18:29:35 <GeKo> #startmeeting tor-browser 2/11
18:29:35 <MeetBot> Meeting started Mon Feb 11 18:29:35 2019 UTC.  The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:29:35 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:29:44 <GeKo> hi all!
18:29:47 <sisbell> hello
18:29:50 <GeKo> let's do our weekly meeting
18:30:08 <mcs> hi
18:30:17 <GeKo> the pad is as usual at https://storm.torproject.org/shared/tHoN4Ii7rLSjPE0OP4gydX4cMGadsXmRQNc-6lwru0N
18:30:29 <GeKo> please enter your pieces and mark those bold we should talk about
18:33:00 <pili> hi
18:33:05 <pili> sorry, still getting used to new time :)
18:33:58 <GeKo> pili: you are not that late :)
18:34:13 <tjr> So perhaps I will go first since I have the first bold :)
18:34:20 <antonela> o/
18:34:22 <GeKo> please do
18:34:58 <sysrqb> o/
18:34:59 <tjr> GeKo: I don't see a way to do that. In Mozilla's build infrastructure, mozconfig changes like that occur well after the toolchain is built.
18:35:48 <tjr> In Tor's you could read the mozconfig ahead of time I suppose, and change the toolchain build (introducing a dependency for the toolchain on the mozconfig...)
18:36:17 <GeKo> hm
18:36:36 <tjr> But I can't really make that change for Mozilla's infrastructure.  I think I could make it so our debug builds used a different toolchain that included the flags and the opt build didn't...
18:36:50 <tjr> But then when you debugged the opt build you wouldn't have symbols for the runtime
18:37:21 <GeKo> i did not read https://hg.mozilla.org/releases/mozilla-esr60/rev/434bde49b9c8 as a toolchain changes
18:37:23 <tjr> It seems like there should be a way to generate a separate pdb for the runtime that you build every time but don't package....
18:37:51 <GeKo> but maybe i was wrong here
18:37:55 <tjr> Oh you mean those flags on the build, not the toolchain!
18:37:59 <GeKo> yes
18:38:04 <GeKo> those are breaking us
18:38:20 <GeKo> no worries about the toolchain you build
18:38:27 <tjr> How are they breaking things?
18:38:29 <GeKo> i took those flags out in our case
18:38:42 <GeKo> i can't link libxull due to out of memory errors
18:38:52 <GeKo> xul.dll fwiw
18:38:53 <tjr> Oh, on the x86 build
18:38:58 <GeKo> yes
18:39:14 <GeKo> we _could_ try doing the cross-compilation from 64bit -> 32bit
18:39:17 <GeKo> while we are at it
18:39:32 <GeKo> but, dunno. sounds like extra fun
18:39:51 <GeKo> but even then
18:39:57 <tjr> Are we building the x86 build on a x86 host?
18:40:02 <GeKo> yes
18:40:13 <tjr> Oh; and that's got a 4GB memory limit?
18:40:18 <GeKo> yes
18:41:05 <tjr> Ew...
18:41:18 <tjr> So --enable-debug implies --enable-debug symbols; so putting it behind enable-debug-symbols won't really help
18:41:39 <GeKo> well, then begin --enable-debug
18:41:48 <GeKo> that's fine too
18:42:09 <tjr> Sorry what do you mean by "then begin"?
18:42:24 <GeKo> *behind
18:42:26 <GeKo> sorry
18:42:44 <GeKo> because right now i get the linking error with "--disable-debug"
18:43:04 <GeKo> i used the same mozconfig as we ship (modulo stylo enabling etc.)
18:43:38 <GeKo> and backing out your patch "fixed" the out-of-memory linking error
18:44:25 <tjr> Ah okay. Alright let's move on, and I will figure out the incantation here that is needed
18:44:32 <GeKo> thx
18:44:51 <GeKo> pili: you are up
18:44:55 <pili> hi
18:45:22 <pili> just an announcement that I'll be travelling next week in India as part of the S9 outreach
18:45:33 <pili> so I'll be online slightly earlier
18:45:42 <pili> and will probably have to skip this meeting next week
18:45:46 <pili> but we'll see :)
18:45:59 <pili> also, I'll need a custom build at some point ;)
18:46:08 <GeKo> pospeselr: ^
18:46:09 <GeKo> :)
18:46:41 <sysrqb> pili: exciting! I hope it goes well!
