18:29:45 <sysrqb> #startmeeting Tor Browser Team Meeting, 6 January 2020
18:29:45 <MeetBot> Meeting started Mon Jan  6 18:29:45 2020 UTC.  The chair is sysrqb. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:29:45 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:29:57 <sysrqb> Hello everyone!
18:30:10 <sysrqb> welcome to the first team meeting of the year 2020!
18:30:13 <antonela> hello!
18:30:14 <boklm> hi!
18:30:49 <mcs> happy 2020!
18:30:54 * sysrqb did not prepare as much as they hoped. updating pad now
18:31:01 <boklm> happy 2020!
18:31:36 <brade> o/
18:32:36 <pospeselr> hi y'all
18:34:54 <sysrqb> alright
18:34:59 <sysrqb> Happy New Year, everyone
18:35:15 <sysrqb> I'm excited for what is coming this year
18:35:43 <sysrqb> lots of changes, but I think the outcome will be worth it
18:36:34 <sysrqb> Okay, pospeselr, i'm a little concerned about the amount of time needed for implementing per-site settings
18:36:42 * antonela needs to go afk a bit early today -- will read backlog
18:36:48 <pospeselr> yeah same
18:37:03 <sysrqb> i wonder if we should start with having a conversation with ma1
18:37:10 <sysrqb> and get his inout on this
18:37:13 <sysrqb> *input
18:37:34 <pospeselr> do we have a way of flagging him down?
18:37:47 <pospeselr> i've tried email and adding them to the ticket :p
18:38:05 <sysrqb> relatedly, i've heard some people talking about umatrix and how it is very flexible in terms of providing per-site control by default
18:38:12 <sysrqb> so that is another option we can look at
18:38:27 <antonela> can you add me to that email thread pospeselr, sysrqb?
18:38:27 <sysrqb> i've never looked at the code, so i'm not saying we should jump to that option
18:38:38 <sysrqb> pospeselr: i think GeKo does
18:38:41 <sysrqb> i'll find out
18:38:51 <sysrqb> antonela: yup
18:39:04 <antonela> thanks!
18:39:09 <pospeselr> alright, i'll look into umatrix today/tomorrow and see if we can bend it to our will
18:39:26 <sysrqb> pospeselr: ah, you tried email. well, idk, i'll see what i can find out
18:39:37 <sysrqb> cool cool
18:40:14 <sysrqb> i do not have a good handle on these implementation, at that low level
18:40:37 <sysrqb> pospeselr: so if you look at umatrix, please document what you find, a bit
18:40:49 <pospeselr> so the main problem with no-script is that preferences are global and not isolated by first-party domain
18:41:09 <sysrqb> yeah, i remember reading that in the ticker
18:41:12 <sysrqb> *ticket
18:41:17 <pospeselr> so if you enable fb.com scripts on facebook it will be allowed everywhere else (if it's included as a tracker for instance)
18:41:21 <pospeselr> yeah ok
18:42:20 <sysrqb> are you planning on working on anything else this week?
18:42:45 <sysrqb> after i look at our tickets for this month, i may ask you to pick up some more work
18:42:58 <pospeselr> i don't have anything else high prioerity immediately popping up on my radar
18:43:00 <pospeselr> yeah please do
18:43:17 <sysrqb> alrighty
18:44:27 <sysrqb> i don't have any other comments right now, and a few people are still out on holiday
18:45:02 <sysrqb> i think we can move onto discussions, and i'll finish looking at our tickets for this month during this week
18:45:39 <sysrqb> so, who wrote those very thought provoking items?
18:46:05 <sysrqb> and what do you mean by "signing documents"?
18:46:26 <pospeselr> it was meee and obviously not serious ;)
18:46:30 <mcs> paper checks? contracts? (I did not write those things)
18:46:34 <pospeselr> like you know, you write a check and put 2019 instead of 2020
18:46:43 <brade> +1
18:46:43 <sysrqb> oooh
18:46:44 <pospeselr> until in mid-febrary/march
18:46:47 <sysrqb> like that.
18:46:56 <sysrqb> probably, yes.
18:47:12 <mcs> at least you did not include a “2020 vision” reference
18:47:16 <sysrqb> the changelog for 9.0.3 nearly said it was released on January 7, 2019
18:47:25 <pospeselr> hahaha amazing
18:47:36 <sysrqb> but then we tagged a new release with the correct date
18:47:47 <sysrqb> because dates are hard
18:47:56 <brade> Can we NOT discuss "when does the decade really begin"
18:48:15 <sysrqb> i'm okay with avoiding that question
18:48:32 <pospeselr> +1
18:49:13 <sysrqb> okay, anyone have anything else they'd like to add?
18:49:47 <sysrqb> oh pospeselr , how are your builds coming?
18:50:03 <pospeselr> assuming we're still doing build4, they are coming along
18:50:09 <pospeselr> looksl ike i'm on linux 64-bit now
18:50:11 <Jeremy_Rand_Talos> (sorry I'm late, had some KVM issues that needed a reboot right before meeting started...)
18:50:13 <sysrqb> that's still the one
18:50:18 <sysrqb> he Jeremy_Rand_Talos
18:50:21 <sysrqb> *hey
18:51:02 <sysrqb> pospeselr: okay. i saw boklm's builds already finished
18:51:03 <pospeselr> mozilla's FTP server was temporarily unreachable for me at some point over the weekend and my build stopped
18:51:14 <sysrqb> ooph
18:51:19 <sysrqb> okay
18:51:50 <sysrqb> hopefully we won't need to rebuild too many platforms for getting reproducibilitiy
18:52:25 <boklm> pospeselr: ah, I think the mozilla ftp error was an other issue, where firefox version had not been bumped (before -build4)
18:53:28 <sysrqb> Jeremy_Rand_Talos: congrats on getting the namecoin patches integrated
18:53:41 <sysrqb> it'll be interesting to see how this experiment goes
18:54:04 <Jeremy_Rand_Talos> sysrqb, thank you!  Should be exciting.
