15:00:23 <sysrqb> #startmeeting Tor Browser weekly meeting 23 August 2021
15:00:23 <MeetBot> Meeting started Mon Aug 23 15:00:23 2021 UTC.  The chair is sysrqb. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:23 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:00:31 <sysrqb> hello!
15:00:37 <Jeremy_Rand_Talos> Hi!
15:00:41 <sysrqb> Pad: https://pad.riseup.net/p/tor-tbb-keep
15:01:30 <richard> o/
15:03:39 <sysrqb> hrm. pad is slow for me today
15:04:04 <Jeremy_Rand_Talos> Same here
15:04:24 <Jeremy_Rand_Talos> Not loading at all in fact, been waiting ~5 mins
15:04:58 <sysrqb> hopefully we didn't break it eith everyone loading for this meeting
15:05:18 <sysrqb> thundering herd
15:05:40 <Jeremy_Rand_Talos> The only thing I was going to add to the pad this week was another polite nudge regarding the GitLab issue I asked about last week
15:05:53 <sysrqb> yeeeah
15:05:56 <sysrqb> please keep nudging
15:06:17 <sysrqb> the 91esr transition has consumed me
15:06:26 <sysrqb> sorry i dropped that last week, despite promising otherwise
15:06:49 * Jeremy_Rand_Talos is closing my browser tab for the pad, so hopefully that'll ease load on the server; no need for me to update the pad this week since nothing else is new here
15:07:36 <sysrqb> huh. "[notice] Have tried resolving or connecting to address '[scrubbed]' at 3 different places. Giving up."
15:08:16 <sysrqb> in any case, I don't have much to discuss this week
15:08:33 <sysrqb> i have tor-launcher 95% converted to async (i think)
15:08:48 <sysrqb> for as much as that is needed to bootstrap
15:08:52 <Jeremy_Rand_Talos> sysrqb, I mean, this is par for the course since I'm the one being paid to care about that issue, not you.  So yeah, I can wait if you're occupied.  :)
15:09:06 <richard> wow nice :D
15:09:28 <sysrqb> Jeremy_Rand_Talos: it's rude of me to continue dropping it though :)
15:09:51 <sysrqb> richard: yeah, it's not the prettiest
15:10:25 <sysrqb> but i'm currently adding torbutton's tor-control-port module as a dependency, and reusing that implementation
15:10:45 <sysrqb> instead of trying to reimplement it (again) within tor-laucnher's custom controller code
15:10:46 <richard> that sounds like a good plan
15:11:03 <richard> i was a bit shocked to discover both torbutton and torlauncher had their own control port logic >:[
15:11:13 <GeKo> i feel there is even a ticket for something like that...
15:11:13 <richard> so hurray for picking one and going with it
15:11:15 <GeKo> iirc
15:11:16 <sysrqb> but they don't exactly fit together, so it's requiring a little re-shaping
15:11:27 <richard> fun fun (?)
15:11:35 <sysrqb> ...yeah :)
15:11:51 <sysrqb> but this is a blocker for 91esr, so it's gotta be done
15:12:15 <sysrqb> GeKo: i remember seeing one of those
15:12:23 <Jeremy_Rand_Talos> richard, seems like maybe the control port lib should be split out into its own project so other people can reuse it, like Stem, Bulb, etc?
15:12:39 <sysrqb> tor-launcher even has a few comments of "this was copied from torbutton"
15:12:40 <GeKo> but gitlab's search is... not its strongest feature it seems
15:12:46 <sysrqb> :)
15:13:23 <richard> Jeremy_Rand_Talos: vOv
15:13:24 <sysrqb> Jeremy_Rand_Talos: i'd probably push-back on making a javascript implementation the common controller library
15:14:10 <sysrqb> but we should get a better interface in arti, which is yet-to-be-designed
15:14:11 <richard> yeah i also suspect the implementatio nis v dependent on Mozilla XPCOM JS type projections
15:14:23 <richard> yeah also arti
15:14:46 <Jeremy_Rand_Talos> Oh cool, Arti is swapping out the control API?
