18:02:09 <GeKo> #startmeeting tor-browser
18:02:09 <MeetBot> Meeting started Mon Oct 19 18:02:09 2015 UTC.  The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:02:09 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:02:23 <GeKo> hi everybody!
18:02:26 <huseby> hola!
18:02:48 <arthuredelstein> hi!
18:03:03 <mcs> greetings!
18:03:17 <boklm> hi!
18:03:17 <GeKo> some updates from me
18:03:45 <GeKo> i mainly worked on #10599 and reviewd #12967, #17329 and #17351
18:04:21 <GeKo> it seems i finally avoided all the dead-ends and we are back on track for our ASan builds, phew.
18:05:02 <GeKo> so, this week i plan to work on #17305 and might make some test builds which i'll send to tor-qa
18:05:26 <GeKo> then i plan to do some reviews, mainly the sponsorU related first
18:05:31 <mcs> Thanks for pushing your way through all of the pain (for ASan)
18:05:43 <GeKo> (i think the only one in this category is the window.name patch)
18:06:01 <mikeperry> bleh I am here
18:06:13 <GeKo> and i'll need to follow up on some mails (bug bounty, #17252 related
18:06:18 <GeKo> hi mikeperry
18:06:21 <mikeperry> things are way too hectic lately. travelling tomorrow
18:06:45 <GeKo> that's it for me now
18:07:39 <mikeperry> last week, I have been occupied with the ED search, discussing bw auths, and meeting other folks in Berlin, and general travel.
18:08:36 <mikeperry> This week I am going to another meeting, and will probably be distracted aftr hours with interviews. I hope to meet up with GeKo and ln5 to smooth out the release process the week after though, and help prep the Nov 3 release
18:08:49 <mikeperry> that's it for me. hope all is going well. sorry for my relative absence
18:10:06 <GeKo> i think we are good
18:10:24 <GeKo> although i am not sure where we are on the testing front
18:10:51 <mcs> mikeperry: I assume there is no movement on the Tor Apple dev account front?
18:13:10 <asn> dgoulet: i will have a good version of the ed25519 branch by the end of the week.
18:13:17 <asn> dgoulet: probably sooner. but im being conservative.
18:14:23 <GeKo> who is next?
18:14:46 * mcs will give a report
18:14:59 <mcs> This past week, Kathy and I posted patches for #12967 and #16620.
18:15:05 <mcs> For #12967, the Tor Launcher part (language prompt) has been merged and we are working on revising the build changes.
18:15:12 <mcs> We also created a patch for #17351 (already merged).
18:15:17 <mcs> Thanks to GeKo for his reviews.
18:15:23 <mcs> This week we plan to finish the #12967 revisions and work on some mochitests for #16620.
18:15:28 <mcs> We also plan to review the patch for #9659.
18:15:46 <mcs> In a perfect world, we will also do some work on #16940 and #17313 this week.
18:15:51 <mcs> That's all for us.
18:16:50 <GeKo> mcs: what is with the child ticket of #12967?
18:16:52 <mikeperry> I got dropped. Apple told us to re-aply. they lost our application in their bureaucracy
18:17:03 <GeKo> is that blocking that bug?
18:17:04 <mcs> Thank you Apple.
18:17:16 <GeKo> wtf
18:17:17 <mikeperry> so we re-applied. we'll see what happens. two people are eagerly standing by the phones for their "confirmation" call
18:18:36 <mcs> I don't think #17344 needs to block the Nov 3 release. It is "nice to have" but a little messy. If we add more locales later (or remove) we will need to change both Tor Launcher and the gitian scripts.
18:18:57 <GeKo> ok
18:19:06 <mcs> (if we fix #17344, Tor Launcher would not need to be changed to pick up the correct locale list)
18:19:15 <mikeperry> I though both tor launcher and torbutton had a makefile rule that said "build all locales"? or no?
18:19:34 <mikeperry> I thought Tails added that for us
18:19:56 <mcs> All locales are built but for #12967 we hard-coded a list of languages/locales in the language prompt ;)
18:20:01 <mikeperry> since they support the combined locale idea already (though via Debian's locale settings, which influence Firefox without UI)
18:20:16 <mcs> And #17344 is about removing the hard-coded list.
18:20:38 <mcs> (the list matches the packages we currently produce… about 15 or 16 languages)
18:22:21 <mikeperry> ok, cool. thanks
18:24:44 * arthuredelstein can go
18:25:03 * huseby will go after arthur
18:25:10 * boklm will go after huseby
18:25:11 <arthuredelstein> This week I wrote patches for #17207 and #17329.
18:25:25 <arthuredelstein> I opened #17367 (related to #17208).
18:25:44 <arthuredelstein> And I wrote a link system that will keep the Mozilla tickets sync'd with the latest version of our patches. For example, https://torpat.ch/16300 .
18:25:56 <arthuredelstein> So next week I'll be updating the bugzilla tickets with links to our patches.
18:26:06 <arthuredelstein> And I will look further into #17250, and also try to work on more Mozilla upstreaming.
18:26:11 <arthuredelstein> That's all for me.