18:46:47 <pospeselr> I'll try to get that to you tomorrow-ish
18:46:53 <pili> thanks! I'm looking forward to it
18:48:19 <pospeselr> I'll try to not to disappoint :)
18:48:36 <GeKo> pili brought up our meeting next week
18:48:48 <GeKo> i guess we can just talk about that topic now
18:49:07 <GeKo> i'll be afk next week (and starting from tomorrow afternoon-sh)
18:49:16 <GeKo> and pili is in india
18:49:18 <pili> thanks pospeselr tomorrow is fine, I have time
18:49:29 <GeKo> and there is a holiday in the US next monday (iirc)
18:49:43 <GeKo> so, i guess we skip the next week's meeting?
18:49:53 <GeKo> and come together on the 25th
18:50:00 <GeKo> at the usual time?
18:50:08 <sysrqb> sure
18:50:09 <GeKo> does that sound right?
18:50:09 <mcs> Sounds good to me.
18:50:11 <antonela> works for me
18:50:13 <boklm> ok
18:50:48 <igt0> +1
18:51:22 <tjr> GeKo: Try --disable-debug with --enable-debug-symbols="" (the ="" is necessary, you can't use --disable-debug-symbols)
18:51:22 <GeKo> sisbell: does that work for you, too?
18:51:34 <GeKo> tjr: ack
18:51:40 <pospeselr> ok
18:52:03 <sisbell> sure, it works fine
18:52:09 <GeKo> okay, great
18:52:21 <GeKo> anything else wrt status updates?
18:52:51 <GeKo> fwiw: i'll drop everything non-critical in the next few weeks to help with tba stuff until we have it ready for stable release
18:53:35 <sysrqb> should we plan on another release after you return?
18:53:45 <sysrqb> or do you think it's too late for another alpha?
18:54:39 <GeKo> it's a pity that i have to start more or less with vacations
18:54:51 <GeKo> but that's life :)
18:54:54 <GeKo> sysrqb: what do you mean
18:54:55 <GeKo> ?
18:55:03 <sysrqb> vacation is important :)
18:55:17 <GeKo> yeah, among many things :)
18:55:21 <sysrqb> do you think we can squeeze in another alpha release before the 8.5 becomes stable?
18:55:37 <sysrqb> or are we at the point where the next TBA releaes will be the first stable?
18:55:46 <GeKo> oh, yes i guess even two
18:56:07 <sysrqb> okay, great :)
18:56:09 <GeKo> i think we should try to get the important tba-a3 things into a release we planned after i am back
18:56:29 <GeKo> that's at least the circuit display, topl, and the ux stuff you have been working on
18:56:37 <sysrqb> okay, that sounds good
18:56:54 <GeKo> then we have the regular bugfix release on 3/19
18:57:05 <GeKo> which would be another alpha
18:57:18 <GeKo> and hopefully then at the end of march we do the 8.5 thing
18:57:37 <sysrqb> that timeline sounds good to me
18:57:45 <GeKo> as an announcement: i'll start tagging tbb-8.5 tickets once i am back
18:58:02 <GeKo> so that we get an overview about possible blockers and can prioritize
18:58:21 <GeKo> alright
18:58:29 <GeKo> discussion-time
18:58:55 <GeKo> do we feel we should talk about the vision meeting we had on friday?
18:59:08 <GeKo> andy questions? concerns? comments?
18:59:20 <GeKo> tjr: thanks for writing that document up!
18:59:51 <pospeselr> yeah it's a pretty good overview
18:59:53 <sysrqb> i appreciated having that discussion
19:00:02 <tjr> I think Arthur will have additional opinions; I just jotted things down this morning
19:01:28 <GeKo> okay. i think the main discussion will happen on tor-project (i hope so at least)
19:01:49 <GeKo> feel free to add your input. it's about your job :)
19:02:08 <pospeselr> I personally think that user safety really ought to be our number 1 concern
19:02:53 <pospeselr> like, in a future where tor is integrated properly into Firefox and Brave and what-not, I don't really see what the point is for Tor Browser trying to also fill that space
19:03:14 <pospeselr> of a 'functioning' browser that most consumers want (saved bookmarks, cookies, history, etc)
19:03:56 <GeKo> what do you mean?
19:04:06 <pospeselr> imho, we should feel free to veer back towards a super security focused tool for whistleblowers and activists and what not
19:04:17 <GeKo> okay
19:04:18 <pospeselr> in a world where usable alternatives exist
19:04:21 <antonela> im a pragmatic person and i don't think we should aim to compete with Chrome. We are not even a for-profit org. Instead, we can focus on being the most secure and usable browser around. It is about to enable users to choice.
