18:01:03 <GeKo> #startmeeting tor browser
18:01:03 <MeetBot> Meeting started Tue Jul  5 18:01:03 2016 UTC.  The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:01:03 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:01:24 <GeKo> welcome everybody to a tor browser meeting on tuesday
18:01:33 <arthuredelstein> GeKo: I just noticed your email said 1900 UTC. But I'm fine with now :)
18:01:45 <GeKo> ugh
18:02:03 <mcs> I am here (with Kathy).
18:02:28 <GeKo> i copy and pasted the wrong mail :(
18:02:40 <GeKo> i meant 18:00 UTC
18:03:11 <GeKo> let's get started though and hopefully folks wanting to show up are showing up a bit earlier
18:03:12 <boklm> hi
18:03:32 <GeKo> okay. i think i can go first today
18:04:26 <GeKo> last week i worked on the design doc update, #18925 and #16652
18:05:24 <GeKo> i had some ideas about #11506 and #19528
18:05:48 <GeKo> i did the admin work for July and triaged a bunch of tickets
18:06:19 <GeKo> this week will be short for me as i plan to take some days off (thursday-tuesday)
18:06:30 <GeKo> we can talk about that later i guess
18:06:54 <GeKo> nevertheless i plan to finish up #16652 and #19200 at least
18:07:02 <GeKo> that's it for me.
18:08:39 * mcs will go next
18:08:52 <mcs> Last week, Kathy and I posted patches for #19273.
18:08:58 <mcs> We helped the Tor Messenger team with a Tor Launcher issue.
18:09:03 <mcs> We reviewed patches for #19478.
18:09:08 <mcs> Then we took a few days off.
18:09:12 <mcs> We have not figured out our entire plan for this week, but we will start with #19532 and we will also look at some Mac OS issues that users are reporting on the blog.
18:09:20 <mcs> That’s all for now.
18:11:38 * arthuredelstein can go
18:11:44 <arthuredelstein> Over the last week, I worked on upstreaming patches to Mozilla, including
18:11:47 <arthuredelstein> https://bugzilla.mozilla.org/show_bug.cgi?id=1281963 (aka #17207)
18:11:52 <arthuredelstein> https://bugzilla.mozilla.org/show_bug.cgi?id=1281959 (#17502)
18:11:57 <arthuredelstein> https://bugzilla.mozilla.org/show_bug.cgi?id=1121643 (#13313)
18:12:03 <arthuredelstein> https://bugzilla.mozilla.org/show_bug.cgi?id=1235520
18:12:14 <arthuredelstein> I also reviewed https://bugzilla.mozilla.org/show_bug.cgi?id=1217238 (#1517)
18:12:20 <arthuredelstein> and #19273.
18:12:43 <arthuredelstein> This week I will work on more patch upstreaming and hopefully look at js/Math fingerprinting.
18:12:57 <arthuredelstein> That's it for me.
18:14:04 * boklm will go next
18:14:15 <boklm> This past week I fixed #19402 to make it work with small resolutions. I worked on #18923 and opened tickets for failing tests. I made a patch for #19528.
18:14:31 <boklm> This week I'm planning to try a build with the patch for #19528. Then work on #19410, #11506, #15994.
18:14:40 <boklm> That's it for me.
18:16:34 <GeKo> arthuredelstein: i forgot to ask last week whether working on #10281 resulted into something
18:16:56 <GeKo> like did you get a functioning build with jemalloc4 enabled?
18:17:29 <GeKo> unrelated i am wondering what to do with #14429
18:17:53 <GeKo> it seems we find more and more holes in it (which is understandable)
18:18:05 <arthuredelstein> Re 10281, I have not, but I can give it a try this week
18:18:30 <GeKo> should we back it out in the alpha series as well until we try again to get it (properly) fixed?
18:19:23 <arthuredelstein> On #14429: I think one approach is to only attempt to round the dimensions on non-maximized screens
18:19:57 <arthuredelstein> I think the rescaling stuff we attempted is probably not practical.
