15:02:32 <richard> #startmeeting Tor Browser Weekly Meeting 2022-10-17
15:02:33 <MeetBot> Meeting started Mon Oct 17 15:02:32 2022 UTC.  The chair is richard. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:02:33 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:02:39 <richard> nailed it
15:02:55 <richard> pad per usual: https://pad.riseup.net/p/tor-tbb-keep
15:04:25 <donuts> o/
15:08:21 <boklm> hi
15:08:27 <richard> o/
15:08:39 <starlingroot[m]> o/
15:09:27 <richard> ok I think we're readyish
15:09:55 <richard> I've only 2 or 3 litle things beore we go to the bolded
15:10:55 <richard> 1. Let's try and go back to the olden days and fill out the pad before the meeting begins so we're not splitting our attention too much between irc and the pad
15:11:28 <richard> yolo'ing worked when it was just a few of us but surprisingly that doesn't really scale :p
15:12:43 <Jeremy_Rand_36C3[m]> works for me.  May have to retrain myself to remember to do it ahead of time, but should get the hang of it
15:12:55 <richard> 2. Outreachy candidates have arrived! If you see some new faces asking for help in IRC please be nice and welcoming and help as much as is reasonable
15:13:09 <richard> I mean do that anyways
15:13:33 <richard> but I've gotten around ~10 or so intro emails and they seem to continue to trickle in
15:13:53 <dan_b> oh wow
15:13:53 <Jeremy_Rand_36C3[m]> Indeed, very much enjoying the new faces on IRC
15:13:58 <PieroV> some are also in this meeting :)
15:14:04 <dan_b> how many are we expecting?
15:14:11 <dan_b> that's amazing
15:14:14 <richard> and to any Outreachy people lurking in the chat, do be kind and helpful to the other candidates as well :)
15:14:45 <richard> dan_B honestly i had expected a handful
15:15:06 <richard> but I've never done this before and we seem to have garnered a fair bit of interest somehow vOv
15:15:22 <Jeremy_Rand_36C3[m]> I've only gotten circa 3-4 candidates for the project I'm mentoring, but I'm not complaining, I was terrified originally that there would be zero :)
15:15:30 <richard> sooo we'll see how it goes
15:16:31 <richard> but yeah, be nice but don't bend over backwards
15:17:20 <Jeremy_Rand_36C3[m]> (speaking of new faces, apologies if I'm oblivious but should I know who starlingroot is?)
15:17:34 <PieroV> (they're an outreachy candidate)
15:17:35 <richard> 3. ok and the only other item which I'm probably woefully behind on is that ux had some tweaks to make for the YEC campaign desigs, so let's be sure those get in with the next release scheudled for later this week
15:17:50 <Jeremy_Rand_36C3[m]> Ah cool!  Welcome starlingroot !
15:18:01 <richard> (I think I've already seen MRs for this come in but just in case)
15:18:01 <starlingroot[m]> Thank you Jeremy
15:18:06 <richard> o/
15:18:14 <dan_b> android was a NOP - blend mode wasn't added till API 29, we target 21
15:18:26 <PieroV> dan_b: not sure I've already told you
15:18:36 <PieroV> But Android break lines in Italian with my old test device
15:18:52 <PieroV> However now that we have translations we can check them easily :)
15:19:06 <dan_b> sure, on "powered by privacy"? on two lines? privacy on another?
15:19:12 <dan_b> it was doing the same in english on my pixel 4
15:19:19 <dan_b> seemed ok with nicob
15:19:25 <dan_b> or is this something worse?
15:19:30 <PieroV> (btw, seeing a different screen is somehow strange :))
15:19:41 <PieroV> Nope, It's "cambiamento" (= change)
15:19:44 <dan_b> (haha :D)
15:19:53 <PieroV> It's broken cambiamen-to (at least grammatically correct :D)
15:20:22 <PieroV> And $100.000 goes in a new line and isn't completely visible
15:20:31 <dan_b> huh... yeah can you send a screen shot when you get a chance in #tor-browser-dev
15:20:37 <dan_b> oph
15:20:38 <PieroV> (for some reason it isn't scrolling, and we could reduce the margin)
15:20:40 <PieroV> sure
15:21:27 <richard> ok pierov looks like the next bullet points are you :)
15:21:43 <PieroV> Yes, basically what's already written :)
15:21:57 <PieroV> Long story is that we've finally merged the mozconfigs between tor-browser.git and tor-browser-build.git
15:22:20 <PieroV> However, Android is 4 mozconfig + 1, so we should've done 5 mozconfig only for dev
15:22:51 <PieroV> Since we already needed to change a few settings (path to the JDK, if different from OpenJDK 11 on Debian, path to the Android SDK/NDK/Gradle)
15:23:02 <PieroV> I've decided to use the same variable that tor-browser-build uses
15:23:18 <PieroV> After I think we all change $PATH before building for Android
15:23:39 <PieroV> So, my suggestion is to create a shell file with these settings to source before starting building for Android
15:23:45 <richard> exciting
15:23:48 <PieroV> And if next local builds fail that might be the reason
15:24:28 <PieroV> That's it for this point. If you have ideas on how to improve the situation just tell me :) (e.g., add a series of if, etc etc etc)
15:25:51 <PieroV> Now, for the next point
15:26:26 <PieroV> 12.0a3 still has Firefox here and there. I think the reason is that the branding is implemented in Fluent
15:26:47 <PieroV> Fluent was implemented in JS in 91, but it's a Rust native vendered lib in 102. So, I think things changed while changing the implementation
15:27:18 <PieroV> It seems packaged locales solve the issue. Are we considering it a blocker for 12.0a4?
