wrongly
16:24:39 <dcf1> my suggestion is to make the 2 issues non-confidential and make a small post on the tor forum acknowledging it
16:25:01 <dcf1> it is very minor and not worth making a big deal about, but I feel things like that should not stay in a private issue forever
16:25:38 <meskio> mmm, I think the issue is the same than the prevous point
16:25:49 <dcf1> oops, one moment
16:25:52 <meskio> :)
16:26:00 <dcf1> https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snowflake/-/issues/40161
16:26:18 <dcf1> I blame it on zenbleed
16:26:35 <shelikhoo> I agree with these measures
16:26:38 <meskio> XD
16:27:11 <dcf1> I don't mind making the forum post, unless someone else wants to
16:27:37 <shelikhoo> I was thinking about avoiding alert fatigue, but a forum post should not be an issue
16:27:49 <meskio> I agree doesn't sound serious, but is good to make it public
16:27:55 <meskio> dcf1: I'm happy if you make the post :)
16:28:16 <dcf1> ok, I will take care of it
16:28:28 <meskio> thanks
16:29:33 <meskio> anything else on this topic?
16:29:34 <shelikhoo> thanks... sorry for making the mistake in the first place...
16:29:55 <dcf1> no apology necessary
16:30:11 <dcf1> we hope to cultivate an engineering culture, not a shame culture :)
16:30:22 <dcf1> when things happen we react appropriately
16:30:44 <shelikhoo> yes...
16:30:49 <dcf1> I also was not paying close enough attention when the issue was created
16:33:32 <meskio> I think we have finished with the discussion points
16:33:47 <meskio> dcf1: thanks for keeping posting the monthly reports in opencollective
16:33:59 <meskio> it looks like opencollective doesn't like me visiting from tor :S
16:34:19 <dcf1> yeah that started recently
16:34:47 <meskio> anything more for today?
16:35:33 <meskio> great, I think we are done for today then
16:35:52 <meskio> #endmeeting