19:04:32 <tjr> The further we veer in that direction, the more 'Tor browser' becomes an indicator of 'person of interest'
19:04:48 <tjr> Although I do generally agree with you I think; just playing devil's advocate
19:04:54 <antonela> well, that person of interest could be a simply citizen
19:06:07 <pospeselr> sure, and if they are concerned about the sort of threat models that don't get you killed, I think we should encourage those users to use future Super Private Browsing Mode enabled FF or Brave
19:06:08 <sysrqb> it's particularly difficult because people like simplicity, they want one browser that can do everything, and it knows when it should protect them and how
19:07:24 <boklm> but Super Private Browsing Mode in FF or Brave don't keep history and cookies either?
19:07:42 <sysrqb> correct :/
19:07:46 <pospeselr> true
19:08:30 <pospeselr> sorry, I was kind assuming that tor would eventually be enabled in general browsing mode in these browsers
19:08:58 <pili> I guess the question is also whether we trust others to do it well enough :)
19:08:59 <GeKo> i think that's not an unreasonable goal for the future
19:09:17 <GeKo> yeah
19:09:44 <tjr> pospeselr: I think if FF ships Tor we can expect it will only be enabled in a Super Private Browsing Mode that does not retain history
19:09:47 <pospeselr> yeah, i think it's too early to tell what the future will hold in that regard
19:10:11 <GeKo> okay, tor-project is your friend and i am happy to read how the discussion went once i am back :)
19:10:32 <intrigeri> I need to think more about how contextual identities come into play in all this (putting aside "reboot on Tails for your other 'you'" for a moment).
19:10:58 <GeKo> okay, let's move on to the hiring topic
19:11:25 <GeKo> for the anticensorship team we did something new in the sense that folks from different teams were involved
19:11:50 <GeKo> one goal was here to start thinking about a unified hiring process across team boundaries
19:12:06 <GeKo> i wonder what we think for the tor browser hiring we are currently doing
19:12:26 <GeKo> do we think it would be helpful to have folks from other teams helping here?
19:12:50 <GeKo> or would that just be a distraction/make the whole process more complicated/etc.?
19:13:02 <antonela> im happy to collaborate
19:13:27 <GeKo> thanks :)
19:13:51 <sysrqb> I think the anti-censorship team hiring process was unique because that team didn't exist, so it relied on existing teams
19:13:56 <antonela> the anticensorship hiring took too long i think, maybe we can reduce that fud time for prospects
19:14:08 <sysrqb> that being said, i think it may be helpful having the perspective of someone on another team
19:14:11 <brade> +1 for reducing time
19:14:11 <GeKo> sysrqb: yep, that's an important point
19:14:12 <intrigeri> One way that has worked rather fine at Tails is to involve 1 person (so far) from other teams at later stages (like interviewing the short list) but not earlier.
19:14:29 <mcs> +1 to involving a few people from other teams (but not everyone)
19:14:38 * mcs is also concerned about speed
19:15:07 <sysrqb> intrigeri: ah, that's good to know, thanks
19:15:09 <GeKo> mcs: you mean not everyone from other teams? or not everyone from the browser team?
19:15:40 <mcs> I mean it might be too much to openly invite everyone from other teams. Ask a few people instead.
19:15:51 <mcs> We should include all of the browser team people
19:16:01 <GeKo> yeah, sounds good
19:16:55 <GeKo> okay, i think a bit more about it but i like intrigeri's idea i think
19:17:21 <pospeselr> yeah same
19:17:24 <mcs> It is also valuable to have a schedule/timeline for the process (and try to stick to it).
19:17:34 <GeKo> that's true
19:17:37 <GeKo> and a good point
19:17:56 <GeKo> i actually brought that hiring question up last week in the vegas meeting
19:18:35 <GeKo> and erin will get back to us with basic policies and some framework if i understood correctly
19:18:48 <GeKo> so, we'll take it from there i guess
19:18:49 <antonela> i got that too
19:18:55 <GeKo> thanks
19:19:06 <GeKo> that's useful feedback
19:19:14 <GeKo> do we have anything else for discussion today?
19:19:39 <antonela> im groot
19:19:47 <sysrqb> :)
19:19:50 <pospeselr> lol
19:20:32 <GeKo> great. then let's call it
19:20:34 <GeKo> *baf*
19:20:46 <GeKo> (thanks all)
19:20:49 <GeKo> #endmeeting