18:00:14 <GeKo> #startmeeting tor-browser
18:00:14 <MeetBot> Meeting started Mon Mar 26 18:00:14 2018 UTC.  The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:14 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:00:18 <GeKo> hi all!
18:00:21 <arthuredelstein> hi everyone!
18:00:24 <GeKo> first meeting after rome
18:00:29 <boklm> hi!
18:00:30 <isabela> !!
18:00:36 <igt0> !
18:00:37 <GeKo> i hope at least some of you had some time to relax :)
18:00:51 <GeKo> please enter your items on the pad: https://storm.torproject.org/shared/tHoN4Ii7rLSjPE0OP4gydX4cMGadsXmRQNc-6lwru0N
18:01:09 <sysrqb> hello hello
18:01:10 <mcs> hi
18:01:13 <pospeselr> hi hi
18:01:18 <GeKo> and start reading the entries of all the other folks if done
18:05:32 <GeKo> okay
18:05:38 <GeKo> let's get started
18:06:01 <GeKo> it seems there is only one non-discussion point we have which is about releases
18:06:15 <GeKo> it seems we are testing getting out releases every week now :)
18:06:32 <GeKo> so far we can keep up with Mozilla which is good
18:06:51 <GeKo> but that reminded me that we walked in Rome about other folks helping with the building at least as well
18:07:13 <GeKo> so, please think about that if you could think helping out with that
18:07:30 <GeKo> the next release is planned in may and i  might not be able to help with the building for that
18:07:41 <GeKo> so it would be a perfect time to step up and it is not that hard
18:07:48 <GeKo> as we have all the infrastructure set up
18:07:58 <GeKo> so, no pressure to decide anything today
18:08:20 <GeKo> i'll ping this group in a couple of weeks again and then we can make proper plans
18:08:57 <mcs> Are there dedicated machines for the builds? Or do you use build-sunet-a.torproject.net ?
18:09:09 <GeKo> i have two
18:09:22 <GeKo> build-sunet-a.torproject.org which i mostly use
18:09:50 <GeKo> and a machine a friend dedicated to do the gitian builds long ago as a fallback
18:10:10 <GeKo> but anyone should be able to get an account on the tpo machine
18:10:20 <GeKo> and the building should be pretty easy
18:10:32 <GeKo> ok
18:10:35 <mcs> OK; thanks
18:10:45 <GeKo> pospeselr: fwiw #20283 is the one with high prio right now
18:11:04 <pospeselr> ok, I'll put the other one on the back-burner for now then
18:11:10 <GeKo> as this is technically speaking still a leftover for sponsor4 and that one ands pretty soon
18:11:12 <GeKo> thanks
18:11:30 <GeKo> let's look at the discussion items
18:11:46 <GeKo> i guess we can leave the roadmap as last item
18:11:59 <GeKo> anything you want to share about the rome experience?
18:12:09 <GeKo> something we should fix ad hoc for the team?
18:12:24 <GeKo> something you already know we should change for the next meeting?
18:12:47 <GeKo> (i.e. tor-dev meeting, not tor browser meeting)
18:13:48 <GeKo> ok, i have one item
18:14:17 <GeKo> i'd like to follow other teams and have a team day before the meeting day starts
18:14:36 <GeKo> because i felt this time that roadmapping for all the pieces is pretty hard
18:14:46 <GeKo> if it still needs to get squeezed into the meeting day
18:14:55 <GeKo> when we need time to coordinate with other teams
18:15:20 <GeKo> we "solved" that by having part of the roadmapping actually on the next day (the TBA parts)
18:15:20 <sysrqb> like the network team? starting a day earlier?
18:15:23 <GeKo> yes
18:15:56 <GeKo> but ideally we would have this at one day and could need this as well to discuss other items
18:16:14 <GeKo> that don't fit in the schedule for the meeting day
18:16:32 <GeKo> it's not something we need to decide now but a thing we can mull over
18:16:44 <GeKo> and keep in mind when thinking about the next dev meeting
18:17:01 * isabela likes this idea
18:17:29 <sysrqb> yep
18:17:29 <mcs> Overall, it seemed like we had enough time throughout the week (the session schedule seemed kind of light).
18:17:59 <mcs> But I guess if we need to have the roadmap done before other parts of the meeting start, having an extra team day makes sense.
18:18:18 <pospeselr> yeah
18:18:22 <GeKo> well, we don't have to
18:18:33 <GeKo> i mean it worked this time as well without the extra day
18:18:52 <mcs> I think it will depend on how packed the rest of the week is with other sessions :)
18:19:24 <GeKo> but i feel it would have worked better with it given that we are a large team now
18:19:55 <arthuredelstein> one option could be to have more extensive IRC-base roadmapping meetings before the in-person meeting and then just discuss the difficult bits in person :)
18:20:00 <arthuredelstein> *based
18:20:03 <GeKo> anyway, that's just a thought
18:20:18 <GeKo> yeah, we could do that, too
18:20:51 <arthuredelstein> not mutually exclusive ideas in any case
18:21:13 <GeKo> yep
18:21:35 <GeKo> okay, if there is anything else you think as immediate feedback from rome, let us know
18:21:49 <GeKo> meanwhile: next tor browser meeting
18:22:07 <GeKo> next monday is holiday here and i won't be available.
18:22:25 <GeKo> how about 4/3 (the tuesday)?
18:22:31 <GeKo> would that work for everyone?
18:22:37 <sysrqb> good for me
18:22:38 <mcs> Tuesday sounds good to me.
18:22:40 <GeKo> or should we postpone the meeting for another week?
