19:00:09 <GeKo> #startmeeting tor browser
19:00:09 <MeetBot> Meeting started Mon Jan 22 19:00:09 2018 UTC.  The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:00:09 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
19:00:13 <isis> (i think they also mean TB team)
19:00:13 <GeKo> hi all!
19:00:15 <isis> oops
19:00:20 <boklm> hi!
19:00:30 * tjr waves
19:00:32 <igt0> o/
19:00:41 <sysrqb> hello
19:00:47 <mcs> hi
19:01:11 <GeKo> okay, let's get started
19:01:26 <GeKo> if you have not filled out the pad, please do now
19:01:32 <GeKo> it is https://storm.torproject.org/shared/tHoN4Ii7rLSjPE0OP4gydX4cMGadsXmRQNc-6lwru0N
19:01:52 <isabela> oh the pad!~
19:01:54 <GeKo> all other folks please read meanwhile the updates and mark things bold you want to tlak about
19:03:20 <igt0> sysrqb, about the prefs, are you talking about the extension prefs the default preferences in the extension? (I am asking, because I was struggling with the same thing)
19:05:30 <sysrqb> igt0: browser prefs, specifically overriding mobile.js
19:06:22 <sysrqb> Orfox directly modifies mobile.js, I think we should use something like the desktop, where we have 000-tor-broserw.js
19:06:33 <sysrqb> err, 000-tor-browser.js
19:06:52 <GeKo> alright
19:07:11 <GeKo> let's go down the marked items
19:07:21 <GeKo> tjr: you have the mic
19:07:22 <sysrqb> I have a patch and (somehow) it works, but one pref is not overridden :/
19:07:45 <arthuredelstein> sysrqb: Is it possible to use 000-tor-browser.js as the baseline with extra prefs on top in another file as needed?
19:07:55 <tjr> I was wondering how the way we do the pref files interacts with code in Firefox that looks to see if there is a user-defined value for a pref
19:08:17 <tjr> And if Tor Browser-set prefs should be considred 'user defined' or not!
19:08:27 <tjr> (I don't know the answer to either)
19:08:43 <sysrqb> arthuredelstein: already do. I have a #inlcude and then override the rest
19:08:54 <arthuredelstein> sysrqb: cool!
19:09:26 <GeKo> tjr: do you want to talk a bit about your item?
19:09:27 <sysrqb> yeah, seemed like the simplest and least error prone
19:09:32 <GeKo> or should we just jump in?
19:09:52 <tjr> If you meane the pref stuff I was just musing out loud, don't have anything more to say on it
19:10:05 <tjr> If you mean the stuff I bolded under my name; then...
19:10:14 <GeKo> yes that one
19:10:22 <sysrqb> tjr: to answer your question
19:10:23 <sysrqb> and based on my reading of the code over the last day
19:10:25 <tjr> Mostly I'm just putting out a call for some way to tell me what you all think is important to get into 60
19:10:26 <GeKo> (the bolded stuff)
19:10:31 <sysrqb> it reads it as a 'pref'
19:10:37 <tjr> And if I haven't mentioned it yet to make sure it's on my radar
19:10:56 <sysrqb> (nt sticky or user)
19:11:01 <sysrqb> *not
19:11:07 <arthuredelstein> So I've been looking at that and I set a couple of bugs to P1, in addition to the existing P1 bugs we had for fingerprinting
19:11:17 <arthuredelstein> Here's the list I have right now: https://mzl.la/2E1eVn2
19:11:26 <GeKo> tjr: your list looks good to me
19:11:47 <GeKo> arthuredelstein: how does that map to our patches we have in tor-browser?
19:11:54 <tjr> arthuredelstein: Yea, I just saw that, thanks for that
19:12:29 <arthuredelstein> Most of them have direct mappings; I think a couple are fixups to existing uplifted patches
19:12:32 <tjr> The font one seems like a very tough one to close out. Do you know the latest status on that? I never followed it very closely.
19:12:58 <arthuredelstein> I think it's pretty close, and Tim was waiting for he kinto server to be ready
19:13:05 <arthuredelstein> There are some fixups that need to be done
19:13:13 <tjr> date one is easy, update channel is easy. The Web UUID one is tough....  but that one I don't think is ours to work on?
19:13:22 <GeKo> arthuredelstein: the fix for #23044?
19:13:44 <tjr> window.name is trying to land tor's patch from scratch right? We haven't rebased or had it reviewd yet?