18:54:23 <sysrqb> okay, anything else before i close this meeting?
18:54:30 <Jeremy_Rand_Talos> sysrqb, At some point (doesn't have to be today) it would be useful to discuss what the next steps should be regarding Namecoin
18:55:05 <sysrqb> yep. i think we should let it bake a little first, and let's see if we get any external feedback on it
18:55:26 <sysrqb> but we should discuss where we should go with it in the near future
18:55:39 <pospeselr> boklm: ah ok that was during the build3 build
18:55:44 <Jeremy_Rand_Talos> sysrqb, yeah sounds good.  Would it be useful for me to start fiddling with porting to macOS and Windows?  Or should we wait before I spend time on that?
18:56:54 <sysrqb> Jeremy_Rand_Talos: i don't think you should spend much time on it, but if it's easy then we get integrate that, too
18:57:34 <sysrqb> but if it'll take another few weeks of time, then i think we can see what happens with the linux integration first
18:57:39 <Jeremy_Rand_Talos> sysrqb, hardest part of macOS/Windows integration is probably moving the launcher code to tor-launcher instead of the shell script where it currently lives.
18:58:16 <Jeremy_Rand_Talos> I'm happy to try doing it, but I haven't ever worked with tor-launcher before, so estimating how long it'll take won't be very accurate
18:59:04 <sysrqb> okay. then let's stay with what we have now, and see how that works
18:59:05 <pospeselr> fwiw (and correct me if I'm wrong here) I don't think we want *new* code in tor-launcher right?
18:59:25 <sysrqb> ideally, i would say no
18:59:49 <sysrqb> but we should start thinking about our plan for that, too
18:59:50 <Jeremy_Rand_Talos> sysrqb, ok, I'll take a break from this for a little while then, and let feedback trickle in.
18:59:57 <sysrqb> sounds good
18:59:58 <mcs> I think it is OK to put new code in Tor Launcher if it fits with what is there… but in this case, for something new, it probably makes sense to wait.
19:00:19 <sysrqb> +1 i agree with that
19:00:39 <Jeremy_Rand_Talos> So I get the impression from what you guys are saying, that tor-launcher is disliked and/or being phased out/replaced?
19:00:59 <sysrqb> the functionality is slowly being integrated directly into the browser
19:01:13 <sysrqb> so, eventually it won't be needed
19:01:43 <Jeremy_Rand_Talos> sysrqb, ah, I see.  Yeah, wherever the code lives that starts the tor daemon, seems like the place that should launch the Namecoin background processes too
19:02:11 <pospeselr> yeah that would probably make the most sense
19:02:13 <sysrqb> you're correct that currently it is tor-launcher
19:02:29 <mcs> How does the Namecoin process get killed or know to exit?
19:02:38 <sysrqb> but i don't have a timeline for when the integration will be complete
19:03:26 <Jeremy_Rand_Talos> mcs, right now, we just send it SIGTERM via the kill command after Firefox exits.  This is okay for a PoC but obviously I don't like that solution.  I don't know how the tor daemon is shut down currently.
19:03:31 <pospeselr> sysrqb: fwiw long-term I've really enjoyed this tor-launcher/tor-button integration work  ;)
19:03:56 <sysrqb> pospeselr: noted :)
19:04:22 <mcs> Jeremy_Rand_Talos: for desktop Tor Browser, Tor Launcher uses the owning controller feature to ensure that tor exits when the browser does.
19:04:58 <Jeremy_Rand_Talos> mcs, is there maybe a wiki page I could look for that would tell me more about how that works?
19:05:12 <Jeremy_Rand_Talos> Or am I just going to have to look at the tor-launcher source code?
19:05:40 <sysrqb> (which means when the controller's socket closes, the tor process sees this and shuts itself down and exits)
19:06:03 <sysrqb> (tor uses poll/epoll/select for socket events )
19:06:17 <mcs> Jeremy_Rand_Talos: probably you could look at the Tor control protocol spec or the Tor Launcher code
19:06:59 <Jeremy_Rand_Talos> mcs, ok thanks.  Sounds like there's no direct analogue for the Namecoin processes, so we may need to find a different approach for that.
19:07:00 <mcs> (looking at code) Tor Launcher passes an __OwningControllerProcess arg to tor that includes the browser pid
19:07:34 <mcs> Jeremy_Rand_Talos: agreed, a different approach may be needed but I am glad you are thinking about it :)
19:08:42 <Jeremy_Rand_Talos> sysrqb, btw as an aside, Namecoin will be applying for more funding from NLnet, probably sending in the submission on Feb 1.  If Tor can provide me with some (non-binding) rough guesses about what kinds of further work would be beneficial, I could make sure it's included in the funding proposal so that we don't have to worry about funding for those tasks
19:09:25 <sysrqb> Jeremy_Rand_Talos: understood
19:09:32 <sysrqb> we can followup on that
19:09:37 <Jeremy_Rand_Talos> cool
19:11:02 <sysrqb> okay, hearing nothing else, i'm going to close this meeting
19:11:15 <sysrqb> Happy 2020 everyone, have a good week
19:11:19 <Jeremy_Rand_Talos> same to you
19:11:22 <sysrqb> and I'm looking forward to everything we accomplish this year
19:11:33 <sysrqb> #endmeeting