18:01:36 <mikeperry> #startmeeting tbb-dev
18:01:36 <MeetBot> Meeting started Mon Jul  6 18:01:36 2015 UTC.  The chair is mikeperry. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:01:36 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:02:14 <mikeperry> Ok, let's get started
18:02:50 <mikeperry> Last week, I was at PETS and mostly distracted. I updated the list pof tags for 5.0 and July, as I said in my email to tbb-dev
18:03:33 <mikeperry> at PETS, I talked to researchers working on website traffic fingerprinting, and also Harry Halpin of the W3C
18:03:38 <mikeperry> among many other folks
18:04:20 <mikeperry> I owe the website traffic fingerprinting people (mostly mjuarez) some followup. That means this week I'll most likely be working on #14952
18:04:40 <mikeperry> I'll likely also need to catch up on email, and finish off the 5.0a3 release
18:05:05 <mikeperry> I think that's it for me for now
18:06:09 * arthuredelstein has to leave early, so will jump in now
18:06:19 <arthuredelstein> Last week I wrote a series of patches that cover #15703, #16416 and #16429 (all related to blob/mediasource URI isolation and regression tests). These patches are in the branch at
18:06:26 <arthuredelstein> https://github.com/arthuredelstein/tor-browser/commits/16429+7
18:06:46 <arthuredelstein> Then I started trying to understand and rebase David's patches for #13313 and also trying to understand the font-loading infrastructure from https://bugzilla.mozilla.org/show_bug.cgi?id=998844
18:06:52 <arthuredelstein> This week I plan to put further effort into #13313.
18:07:06 <arthuredelstein> I think I may also look at #16326 -- this seems like a natural extension to the regression tests from our #13749.2 patch.
18:07:29 <arthuredelstein> That's it for me.
18:09:48 <mikeperry> that sounds great arthur
18:11:11 * mcs can go next
18:11:22 <mcs> Last week, Kathy and I did some testing of the 5.0a3 updater.
18:11:27 <mcs> We used our own cert for signing MAR files so we could test a few scenarios (which means we had to produce and use our own builds).
18:11:35 <mcs> We did not find any problems other than #16496 which GeKo has already fixed (not updater related).
18:11:41 <mcs> We compiled and submitted our June status report to tor-reports.
18:11:46 <mcs> And we spent the rest of our time trying to find the root cause of #16495.
18:11:55 <mcs> We have not found it yet, so this week we will keep working on that bug.
18:12:03 <mcs> Hopefully we can find and fix it before we are out of the office without reliable net access for a little more than a week (July 10th – which is this Friday – through July 18th).
18:12:12 <mcs> If we have extra time we will look at #14205.
18:12:16 <mcs> That's all for now.
18:15:03 * boklm can go next
18:15:44 <boklm> Last week I fixed acid3 and https-everywhere tests, updated my split branches repo
18:15:47 <boklm> This week I will try to add a test for #15802, I'm also planning to setup a new machine that I will use for building Tor Browser
18:15:50 <boklm> I can also look at #15864
18:16:10 <boklm> Later this month I'm planning to be mostly offline from July 25th to August 1st
18:16:19 <boklm> That's all for me.
18:16:56 * n8fr8 n8fr8 and amoghbl1 can go next
18:17:10 <n8fr8> we publicly announced orfox, and gave access to our alpha builds: https://guardianproject.info/2015/06/30/orfox-aspiring-to-bring-tor-browser-to-android/
18:17:31 <n8fr8> now focused on getting all the extensions installed by default, and working where there are no current mobile versions (tor button)
18:17:57 <n8fr8> finally, we're still pondering the user-agent issue, and will continue the thread started on tbb-dev to sort it out
18:18:23 <n8fr8> oh, and documenting everything possible here: https://dev.guardianproject.info/projects/orfox-private-browser/wiki/Orfox_vs_Tor_Browser_FAQ
18:19:23 <amoghbl1> I also gave a rump session talk at PETS, trying to get a few researchers involved in testing the app out for leaks and stuff
18:24:23 <mikeperry> ah, so OrFox has an f-droid repo. nice
18:24:37 <mikeperry> I didn't see it in the default guardian project one.
18:24:44 <mikeperry> but I guess I need to add the alpha
18:25:59 <Yawning> n8fr8: random quick question, that's vaguely tor browser related (because I say so)
18:26:06 <Yawning> you build 0.2.6.x for orbot right?