18:21:01 <arthuredelstein> Were you maximizing the window or using fullscreen mode?
18:21:02 <GeKo> good idea
18:21:07 <GeKo> maximizing
18:21:16 <GeKo> i did not test fullscreen
18:21:30 <arthuredelstein> I wouldn't be surprised if that doesn't work either :P
18:21:44 <GeKo> yeah, i somehow assumed that :)
18:22:14 <GeKo> adapting that is nothing urgent fwiw. i just thought to bring it up
18:22:17 <mcs> r.e. #14429, I assume we will break too many sites if we report rounded dimensions (that is, not report the actual dimemsions)?
18:22:56 <arthuredelstein> That has been my untested assumption.
18:23:21 <arthuredelstein> But also, I think that there are many ways to measure the displayed content bounds besides the obvious screen.width
18:23:32 <GeKo> yes
18:23:36 <mcs> Ah, good point.
18:23:46 <GeKo> and probably a bunch of side-channels
18:24:05 <arthuredelstein> One possible way to improve #14429 is to create some kind of resizeend listener
18:24:08 <GeKo> i'd rather avoid opening that can of worms
18:24:25 <arthuredelstein> (in C++)
18:25:12 <arthuredelstein> But that probably needs a separate implementation for each platform, so it's probably quite a bit of effort.
18:25:54 <GeKo> i agree
18:26:10 <arthuredelstein> And there's still the annoying problem of tiling window managers being hard to detect.
18:27:37 <GeKo> arthuredelstein: oh, another thing i wondered was whether you could try the test case in the embargoed bug related to #8725.
18:27:49 <GeKo> i don't have the means right now, alas
18:28:36 <GeKo> but i thought figuring out whether the proposed defense copes with that trick seems worth knowing before merging the patch
18:29:25 <arthuredelstein> Which is the embargoed bug?
18:30:32 <GeKo> 1120398
18:30:41 <GeKo> you are added afaict
18:31:12 <GeKo> that said do we have anyone else for status updates?
18:31:25 <arthuredelstein> Yes, I can see it. I will try the test case.
18:32:45 <GeKo> thx
18:33:14 <GeKo> okay, moving to the discussion part
18:33:31 <GeKo> does anybody has an item for it?
18:34:15 <GeKo> *have even
18:34:28 <Sebastian> I would
18:34:39 <GeKo> go ahead
18:35:07 <Sebastian> hope it's not offtopic. We'd really like to deploy an installed version of TB. Is that something that's planned for some point in the future?
18:36:44 <GeKo> it is not really on any roadmap as far as i can tell right now
18:37:02 <Sebastian> We have a system with 15k users, we'd like to make TB either the default or one of two choices. Expecting them to download it would mean a lot of wasted space as well as roughly nobody does it on their own
18:37:44 <GeKo> i see
18:37:57 <mcs> On Linux or ?
18:38:04 <Sebastian> Debian, yeah
18:38:17 <Sebastian> also windows, but windows isn't used so much
18:38:40 <mcs> #18367 and #18369 would probably help
18:39:21 <GeKo> these are at least blockers for that i think, yes
18:39:30 <Sebastian> right
18:39:30 <GeKo> not sure what else we would need
18:40:40 <Sebastian> Anyway, just to point out that there's at laest some demand for this kind of feature
18:40:44 <Sebastian> thanks :)
18:40:51 <GeKo> ok, thanks.
18:41:03 <GeKo> we'll keep it in mind.
18:41:15 <boklm> for now, an option could be using torbrowser-launcher configured to use a local mirror
18:41:29 <GeKo> but i can't promise anything :(
18:41:38 <Sebastian> There are never promises :)
18:41:55 <Sebastian> well, sometimes there are promises but there are never guarantees :)
18:42:12 <GeKo> i was about to write that, too ;)
18:42:43 <GeKo> looking at my list i have some items:
18:43:45 <GeKo> first, a boring one: we have some tbb-* mailing lists. there was the question of who wants to admin these.