15:27:43 <Jeremy_Rand_36C3[m]> PieroV: relatedly, Tor Browser Nightly's taskbar group (on Linux KDE) identifies itself as "firefox-nightly" at the moment
15:27:52 <Jeremy_Rand_36C3[m]> Not sure if that's the same bug
15:28:48 <Jeremy_Rand_36C3[m]> (happy to take a screenshot if you think that's a previously unreported bug)
15:28:51 <richard> what are the implications of this issue w/ regard to multi-locale bundles?
15:29:03 <PieroV> Initially I thought it was required
15:29:33 <PieroV> But things seem to keep working, since we're still using old properties and dtds
15:29:54 <PieroV> Henry investigated this and wrote a few comments in tor-browser!374
15:30:27 <PieroV> (currently Fluent is blocked on Weblate not supporting all its features, but the branding was already in Fluent in 91)
15:30:46 <richard> our branding was?
15:31:07 <PieroV> We have a few lines in import-translations.sh to handle it
15:32:34 <PieroV> (not sure it answers your question)
15:33:02 <richard> ok, I think I would rate fixing the branding as lower priority than multi-locale bundles fwiw, and I don't think we should block 12.0a4 on it unless y'all have a good reason for me
15:34:09 <Jeremy_Rand_36C3[m]> Do any of the branding issues expose us to trademark violation liability?
15:34:52 <Jeremy_Rand_36C3[m]> (I realize Mozilla is probably not interested in suing us, but still maybe worth keeping in mind?)
15:35:09 <richard> I am not a lawyer!
15:35:10 <PieroV> I didn't think of that
15:35:25 <richard> but it would def block stable
15:35:41 <PieroV> I was more worried about getting the higher number of fixes already (since as richard just said, it's a blocker to stable for sure)
15:36:06 <richard> in the past we've had various branding leak issues in the past and so far have not been sued vOv
15:36:21 <richard> but anyway
15:37:06 <Jeremy_Rand_36C3[m]> PieroV: is the taskbar bug I mentioned the same set of branding issues you're talking about or have I stumbled onto a new one?
15:37:18 <PieroV> Jeremy_Rand_36C3[m]: I have no clue
15:37:34 <PieroV> I don't even know how it works, never used KDE :D
15:37:36 <Jeremy_Rand_36C3[m]> PieroV: OK, in that case maybe I'll post a screenshot onto #tor-browser-dev after the meeting
15:37:42 <PieroV> Ok, thanks!
15:37:51 <PieroV> Jeremy_Rand_36C3[m]: do you happened to know which API could set that, too?
15:38:08 <Jeremy_Rand_36C3[m]> PieroV: I don't know :(
15:38:16 <PieroV> ok never mind, thanks anyway :)
15:39:21 <richard> ok then
15:39:26 <richard> dan: you still need git help?
15:39:45 <dan_b> just the question,
15:40:03 <richard> just push -f over it :D
15:40:05 <dan_b> for 41125 i previsously had a commit that was accepted, but not want to replace it
15:40:06 * Jeremy_Rand_36C3[m] uploaded an image: (20KiB) < https://matrix.org/_matrix/media/r0/download/matrix.org/ezuvRcqwyziMrRXdHAgibmVF/Tor%20Browser%20branding%20bug%202022%2010%2017.png >
15:40:10 <dan_b> ah ok
15:40:12 <dan_b> on my branch
15:40:12 <richard> unless you mean *actually* delete
15:40:15 <Jeremy_Rand_36C3[m]> PieroV: that's a screenshot ^
15:40:24 <dan_b> yeah that commit needs to be removed
15:40:29 <Jeremy_Rand_36C3[m]> (got it faster than expecting)
15:40:35 <dan_b> so the uplift can be pulled in to replace
15:40:39 <PieroV> there are a few additional changes in the MR
15:40:52 <richard> oh
15:40:55 <dan_b> yeah but those are being dropped
15:40:55 <richard> ok
15:40:58 <Jeremy_Rand_36C3[m]> (I'll post in #tor-browser-dev after the meeting too)
15:40:59 <PieroV> Like the HTTP things (that we don't need, in my opinion)
15:41:03 <dan_b> and replaced with fixes in different spots
15:41:04 <richard> so make a new MR for it
15:41:23 <richard> with a revert commit, and then a cherry-pick of the official mozilla one
15:41:59 <richard> and we'll do the rebase/commit shuffle as part of the next rebase
15:42:35 <PieroV> richard: the original MR also has a pref lock
15:42:37 <dan_b> ok cool
15:43:06 <dan_b> and PieroV i think you might be right. it was the moz recommend but with the other prefs I think its prolly superfulous too, I'll take one last look and prolly let it go
15:43:11 <PieroV> But I guess that commit can be cherry-picked in the new branch
15:43:15 <richard> (in the case the pref lock should be pulled out and re-implemented as a fixup to the appropriate pref commit)
15:43:47 <dan_b> yep, that already is
15:44:05 <dan_b> do we prefer muyltiple MRs for this or one MR with several commits fixing up different things and the revert
15:44:24 <richard> I'm fine with one MR
15:44:33 <dan_b> cool
15:44:38 <dan_b> thanks I think I'm good
15:44:46 <richard> probably create a new issue for it "swap out fix bug XXX with upstream mozilla fix for bug YYY"
15:44:59 <richard> but grammatically correct/etc
15:45:03 <dan_b> hahaha
15:45:11 <dan_b> 👍
15:45:22 <richard> ok anything else?
15:45:46 <PieroV> No from me
15:46:00 <boklm> not from me
15:46:25 <ma1> neither for me
15:47:08 <richard> ok then
15:47:19 <richard> have a good week everyone!
15:47:20 <richard> #endmeeting