19:00:17 <mikeperry> #startmeeting tbb-dev
19:00:17 <MeetBot> Meeting started Mon Nov 30 19:00:17 2015 UTC.  The chair is mikeperry. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:00:17 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
19:00:31 <mikeperry> Hello everyone!
19:01:02 <boklm> Hi
19:01:11 <mikeperry> Welcome back from the holiday (if it was a holiday for you :)
19:01:21 <mikeperry> Let's get started with the standups.
19:02:29 <mikeperry> Last week, as expected, my couple days of work were distracted by the ED search again. I did not have any time for TBB
19:04:00 <mikeperry> this week, I need to write our status report, handle a bunch of mail, announce+run the applications meeting (this time tomorrow), actually get us a gatekeeper cert, and update my tor-core patches based on feedback.
19:05:13 <mikeperry> I think we may be down a GeKo for much of this week though, so I will try to pick up some of that slack
19:05:16 <mikeperry> that's it for me
19:07:45 * mcs will go next
19:07:50 <mcs> This past week, Kathy and I revised our patch for #16940 (merged by GeKo - thanks!)
19:07:57 <mcs> We created a patch for #16863.
19:08:05 <mcs> Also, we commented on #17662 and triaged a few Tor Browser tickets.
19:08:10 <mcs> Next on our list:
19:08:15 <mcs> - Review some patches.
19:08:20 <mcs> - Work with arthuredelstein to find a good solution for #17661.
19:08:27 <mcs> - Fix some Tor Launcher usability problems in the setup wizard.
19:08:31 <mcs> That's all for us.
19:10:17 * arthuredelstein can go
19:10:46 <arthuredelstein> Last week, we checked in bugzil.la/1200802.
19:10:55 <arthuredelstein> And I wrote revised patches for bugzil.la/1174386 and bugzil.la/867501
19:11:00 <arthuredelstein> I did some investigation of #17009.
19:11:11 <arthuredelstein> And I posted a patch for a TBB banner #17565.
19:11:28 <arthuredelstein> I'm hoping, with revisions, the banner can be good enough to be included in the next TBB update (Dec 15).
19:11:41 <arthuredelstein> This week I will look more at the font fingerprinting patch, the TBB banner, #17009, #17661 and #17568.
19:12:04 <arthuredelstein> That's it for me.
19:13:16 <mikeperry> arthuredelstein: nice work on the banner. I was going to comment that you take some of Sue's color scheme and other points into account, but it looks like you made a solid effort to do that
19:13:31 <mikeperry> we should bounce this off of her and see what she thinks
19:14:27 <arthuredelstein> mikeperry: Good idea.
19:15:06 <arthuredelstein> (I suppose we could still include a picture of Jimmy Wales)
19:15:29 <mikeperry> haha
19:15:50 <mikeperry> or some random white middle-aged dude ;)
19:16:17 <mikeperry> it sounded like they may not have had enough data points to call that result definitively controlled, though :)
19:16:18 <GeKo> arthuredelstein: so the paypal is blocking tor is not an issue any more?
19:16:36 <GeKo> (i guess the click on the donate button links to some paypal donate option)
19:16:53 <GeKo> or how is the flow supposed to be
19:16:56 <GeKo> ?
19:17:13 <arthuredelstein> I don't have the latest information on paypal blocking. isabela might know more.
19:17:43 <mikeperry> paypal is trying to be nice to Tor, and usually works. however, every once and a while it just returns a white page with some exit IPs
19:18:03 <mikeperry> at least, that was the state as of last week
19:18:36 <GeKo> i fear that users are throwing their tor browser away if they see that this is not working because they think tor browser is broken or something
19:18:44 <mikeperry> as for this think, I think we were going to have a tbb-specific link on our website, so we can do some basic hit counting for this specific banner
19:18:57 <GeKo> or is that no worry anymore?
19:19:17 <mikeperry> s/this think/this thing's link
19:19:54 <mikeperry> well, that is another argument for having the banner go to a page we control, so if paypal goes nuts, we siwtch it to Amazon or something
19:20:09 <mikeperry> (we have an Amazon page ready as a backup already)
19:20:17 <GeKo> okay, sounds good to me
19:20:17 <mcs> I thinkl there should also be a way to hide the banner and maybe we should stop showing it after a certain date (e.g., 31-Jan-2016)
19:20:35 <mcs> (because I bet some people will not like it ;)
19:20:56 <mcs> I will comment in the ticket though.
19:21:03 <arthuredelstein> mcs: I think that's a good idea. Should be pretty easy to do that.