18:30:13 <mikeperry> well, anything else browser related?
18:31:01 <Yawning> I'll probably have a new obfs4proxy tag that adds a bunch of dependencies for the next 5.0.x alpha build
18:31:58 <mikeperry> I am pretty satisfied with what folks are working on now. also as a heads up, I will be at the HTTP/2 workshop from July 26-31, which may cause some issues with the near overlap with others' vacations, but hopefully not
18:31:59 <Yawning> but that should be minimal work from y'all
18:32:10 <mikeperry> ok
18:32:13 <mcs> What's the plan for the TB 5.0a3 release?  Are you going to release build5?
18:32:18 <mikeperry> yes
18:32:23 <Yawning> I'll just be like "yo, merge dis branch"
18:32:23 <mcs> OK
18:32:38 <mcs> At this point we do not know when the next alpha will be, or do we?
18:32:39 <mikeperry> apparently I forgot to upload signatures, which blocked GK
18:33:17 <mikeperry> I was thinking we probably want to aim for the end of the month, but that is complicated by the HTTP workshop :/
18:33:28 <Yawning> #16467 *might* be worth cherrypicking, (technically #9663, but there's code overlap)
18:33:41 <Yawning> but it's kind of scary, up to y'all
18:33:46 <mcs> Will gk also be at that workshop?
18:33:59 <mikeperry> that probably means we should try to have a code freeze by Friday, July 24th tentatively
18:34:02 <mikeperry> I believe so, yes
18:34:03 <mcs> OK
18:34:10 <mikeperry> so I suppose we can coordinate there
18:34:22 <mcs> Code freeze targets are good to have in mind (at least they help me).
18:34:35 <Yawning> (it should just apply if you omit the part that wires in donna into the ed25519 code that doesn't exist in 0.2.6.x)
18:34:36 <mikeperry> so if you return July 18th, you'll have 5 business days to wrap up loose ends before we freeze?
18:34:57 <Yawning> (but, tor clients usually don't care about circuit build crypto performance)
18:34:57 <mcs> about that many days, yes.
18:36:17 <Yawning> even if you're on a potato86 tor browser users probably don't notice the ntor cpu consumption
18:36:18 <n8fr8> Yawning: okay. i need to get my head back into obfs/pt land anyhow in the next few weeks
18:36:48 <Yawning> n8fr8: might be worth just waiting till it hits master and a real release
18:37:02 <mikeperry> I think it is also likely we may call that release 5.0-stable
18:37:12 <Yawning> mikeperry: oh wow ok
18:37:13 <mikeperry> but not upgrade the 4.5 users until Aug 11th
18:37:27 <Yawning> 5.0 builds 0.2.7.x right?
18:37:32 <mikeperry> which may or may not involve a rebuild in that last couple weeks
18:37:44 <Yawning> are you gonna go back to 0.2.6.x for the release?  it's likely to still be alpha at that point
18:37:51 <pirate4096> Any new pluggable transports planned for 4.5?
18:38:10 <mikeperry> yes. right now it does. will that cause any problems for you, yawning?
18:38:16 <mikeperry> going back to 0.2.6.x?
18:38:30 <Yawning> no
18:38:33 <Yawning> was gonna say
18:38:37 <Yawning> going back is probably a good idea
18:39:02 <Yawning> as much as 0.2.7.x has a ton of really cool things I don't think it will be production ready by the end of this month
18:39:13 <Yawning> if there's things that need backporting let me know, naturally
18:40:45 <mikeperry> ok. similarly let me know if there is anything in 0.2.7  that we'll really want/need for PTs or other stuff
18:41:02 <Yawning> prolly not urgently
18:41:27 <Yawning> I'll let you know if I spot anything that needs cherrypicking, like I usually do
18:43:00 <mikeperry> ok, so is everyone OK with the plan? it may change by a few days in either direction depending on GK's travel plans for the HTTP workshop
18:44:28 <boklm> ok
18:45:19 <mikeperry> any other questions?
18:46:52 <pirate4096> What are some new features in 5.0?  Thanks!
18:48:54 <mikeperry> https://gitweb.torproject.org/builders/tor-browser-bundle.git/tree/Bundle-Data/Docs/ChangeLog.txt
18:49:28 <mikeperry> ok, I think that covers the meeting for today. I will be getting 5.0a3 up ASAP.
18:49:35 <mikeperry> thanks everyone
18:49:40 <mikeperry> #endmeeting *baf*