19:13:49 <tjr> jkt asked me about that this morning actually
19:14:26 <arthuredelstein> GeKo: Yeah, actually I wanted to ask you about that because I'm not sure which parts we should try to uplift
19:14:35 <tjr> 1330890 is a blocker for a diff one and needs no work IIRC
19:14:37 <GeKo> or the onr for #21321
19:14:41 <tjr> 1330882 I'm not sure of the status of....
19:14:52 <GeKo> (i mean to contribute the missing part to that one)
19:15:12 <arthuredelstein> tjr: I would propose making a FPI patch for window.name
19:15:27 <tjr> Ah okay...
19:16:11 <GeKo> arthuredelstein: especially the fix for #19273
19:16:25 <GeKo> is pretty important
19:16:34 <GeKo> what is up with that one?
19:17:21 <arthuredelstein> #19273 = https://bugzilla.mozilla.org/show_bug.cgi?id=1305177
19:18:25 <arthuredelstein> I didn't put it as P1 (this list) but we can do that. Basically I was trying to gauge what's doable in the next 6 weeks or so
19:18:54 <tjr> If we want it for 60 I think we should put it as P1 and re-evaluate in 3-4 weeks if it's not going to land
19:19:12 <tjr> And assign it to either me or you depending on who you think is going to do it :)
19:19:20 <arthuredelstein> OK!
19:19:27 <GeKo> yes, please
19:19:31 <boklm> tjr: regarding https://gitweb.torproject.org/tor-browser.git/patch/?id=ba9e5d6 it is from https://trac.torproject.org/projects/tor/ticket/24197#comment:1 and I think it is already in mozilla-central
19:19:40 <tjr> There's a couple people in webextensions (not McKay, but others) who have been very helpful so they may be able to help us with the webext thing mentioned
19:19:49 <GeKo> arthuredelstein: and pospeselr could be helping you as well
19:20:00 <GeKo> so, we should be able to get quite some patches upstreamed
19:20:05 <tjr> boklm: great, thanks
19:20:08 <pospeselr> aye aye
19:20:14 <arthuredelstein> OK! I can add quite a few to this P1 list.
19:20:27 <GeKo> thanks.
19:23:07 <GeKo> arthuredelstein: there is #21569 as well
19:23:24 <GeKo> so there seems patches we have missing on your list
19:23:35 <arthuredelstein> There are many missing; these are only P1
19:23:52 <GeKo> who decided that?
19:24:19 <GeKo> the idea was to have a list of all patches and then decide in this meeting which we want and which we did not want to prioritize
19:25:04 <arthuredelstein> Sorry -- I thought you had asked me to triage them.
19:25:30 <tjr> I think (hope) that anything we may care about is somewhere in this list:     https://bugzilla.mozilla.org/buglist.cgi?quicksearch=sw%3A%5Btor%20OR%20fingerprinting&list_id=13978086
19:25:34 <GeKo> oh, sorry for that misunderstanding.
19:25:46 <arthuredelstein> But we can still do that in this meeting. I will paste the list I have.
19:25:52 <GeKo> no, i thought we should have a list based on our patches in tor-browser.git
19:26:07 <arthuredelstein> Right, please give me a couple minutes to get it in a readable state
19:26:09 <GeKo> and then deciding which out of those we want to get upstreamed
19:26:16 <GeKo> no worries
19:26:36 <GeKo> mcs: hi!
19:27:05 <GeKo> i've uploaded the signed mar files to the builds dir on people.tpo
19:27:14 <GeKo> (i am almost done with the signing)
19:27:24 <GeKo> could you test those like today and check whether they work
19:27:25 <GeKo> ?
19:27:33 <GeKo> in particular on macOS?
19:27:46 <GeKo> for the alpha we are using the backup signing key after rotation
19:28:02 <mcs> GeKo: Yes., some testing is a good idea :)
19:28:15 <GeKo> so, while i tested that myself on linux having someone else doing the testing would be good
19:28:23 <GeKo> 7.5-build3 and 8.0a1-build3 are the dirs
19:28:49 <GeKo> thanks
19:28:56 <GeKo> let me know if there are any issues.
19:29:00 <mcs> Will do.
19:29:06 <GeKo> or better the tor-qa list
19:29:09 <GeKo> cool
19:30:23 <GeKo> igt0: i am curious
19:30:39 <GeKo> did you check whether the desktop browser still worked after moving torbutton?
19:31:01 <GeKo> i guess that's the first stage of arthur's three stage proposal
19:31:29 <GeKo> isabela: yeah, meeting some time next week for the circiot display thing is fine with me
19:31:35 <GeKo> arthuredelstein: ^
19:31:48 <isabela> you know
19:31:57 <isabela> if we have to do it this wed is also good?