19:21:37 <arthuredelstein> I think 31-Jan-2016 is roughly when the next update is anyhow?
19:21:45 <mikeperry> oh yeah, right. it should have an x to hide it forevermore
19:21:59 <arthuredelstein> also a good idea.
19:22:51 <GeKo> arthuredelstein: is that a static image or do we think we should pick a random one out of, say 3?
19:23:26 <arthuredelstein> GeKo: Right now it's static. I don't have image of other people at this point.
19:23:34 <mikeperry> the next scheduled Firefox release is 25-Jan. which supposedly is fine for EU folks, but maybe a bit late to keep pestering the Americans
19:23:42 <arthuredelstein> Although the text is HTML, not part of the image.
19:23:47 <mikeperry> so hrmm
19:24:14 <arthuredelstein> Isn't the next Firefox release Dec 15?
19:24:20 <GeKo> yes
19:24:22 <mikeperry> yes.
19:24:24 <mikeperry> the next next one :)
19:24:45 <arthuredelstein> oh right. duh :P
19:24:59 <mcs> We could also stop showing the donate banner after so many times, e.g., after 10 times.
19:25:13 <mcs> (I am not sure what is best)
19:25:22 <arthuredelstein> Or maybe after the user clicks donate?
19:25:28 <GeKo> i think i am in favor of such an option
19:25:29 <mikeperry> I think we probably ask all these questions to the funding list
19:26:19 <GeKo> ideally the user don't have to do anything
19:26:35 <mikeperry> I bet Sue and Katina probably have some numbers to put on these things based on experience. also they may have other images/text in mind, and would be able to tell us if we should rotate (as GeKo mentioned)
19:27:41 <arthuredelstein> I like the rotating idea.
19:28:33 <mikeperry> yeah, it may cause people to get tired of it less quickly
19:28:59 <mikeperry> or at least, be more likely to think longer about donating before clicking the x or tuning it out or whatever
19:29:00 <arthuredelstein> Phoul said he might be able to get the text translated pretty fast. I don't know if it can be in time for the next release though, and if we want to go down that road.
19:29:23 <arthuredelstein> If we don't translate, we need to consider whether we show the banner at all on non-English versions of TBB.
19:30:04 <GeKo> but the users get forwarded anyway to en en-US only page, right?
19:30:08 <GeKo> *an
19:30:18 <arthuredelstein> That's true.
19:32:06 <GeKo> i think getting the text translated if we can would still be a win
19:32:27 <arthuredelstein> We could try to get the website donate page translated as well.
19:32:59 <arthuredelstein> I don't know how hard it is and how much time it takes.
19:33:04 <arthuredelstein> But it's not a huge amount of text.
19:33:09 <mcs> arthuredelstein: Will you summarize this discussion on the ticket (when we are done)?
19:33:28 <arthuredelstein> mcs: Will do.
19:33:31 <GeKo> mcs: maybe better sending a mail to the funcing list
19:33:34 <GeKo> *funding
19:33:44 <GeKo> to get feedback from more people
19:33:47 <mcs> Whatever you think is best.
19:33:56 <mcs> Feedback from funding people makes a lot of sense.
19:34:31 <GeKo> arthuredelstein: so i still think the translated text might make people more comfortable to click on the donate button
19:34:56 <GeKo> as they might understand much better what this is all about
19:35:37 <arthuredelstein> GeKo: Yes, I imagine it might be especially jarring for people who don't read English.
19:36:08 <arthuredelstein> After the meeting I'll talk to Phoul about localization.
19:39:31 <mikeperry> ok. I can mail the funding list if you'd prefer that. or you can. either one
19:40:18 <arthuredelstein> I'm happy to mail the list.
19:41:03 <mikeperry> ok, sounds good then
19:41:51 * boklm can go next
19:42:05 <boklm> This week I have mainly been making the testsuite work on OSX
19:42:06 <boklm> During the next 3 days, I will be going to https://reproducible-builds.org/events/athens2015/
19:42:13 <boklm> Then I will try to finish making the testsuite work on OSX, and fixing marionette tests on Windows, and look again at #17662
19:42:19 <boklm> That's all for me
19:43:22 <GeKo> here is what i did:
19:43:34 <GeKo> i finally made #15578 review ready
19:43:48 <GeKo> i am almost done with reviewing #9659
19:44:16 <GeKo> there is still some tsting i need to do
19:44:18 <GeKo> *testing
19:44:46 <GeKo> then i started with #15988
19:45:17 <GeKo> and thought a bit about #16863
19:45:43 <GeKo> mcs: i think a good option for now would be to review that ticket and merge it into to rbrowser
19:45:46 <GeKo> *tor browser
19:45:58 <GeKo> and start bothering mozilla after the esr45 swtich
19:46:15 <mcs> GeKo: That makes sense to me.