18:26:12 <GeKo> nice!
18:26:37 <huseby> I spent last week auditing call sites for the contextual identities work we're doing
18:26:43 <huseby> it's our #1 priority ATM
18:26:54 <huseby> because if we can get that landed and owrking
18:27:03 <huseby> then all of the isolation patches from TBB will get easier
18:27:12 <huseby> ThirdPartyUtil will probably not be necessary
18:27:22 <huseby> it is my goal this week to answer that question.
18:27:43 <huseby> so not much progress on my front with regards to uplifting TBB patches, mostly foundation work
18:27:44 <anonym> mikeperry: correction: we don't use Debian's locale settings in Tails, but rather set the general.useragent.local pref according to $LANG in the environment (using some probably buggy heuristics).  intl.locale.matchOS doesn't work for whatever reason.
18:27:53 <GeKo> arthuredelstein: could you do a review for #16620? i plan to look at it but more eyes are alwasy good
18:27:54 <huseby> stuff that will be in the next ESR
18:28:18 <arthuredelstein> GeKo: Sure, will do.
18:28:20 <huseby> oh, and, I have a 1:1 briefing with chris beard next week on how Mozilla can best help the Tor Project
18:28:40 <huseby> I circulated the proposals to most of you and received great feedback.  thank you.
18:28:43 <huseby> that's it for me.
18:29:57 <anonym> in Tails, our hope is that once you on your (Tor Project's) side support multiple locales, then we can use whatever facility you use for detecting the initial language in the Tor Launcher
18:30:07 <anonym> or intl.locale.matchOS if it works then
18:31:29 * boklm can go next
18:31:39 <mcs> The Tor Launcher language prompt relies on intl.locale.matchOS to show the lnaguage prompt in the system language.
18:32:04 <boklm> This past week I worked on some marionette tests and fixed #17125
18:32:05 <boklm> On unnamed-messenger, I looked at the reproducibility issues, and after fixing them have been able to reproduce the Linux and Windows builds (ticket #10942)
18:32:06 <anonym> mcs: excellent! then it must be working again :)
18:32:23 <boklm> While doing that I noticed that it's now possible to build gecko reproducibly without using faketime.
18:32:28 <anonym> (actually I think I already know this from the ticket :))
18:32:35 <boklm> After discussing this morning with GeKo about how we should use the same tools to make maintenance easier, I opened ticket #17379
18:32:46 <boklm> This week I'm planning to do more marionette tests, try to have unnamed-messenger OSX builds reproducible. I will also try to help on #17305.
18:32:51 <boklm> That's all for me.
18:33:09 <mcs> anonynm: It worked for GeKo and for us on a couple of Linux systems… but we will need to gather more data points ;)
18:33:32 <GeKo> and even on windows, *cough*
18:34:19 <mcs> yes, we tested on Mac OS and Win7 as well.
18:34:20 <anonym> mcs: as soon as there is an alpha/beta of such a Tor Browser I'll give it a shot
18:34:46 <anonym> perhaps there already is one? pre-alpha? :)
18:35:19 <anonym> ah, there's a meeting. won't continue highjacking it. sorry!
18:36:12 <mcs> R.e. #17379, is that our plan then (to comverge on a common build process)?
18:36:41 <boklm> That's a proposed plan
18:36:46 <mcs> OK
18:38:08 <GeKo> given the overhead of reproducible builds we currently have i don't want to have n different systems being maintained for n products
18:39:39 <tjr> I documented the current status WRT alternate memory allocators (and partitioned heaps) - I haven't heard anything about it so I'm not sure if no one has questions or no one read it =)
18:40:06 <GeKo> tjr: i read it and a response mail is on my pile of things to do
18:40:13 <GeKo> thanks for doing that :)
18:40:34 <mcs> I plan to read it soon!
18:40:41 <tjr> no problem; I'm sorry I wasn't able to find time to really drive it forward
18:42:28 <GeKo> no worries, tjr
18:43:02 <GeKo> alright, is there someone else who wants to report things?
18:43:09 <ln5> i have a thing
18:43:28 <ln5> tjr, dkg and myself are working on an IETF draft on CT gossip
18:43:39 <ln5> it'd be great if people involved in tor browser commented on it
18:43:50 <ln5> there's a draft of the draft at https://www.adb-centralen.se/~linus/volatile/draft-ietf-trans-gossip-01-dev-20151017.txt
18:44:04 <GeKo> thanks. is there a deadline.
18:44:10 <GeKo> ?
18:44:13 <ln5> yes, 23:59 UTC :)
18:44:28 <ln5> that's for -01. but there'll be more time after that.
18:45:53 <ln5> (-01 being the version of the draft that will be discussed at the next IETF meeting, beginning of november)
18:46:55 <ln5> happy to take comments on the TRANS working group mailing list or in private if you prefer that
18:48:48 <ln5> oh noes, i killed bambi^Wthe meeting!
18:49:11 <GeKo> nah, anything else?
18:49:49 <ln5> nothing more from me
18:49:57 <GeKo> thanks, everybody *baf*
18:50:00 <GeKo> #endmeeting