19:32:12 <isabela> i might be scheduling for a day antonela is flying :( sorry
19:32:16 <GeKo> could be!
19:32:27 <isabela> ok i will confirm via email :)
19:32:33 <GeKo> thanks
19:33:10 <igt0> yep, it is working. However after moving the default preferences didn't work, so I updated the moz.build and added it in the package-manifest.in.
19:33:27 <GeKo> neat.
19:33:47 <GeKo> fwiw. i am fine if we try to move this in the next alpha
19:34:00 <GeKo> so, if there is a patch  to review, let us know :)
19:34:24 <GeKo> sysrqb: interesting bugs you find :)
19:34:33 <GeKo> do we have a ticket on trac for the pref things?
19:34:48 <arthuredelstein> isabela: GeKo: I can do this Wed or next week
19:34:48 <sysrqb> GeKo: not yet. I'll open
19:35:01 <igt0> ah great, I will send it.
19:35:03 <GeKo> if not could you create one, so that the knowledge about it and what we plan to do to fix those is not lost?
19:35:09 <GeKo> sysrqb: ^
19:35:12 <GeKo> ah, okay, thanks
19:35:20 <isabela> arthuredelstein: ack
19:35:22 <sysrqb> but yes, ...interesting
19:35:30 <GeKo> and scary
19:35:45 <sysrqb> yes
19:36:12 <GeKo> alright, discussion time i guess
19:36:34 <GeKo> could someone explain the first point?
19:36:51 <tjr> arthuredelstein: do you want to keep that meeting (
19:36:55 <tjr> at a different time)
19:37:01 <GeKo> which meeting?
19:37:11 <tjr> Arthur and I would meet with Ethan's team Wens nights
19:37:30 <arthuredelstein> tjr: Yeah, I think that would be useful. Would be great if we can find a time pospeselr can attend as well
19:37:30 <GeKo> i see
19:37:48 <tjr> Yea, most times between 10 and 5 Central time will work for me, with a few exceptions
19:38:06 <pospeselr> when is the meeting scheduled?
19:38:11 <arthuredelstein> Yup, me too, but Pacific time.
19:38:42 <tjr> Tuesday at this meeting's time?
19:39:21 <arthuredelstein> Yeah, that would work for me.
19:39:40 <pospeselr> ok, i can't do anything from Wednesday on (unless it's in the evening pacific) as I have jury duty
19:39:53 <pospeselr> so tuesday should work
19:40:25 <tjr> Cool, done, we'll save the time and figure out the arrangements 5-10 min before this time tomorrow
19:40:32 <arthuredelstein> pospeselr: It's a weekly meeting just in case that wasn't clear
19:40:49 <arthuredelstein> tjr: Sounds good
19:41:06 <pospeselr> ah ok
19:41:11 <pospeselr> will add to my calendar
19:41:51 <GeKo> thanks
19:41:52 <arthuredelstein> So here's my full list of patches that haven't already been uplifted and could conceivable be uplifted: https://gist.github.com/arthuredelstein/c06d33cbd306fc1d7671c433378eee01
19:42:05 <GeKo> so, the GSOC thing
19:42:55 <arthuredelstein> (For my P1s, I had tried to pick a few that seemed doable before Firefox 60 beta, but we can definitely add more)
19:42:58 <GeKo> tjr: so, you think we should have a crashreporter project 2?
19:43:15 <GeKo> what's the scope of that one?
19:43:18 <tjr> GeKo: Yea, I think with more effort we could finish it up and get it in a shippable state
19:43:30 <GeKo> ok. fine with me.
19:43:43 <tjr> Fix the proxying of the submission, get it building reproducibly on win/mac, and tweak/re-review the data we're submitting
19:44:00 <GeKo> yeah, great
19:44:21 <GeKo> do we have any other ideas for GSoC? and folks willing to mentor?
19:45:31 <arthuredelstein> I was thinking it would be nice to have something that removes EXIF data from uploaded images
19:45:47 <arthuredelstein> Maybe that's a GSoC-worthy project?
19:45:57 <pospeselr> arthuredelstein: now that is an awesome idea
19:46:14 <tjr> That is cool, yea
19:47:23 <sysrqb> yeah, +1
19:47:25 <arthuredelstein> I think I'd be willing to mentor it but I'm not really familiar with how mentoring GSoC works.
19:48:15 <arthuredelstein> Although maybe it's especially a good thing to do on mobile
19:49:08 <arthuredelstein> Where I probably wouldn't be useful as a mentor
19:49:08 <GeKo> arthuredelstein: i guess a good start could b asking colin
19:49:14 <GeKo> about deadlines etc.
