18:00:35 #startmeeting tor browser 18:00:35 Meeting started Tue Nov 1 18:00:35 2016 UTC. The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:35 Useful Commands: #action #agreed #help #info #idea #link #topic. 18:00:39 hi all! 18:00:44 hi 18:01:02 hi 18:01:07 howdy. 18:01:11 welcome to the first tor browser meeting in november at an unusal day and at some places in the world an usual time 18:01:24 hi 18:01:32 i guess we start with the status updates 18:01:37 i can go first today 18:02:54 last week i worked on a bunch of reviews, on #20352 18:03:22 and tested a mozilla fix which is big for an esr context 18:03:49 but it still seems to compile in all our cross-(compile) envs and nothing blows up 18:04:11 i guess we'll be fine when it lands (it might already have, need to double-check) 18:04:21 then i tool two days off and am somewhat back 18:04:29 which brings me to this week 18:04:49 there is the begin-of-the-month admin stuff i need to do 18:05:13 and i plan to the morst urgent things in the coming three day 18:05:14 s 18:05:31 there is still kind of vacation here and i need to adapt to that for personal reasons 18:05:48 thus, a lighter week for me and i plan to be back working normal next week 18:06:08 if there is stuff that is urgent please shoot an email as i might not be on irc 18:06:17 that's it for me for now 18:07:03 * mcs will go next 18:07:11 Last week, Kathy and I created a fixup for #20204 and we squashed #20429. 18:07:18 We also finished #20185 and posted “work in progress” files for #20121. 18:07:26 And, as usual, we reviewed some patches and spent some time on bug triage. 18:07:33 This week we plan to create a fix for #19646 (the fix should be small and we would like to have it ready for the next Tor Browser release). 18:08:03 We will work some more on MacOS sandboxing (#20121). 18:08:09 We also plan to comment on #20497 and review various patches. 18:08:18 That’s all for us. 18:08:35 Sounds great :) 18:09:08 * femme can go next but it's a semi-discussion point which can wait 18:09:26 I'll introduce myself? 18:09:41 hi! 18:09:46 Sure! Or I can :) 18:10:06 Hi, all. I work on desktop team for Canonical. If it's about web browsers in Ubuntu, I'm probably involved somehow. 18:10:14 For a few weeks, as a tangent or intersection of personal interests and day-job, I have been thinking about packaging TBB as a cross-distro, auto-updating "snap" package. I thought I should look at your build-scripts first. I've checked out the scripts and am learning how they work now. I will probably have questions soon. 18:10:25 EOF 18:10:47 interesting 18:11:37 yes, it sounds interesting 18:11:58 qengho: let us know if things are unclear and we can help 18:12:04 Hopefully not chinese-curse "interesting". 18:12:10 I will. Thanks. 18:12:25 I was looking at snap packages also to solve the issues small t tor was having being supported especially on lts releases https://pad.riseup.net/p/tor-lts 18:13:17 There's some overlap with the sandboxing work Yawning is doing but it should mostly be complimentary. 18:13:55 I'm happy to steal ideas from Yawning! 18:14:00 This week I'm going to look more into this and examples of it being done with gitian 18:14:45 (hopefully together with qengho :) 18:15:31 That's it for me! 18:15:36 thanks! 18:15:38 * arthuredelstein can go 18:15:47 This week I worked on #20414, #20264, #20347, #19459 and #16622. 18:17:29 zwiebelbot? 18:17:40 got eaten maybe? 18:17:53 * femme thought it was on her end 18:17:57 Anyway, I also worked on a first-party isolation test for cookies (bugzil.la/1312541). 18:18:15 This week I want to finally finish up #16622 and start working on the jemalloc stuff for SponsorU. 18:19:04 arthuredelstein: so the strings are okay to get merged as far as i can tell? 18:19:10 the donation banner ones 18:19:59 GeKo: Ok, great. I think they look OK as well. 18:20:25 We can merge them without merging the rest of the patch to give us time to look at the JS etc. 18:20:34 well, i was wondering if all folks said "yes" to them 18:20:38 "rest of the patch" -> "second patch" 18:20:43 i have seen mcs commented 18:20:56 did you get feedback from ssteele, isabela etc.? 18:21:06 (i have not read all my mails in my backlog yet) 18:21:29 I haven't, but I can send another email asking them to have a look. 18:21:46 that might be good 18:22:07 and, yes, i thought of picking just that commit until we get the other stuff reviewed 18:22:15 we have another week it seems 18:22:26 but better to get it done as early as possible 18:22:34 Yes, just so the translators have time. 18:22:39 yes 18:25:07 Thanks. Anyhow that's all for me! 18:25:12 who is next? 18:25:13 * boklm can go 18:25:29 This week I made a first version of testsuite bundles for #19067, added some hardenning tests on #15138 18:25:32 I started investigating the build error I have on #20439 (for a better fix than the one I posted) but did not fully understand it yet 18:25:40 This week I'm planning to work on #19817, #20439, #20443 18:25:50 That's it for me 18:26:57 boklm: it seems you think we are done with #15138 then? 18:27:10 GeKo: yes 18:27:17 good, thanks. 