15:14:52 <sysrqb> yeah, thisone is dependent on Firefox internals for socket-types, threading, etc.
15:15:00 <richard> long term yeah
15:15:19 <sysrqb> the plan is a complete redesign
15:15:46 <Jeremy_Rand_Talos> Nice.  The control API very visibly was designed long ago, would be cool to eject all that technical debt
15:15:49 <sysrqb> better/easier than a bespoke ascii protocol
15:16:00 <sysrqb> yeah
15:18:10 <sysrqb> richard: you're back online on thursday?
15:18:27 <richard> i *should* be online Wednesday afternoon
15:18:33 <richard> barring catastrophe
15:19:06 <sysrqb> okay, i'll ignore that possibility in that case
15:19:40 <sysrqb> can you look over Moz92 tickets when you're back?
15:19:53 <sysrqb> i think i still need to open a ticket
15:19:59 <richard> yeah definitely
15:20:04 <sysrqb> but i'll put in details and assign it to you?
15:20:08 <richard> perfect
15:20:12 <sysrqb> thanks
15:21:07 <sysrqb> i'm going to try building Fenix91 using geckoview92, and we'll see how this experiement goes
15:21:21 <sysrqb> and if it's actually worth experiementing
15:22:29 <sysrqb> hrm. i thought boklm was only afk for part of this week.
15:23:09 <sysrqb> GeKo: can i ask you for a tor-browser-build patch review in a couple days, if/when I get Android building
15:23:10 <GeKo> nope
15:23:17 <GeKo> yes
15:23:22 <sysrqb> thanks
15:23:41 <sysrqb> fun times.
15:23:53 <GeKo> what are out aarch64 plans for macos?
15:23:56 <GeKo> *our
15:24:31 <GeKo> because we need that at some point and that point might be closer than we expect
15:24:37 <sysrqb> GeKo: that would be nice to have
15:24:42 <sysrqb> yeah.
15:25:13 <sysrqb> i hope we can continue using the universal/combined binary right now
15:25:25 <sysrqb> but i haven't looked at it closely
15:25:40 <sysrqb> oh. hrm.
15:25:43 <GeKo> well, there is no combined one yet as there is no aarch64 one
15:25:48 <sysrqb> right, we don't ship that yet
15:25:52 <sysrqb> yeah.
15:26:05 <sysrqb> how's that working. mocOS has a compatibility layer?
15:26:10 <GeKo> yep
15:26:12 <sysrqb> *macOS
15:26:19 <GeKo> but that stops working at some point
15:26:39 <GeKo> because, you know, app devs are supposed to have upgraded :)
15:26:47 <GeKo> i thought it was 2 years
15:26:49 <GeKo> but not sure
15:26:59 <GeKo> and we already burnt like 1 year at least
15:28:08 <GeKo> anyway, nothing to decide now
15:28:26 <GeKo> but we should avoid the situation where we are like "shit, that thing stops working next month"
15:28:43 <sysrqb> yeah.
15:28:44 <GeKo> and we used to work on toolchains during esr transition
15:28:50 <sysrqb> https://www.notebookcheck.net/The-end-for-Intel-compatible-x86-apps-on-M1-Macs-macOS-11-3-could-drop-Rosetta-in-certain-countries.526072.0.html
15:28:55 <GeKo> and i thought i'd bring this up this time
15:28:55 <sysrqb> that could be super cool.
15:29:34 <sysrqb> this is good to bring up
15:29:35 <sysrqb> thanks
15:31:06 * sysrqb updates tor-browser-build#40158
15:31:11 <sysrqb> anything else?
15:32:09 * sysrqb hears nothing
15:32:20 <sysrqb> thanks everyone, have a good week
15:32:24 <sysrqb> #endmeeting