19:46:20 <GeKo> the thing is that the code changed/is changing quite a bit
19:46:35 <GeKo> and we would do double work if we tried to prepared it now for m-c i fear
19:46:50 * GeKo can't type anymore
19:46:56 <GeKo> *prepare
19:47:05 <mcs> OK. So after we update it for ESR 45 maybe the code will be more stable.
19:47:11 <mcs> (if we get lucky)
19:47:16 <GeKo> i hope so, yes
19:47:33 <mcs> Sounds like a good plan then.
19:47:53 <GeKo> this we i'll plan to continue with review stuf, #15988 andmaybe start working on #13893
19:48:15 <GeKo> but i guess the reproducible builds meeting will swallow a lot of this
19:48:21 <GeKo> we'll see
19:48:44 <GeKo> that's it for me
19:50:00 <GeKo> oh, i plan to do the ticket triage for december as well
19:52:08 <mikeperry> I can do that. I was planning on writing the status report anyway
19:53:20 <GeKo> ok, then do it
19:54:59 <mikeperry> jacek's mail about SEH was not terribly encouraging btw :/
19:55:20 <GeKo> yeah, that's true
19:55:50 <mikeperry> I am now uncertain as to if we should try to find clang hackers, GCC hackers, or just patch the sandbox ourselves
19:56:19 <mikeperry> is there a reason why we don't use clang in gitian (other than never tried it and it may be a lot of work)?
19:56:57 <GeKo> jacek mentioned that the clang cross-compile support for windows is not as good as the gcc one
19:56:59 <mikeperry> it seems like all of the new compiler and hardening features are appearing in clang first.. and it also seemed like your struggles with ASAN and such were exacerbated by using GCC and not clang
19:57:07 <mikeperry> ah
19:57:30 <GeKo> yeah, the clang option is a real one though
19:57:51 <GeKo> sounds like a real fun project
19:58:14 <mikeperry> heh. so then we'd need to find someone to do cross-compilation support for clang, and SEH?
19:58:31 <GeKo> i'll have to think about the mail a bit
19:58:35 <GeKo> yeah
19:58:42 <mikeperry> are the newer macos compilers by crosstools-ng/etc clang-based, also?
19:58:55 <GeKo> es
19:58:59 <GeKo> yes
20:01:07 <GeKo> we still have e.g. openssl etc. cross-compiled with gcc for os x
20:01:34 <GeKo> as we'd basically need to add support for clang cross-compilation ourselves
20:01:53 <GeKo> and that is just one lib
20:02:20 <GeKo> so, i suspect this will be a ton of work for windows where this is even more unusual
20:02:58 <mikeperry> maybe we ask Jaceck to ballpark costs/efforts for us for these various options?
20:03:14 <mikeperry> and if we could actually find a market for paid developers in each case?
20:04:17 <GeKo> yeah, i think that is a good idea. i'll do that tomorrow
20:05:50 <mikeperry> it's OK, since you're traveling I will add it to my pile of mails. what's one more? :)
20:06:32 <GeKo> fine with me ;)
20:06:51 * isabela is around now (sorry for not joing the meeting at the beginning I was in a call with OTF, will read backlog)
20:07:24 <GeKo> do we have a huseby today by chance?
20:08:56 <isabela> fundraising has a meeting today at 1:30pm pacific time
20:09:01 <isabela> in around 1hr
20:09:12 <isabela> I can bring any questions to the meeting and get a response right there
20:09:23 <isabela> or someone from tbb can join it
20:09:37 <arthuredelstein> Where is it?
20:09:42 <isabela> is on google hangout
20:10:01 <isabela> i have a link
20:10:55 <arthuredelstein> I can join if that would be useful.
20:12:03 <isabela> I think it will
20:12:14 <arthuredelstein> 1 pm PST?
20:12:27 <isabela> 1:30
20:12:45 <arthuredelstein> ok
20:13:41 <isabela> cool
20:14:00 <isabela> I will be a little offline as I need to go meet katina for this meeting
20:16:06 <mikeperry> alright, anything else for here?
20:17:09 <GeKo> i don't have anything
20:17:53 <mikeperry> alright. calling it!
20:17:56 <mikeperry> #endmeeting *baf*