18:27:35 * tjr can go 18:27:54 rbarnes and dveditz met last week to talk about Tor at Mozilla and we came up with a proposed roadmap for Tor/Mozilla stuff. It seems to have gotten generally good reception but is still a proposal. 18:27:58 I'm starting some conversations internally about non-uplift things like mingw builds as part of TaskCluser. 18:28:03 I'm worried some of this may stall as I will go on leave at the end of the week until January and will have little or no time to drive these things. So I would encourage Arthur/GeKo/whomever to prod regularly on individual points to try and keep them moving along. 18:28:06 I've asked Nathan about the patchlist Orfox has regarding proxies and any other mobile-only patches - it's important we get these this week so if you know where these are I'd appreciate any additional help. 18:28:11 Entirely unrelated to Mozilla - I noticed that GCC 4.9 merged --enable-vtable-verify - it might be interesting to do a build with that turns on and see if FF crashes immediately. https://gcc.gnu.org/wiki/vtv It would probably count towards the memory allocator hardening for SponsorU. 18:29:14 (that's it) 18:30:54 tjr: i guess https://bugzilla.mozilla.org/show_bug.cgi?id=1046600 might answer the vtv thing? 18:31:30 see froydnj's comment 18:31:32 Ah! 18:31:47 Ew 18:31:48 Ok 18:32:05 yeah, i spent quite some time on it until i got there 18:32:06 :) 18:32:10 tjr: Which things are you worried about stalling or would like me to keep tabs on while you're on leave? 18:33:22 Well, basically everything except uplift, which I think will keep chugging along as you work with Ethan and crew. 18:33:27 tjr: i poke jacek caban 18:33:45 In addition to the email I sent yesterday? 18:33:45 and i planto get back to you folks early next week regarding the road map 18:33:50 no 18:33:58 i had that mail in mind 18:34:31 Stuff like mingw/TaskCluster integration are a good exmaple of something that will be easy to be stalled 18:35:12 And the process changes I'm trying to bug people about regarding prefs and fingerprint prefs. As well as more robust testing architecture for proxy escapes and general browser testing with tor prefs/compile options 18:36:41 tjr: Makes sense. 18:36:55 Do we have tickets for all these things? 18:37:45 Nope! Let me confirm with my boss and I'll hopefully mass-open tickets this week on both bugzilla and trac. IS that a good approach? 18:38:26 I think that will help get everyone on the same page. 18:38:37 Sounds great to me. 18:38:48 tjr: it seems i dug out the orfox repo: https://github.com/guardianproject/tor-browser 18:39:01 the guardianproject patches are on top of ours 18:39:31 but i am not sure how to weigh them for uplift concerns etc. 18:39:37 I probably want the 45.4.0 branch right? 18:40:00 tjr: Seems like these things you listed are perhaps more needed on bugzilla than trac. But I'd be glad to be CC'd on all of them. Also whiteboard:[tor] is a good tag. 18:40:00 orfox-tor-browser-45.4.0esr-6.5-1, yes, i think 18:40:12 tjr: yes, bugzilla 18:40:36 and we can then open trac tickets on our side if we need to work on tor browser patches in the mean time 18:41:10 tjr: and cc'ing me would be neat, too 18:41:42 okay, anyone else here for status updates? 18:41:43 sure! 18:41:59 otherwise let's get some discussion going. 18:42:10 i guess the first point would be next meeting time. 18:42:35 would 19UTC be okay next week (in case little-t-tor is moving the time as well)? 18:42:52 ok for me 18:42:52 (and from next week on i mean until DST comes again) 18:42:53 GeKo: fwiw, we are scheduled relative to UTC 18:43:11 so, our meeting is always 1700 UTC 18:43:35 okay, cool. thanks 18:44:06 ok for me 18:44:10 ok for us 18:44:44 I'm OK with that too. 18:45:05 okay, i'll announce it on tbb-dev then later this week 18:45:20 do we have something to dicsuss regarding the postponed release? 18:45:46 Could you elaborate on the postponed release, I must've missed that 18:46:24 mozilla announced that they need to move all releases by one week 18:47:57 actually they seem to start with candidate builds this week already 18:48:05 If we're good with the baked in key pinnings, we should be good I think 18:48:19 but esr 45.5 will go live on nov 15 18:48:23 i think we are. 18:48:35 there was more than one week buffer iirc 18:49:00 mcs: i thought about fixing #20118 for stable as well 18:49:21 it baked in our alpha series and is important for some users 18:49:37 do you think that there is something speaking against it? 18:49:40 GeKo: That seems like a good fix to pick up. 18:49:44 okay 18:50:34 do we have something else for today? 18:50:59 https://hacks.mozilla.org/2016/10/webassembly-browser-preview/ 18:51:49 They say it could end up being enabled as soon as ESR 52, so maybe something we should keep our eyes on, as a potential fingerprinting vector 18:53:37 yes, thanks. 18:54:57 it might solve some of our issues with asm.js though, we'll see. 18:55:09 that too 18:55:16 okay. time to wrap up it seems. 18:55:33 thanks all for being here today. *baf* 18:55:40 #endmeeting