18:44:00 <GeKo> i am doing that right now with helix. if anybody wants to help just say so
18:44:19 <GeKo> then another question is what to do wuth #19417
18:44:21 <GeKo> *with
18:44:48 <GeKo> it seems we have linkability issues as well and i am wondering whether we should disable asmjs by default as well
18:45:14 <GeKo> until we have proper patches for it
18:45:33 <mcs> Does disabling mean that asmjs sites will stop working or will they just run more slowly?
18:45:47 <GeKo> more slowly
18:46:15 <mcs> Then I vote for “disable via pref” (but I think we may need to add support for a pref)
18:46:18 <GeKo> but we had bug reports in the past about mega.nz e.g. where users complained it was highly unusable
18:46:40 <mcs> I remember that now that you mention it. Hmm.
18:46:41 <GeKo> although we never investigated whether that was due to asmjs or just disabling all the other JITs
18:47:54 <GeKo> well, there is already a pref which is currently governed by the slider
18:48:13 <GeKo> and just exempting that one and disabling it by default seems to solve the "problem"
18:48:28 <GeKo> so, that part is the easy one it seems
18:48:33 <mcs> (I guess there is a pref)
18:48:59 <GeKo> yes, javascript.options.asmjs
18:49:24 <GeKo> i guess i prepare a patch for disabling it until anybody raises serious concerns in the ticket
18:49:25 <mcs> It would be nice to solve the linkability problem and persistent caching issue soon. It would be better if Mozilla could help ;)
18:49:41 <mcs> sounds like the best thing to do for now (disabling)
18:49:46 <GeKo> ok
18:50:27 <GeKo> another item i have: next meeting
18:50:41 <GeKo> would it work for you doing it 18:00 UTC next tuesday again?
18:50:54 <arthuredelstein> Works for me
18:51:07 <mcs> That works for us too.
18:51:25 <boklm> Works for me
18:51:49 <GeKo> okay. then i can try writing a proper mail to tbb-dev tomorrow :)
18:52:23 <GeKo> just as a heads-up: july is travel-wise and vaction-wise a bit tricky for me
18:52:48 <GeKo> i'll be at PETS the week from July 18. not sure about work and meetings yet
18:53:04 <GeKo> the two weeks after that i have long-planned vacations
18:54:17 * boklm has some vacation planned the week from July 18
18:54:30 <GeKo> which kind of got messed with due to mozilla's new release schedule
18:54:46 <mcs> While we are talking about vacation, I will mention that Kathy and I will be out July 29 - August 5th.
18:54:47 <GeKo> the first week of my vacation i won't have any internet
18:55:05 <arthuredelstein> I will also be starting vacation on the week of July 24
18:55:21 <GeKo> and mike agreed to tag things for the stable build
18:56:18 <GeKo> mcs: arthuredelstein: sounds fine.
18:56:51 <GeKo> boklm: sounds fine as well. just one week?
18:57:05 <mcs> I guess it will be helpful to put together a list of start and end dates for vacation this month and next!
18:57:08 <GeKo> i am asking because August 2 seems to be a scheduled Mozilla release
18:57:45 <boklm> yes, I will be there before the August 2 release
18:57:50 <GeKo> cool
18:58:23 <GeKo> as i said mike will help with the stable one and i plan to get the alphas taged on the weekend + the stable built there
18:58:42 <GeKo> and then after travelling back on monday i hope i can sign everything :)
18:59:06 <GeKo> it's a bit more challenging this time than usual but, hey :)
18:59:27 <GeKo> that's all i have at least
19:00:56 <GeKo> mcs: yes. i send a mail around so that all of us have a rough understanding who is working when
19:01:05 <GeKo> anything else?
19:01:15 <mcs> GeKo: sounds good (and nothing else from me)
19:01:42 <GeKo> thanks all *baf*
19:01:47 <GeKo> #endmeeting