19:49:22 <GeKo> and then take it from there
19:49:37 <arthuredelstein> OK!
19:49:58 <sysrqb> arthuredelstein: yeah. usually GSOC require two menotrs, main and backup
19:50:22 <sysrqb> I cn help you with that on th emobile siade
19:50:33 <sysrqb> ugh, lag.
19:50:39 <arthuredelstein> Thanks, that would be great. I'll email colin and cc you.
19:50:45 <sysrqb> sgtm!
19:50:55 <GeKo> arthuredelstein: so, what does "?", "C", and "F" mean? on your updated patch list?
19:51:59 <GeKo> so, just skimming the list
19:52:12 <GeKo> i think we want the fix for #23016
19:52:51 <pospeselr> for uplift?
19:52:57 <GeKo> yes
19:53:01 <arthuredelstein> pospeselr: Right, would be great if you want to work on that
19:53:27 <GeKo> the fix for #5282 can get ignored as the whole pipelining code is gone
19:53:27 <pospeselr> looked into that already
19:53:32 <pospeselr> doesn't need uplift, as the 'broken' component no longer exists
19:53:47 <GeKo> great.
19:54:01 <arthuredelstein> GeKo: Those are codes related to my rebasing work and not relevant
19:54:57 <arthuredelstein> pospeselr: So I guess 23016 won't need to be in the ESR60-based version of TBB, right?
19:56:06 <pospeselr> right
19:56:15 <GeKo> ok, so then going over the list takes longer
19:57:05 <GeKo> but it seems to me starting with the usual things like uplifting our newly made fingerprinting and linkability patches is fair game
19:57:08 <GeKo> and important
19:57:21 <GeKo> plus the patch(es) for fixing #19273
19:57:35 <GeKo> that bug caused quite some pain for mcs/brade
19:58:24 <GeKo> i guess that could be a good start to work on things, as time is running short
19:58:47 <GeKo> does that make sense?
19:59:32 <arthuredelstein> I think so. What should we do about triaging this list?
20:00:15 <arthuredelstein> Basically the Tor uplift team was focused on trying to get parity with TBB fingerprinting defenses
20:00:34 <arthuredelstein> So that was where the RFP bugs in that P1 list came from
20:01:20 <arthuredelstein> I think that's actually quite a good goal to have for ESR60, and then as you say, we can try to add more linkability/fingerprinting bugs.
20:01:32 <GeKo> okay. but https://bugzilla.mozilla.org/show_bug.cgi?id=1303456 ?
20:02:46 <arthuredelstein> Yeah, I already added that to the list
20:03:29 <mcs> arthuredelstein: I think the followup fix for #19273 (along with some background) can be found in #22618
20:04:22 <GeKo> arthuredelstein: yeah, but that bugs is likely the reason for weird crashes we sometimes have
20:04:42 <GeKo> so, i actually pondering backing it out instead
20:05:18 <GeKo> *i am
20:05:30 <arthuredelstein> GeKo: oh, I had missed that
20:05:45 <arthuredelstein> now I see #19910
20:05:56 <GeKo> yep
20:06:35 <GeKo> so, i think if i should not be at that triage meeting which is perfectly fine to me
20:06:40 <arthuredelstein> OK, I'll remove that from our uplift batch for now
20:06:56 <GeKo> then a way forward would be to just start working on fingerprnting and linkability bugs
20:07:08 <GeKo> and have #19273 on the radar
20:07:22 <GeKo> and we determine the order and set of patches in the meeting next week
20:07:47 <GeKo> (the ones we should prioritize in the remaining time)
20:08:03 <GeKo> does that make sense?
20:08:44 <GeKo> meanwhile, do we have anything else for today?
20:09:09 <arthuredelstein> You want to do triage in the tbb-dev meeting next week?
20:09:42 <GeKo> well, i think we should work as fast as possible on upstreaming patches
20:09:43 <arthuredelstein> Or should we triage in the meeting tomorrow with tjr and pospeselr?
20:10:03 <GeKo> you can do the triage tomorrow that's fine
20:10:13 <GeKo> but we should look over the final result next monday
20:10:38 <arthuredelstein> OK, sounds good. I agree on making upstreaming a high priority now as well
20:10:51 <GeKo> okay! we made it it seems
20:10:58 <GeKo> sorry for the long meeting
20:11:12 <GeKo> thanks all *baf*
20:11:15 <GeKo> #endmeeting