18:22:41 <arthuredelstein> me too
18:22:54 <boklm> Tuesday works for me
18:22:56 <pospeselr> tuesday is fine
18:23:01 <igt0> it works for me
18:23:10 <GeKo> great
18:23:21 * antonela is around
18:23:27 <GeKo> then let's have it on 4/3 same time, same place
18:23:28 <GeKo> o/
18:23:32 <sysrqb> hi hi
18:23:36 <GeKo> i'll send an email to tbb-dev later
18:23:48 <GeKo> so, the final item is the roadmapping one
18:24:25 <GeKo> but wait it seems some bold things popped up
18:24:55 <GeKo> arthuredelstein: so the idea with that ticket is to disable update from HTTPS-E and only update it via tor browser updates
18:25:12 <GeKo> as a first step to not pinging any external source anymore for extension updates
18:25:35 <GeKo> alas, the torbutton/torlauncher hack is not working anymore with webextensions
18:25:45 <GeKo> so we need to find something new
18:26:05 <arthuredelstein> sorry, which hack do you mean?
18:26:10 <GeKo> the hack is setting the update url to data:text/plain,
18:26:26 <arthuredelstein> Oh I see.
18:26:39 <GeKo> which lets the update request get silently dropped by the browser
18:27:01 <arthuredelstein> I'm just wondering how we could keep the HTTPS-E rewrite database up to date.
18:27:12 <GeKo> while avoiding annoying warnings like those in #13129
18:27:26 <arthuredelstein> Or if there is any danger from not updating the extension, like a security update or similar.
18:27:33 <GeKo> arthuredelstein: HTTPS-E is soon shipping with an in-extension ruleset updater
18:27:55 <arthuredelstein> So that will presumably be pinging some third-party site?
18:27:58 <GeKo> yeah, we'd need to keep track of https-e updates
18:28:07 <GeKo> for security purposes
18:28:23 <GeKo> yes, some eff site to just upadate rules but no extension code
18:28:35 <GeKo> which is quite some win if done right
18:29:27 <arthuredelstein> I see
18:29:43 <arthuredelstein> What do you see as the biggest win?
18:30:28 <GeKo> that users don't get shipped third party extension code automatically anymore
18:31:03 <GeKo> by whoever is able to sign and deliver updates
18:31:28 <arthuredelstein> Got it. Makes sense. I guess as long as we have a way of monitoring any chemspill-like situations with https-e it seems like a good thing to me.
18:31:42 <GeKo> yes, that's the plan at least
18:31:56 <GeKo> alright
18:31:58 <arthuredelstein> thanks
18:32:07 <GeKo> isabela: your fp proposal point
18:33:33 <GeKo> it's the "vanilla tor data analysis" one on tor-internal
18:33:48 <GeKo> that should help us building the "click to flag" thing for our users
18:33:59 <isabela> yes
18:34:36 <isabela> ppl dont need to give feedback now but i do need some input on that asap cuz deadline is this friday
18:34:44 <isabela> inputs i am looking for:
18:35:06 <isabela> 1. will some labor from tb team be needed in order to research those 2 questions?
18:35:48 <isabela> 2. what could be the final deliverable? would something like a tech proposal on how to use the data researched (to answer those 2 questions) be something doable?
18:36:37 <GeKo> i can think about that this week
18:36:43 <GeKo> and we can coordinate a bit
18:37:09 <GeKo> i'll put it on my items for tomorrow
18:37:29 <isabela> tx
18:37:35 <isabela> i was about to say that :)
18:37:37 <sysrqb> i wonder if we should make the goal smaller, only target places where facebook are facing censorship?
18:37:51 <sysrqb> we can expand the countries we support from there?
18:37:59 <sysrqb> (ISPs?)
18:38:04 <isabela> to try to do it tomorrow (cuz i know is late over there) because once you share it other folks might comment on it etc
18:38:17 <isabela> and me and tommy are trying to make sense of what to propose and submit it by friday
18:38:45 <isabela> sysrqb: tbh that would be the same amount of effort
18:38:48 <GeKo> sure, will do
18:38:56 <isabela> sysrqb: we wont be promissing applying any solution
18:39:12 <sysrqb> okay
18:39:14 <isabela> just proposing a way to use the data (if our answers are all positive)
18:39:38 <isabela> like 'yes, if we get data X, Y and Z we could do something with that'
18:40:36 <isabela> but it can be negative where 'X, Y and Z are things we would need but right now there are limitations to use it that are xxx, xxx and xxx'
18:40:39 <isabela> idk
18:41:11 <GeKo> sounds good
18:41:27 <GeKo> okay, finally the final item the roadmap draft
18:41:46 <GeKo> if got tricked into using google by isabela and the famous spreadsheet master asn
18:41:56 <GeKo> it's not fancy yet
18:42:02 <GeKo> but it has content!1!
18:42:09 <sysrqb> heh :)
18:42:12 <pospeselr> lol
18:42:34 <GeKo> this is up for disucssion and i hope we can finalize this piece over the week
18:42:35 <antonela> haha
18:42:49 <GeKo> so that we have a roadmap for the next 6 months by next tuesday
18:43:13 <sysrqb> sounds good
18:43:27 <sysrqb> spread sheet looks good!
18:43:28 <GeKo> the assignment to people is not set in stone, just a thing i think could work
18:43:32 * isabela is writing a lot of comments about it
18:43:35 <isabela> and might move things around :P
18:43:44 <isabela> but will share with the team before i do it
18:43:48 <isabela> so y'all dont kill me :P
18:43:56 <GeKo> so, if you are familiar with that google interface please add comments there otherwise
18:44:07 <GeKo> i am fine reading txt files or pads or whatever
18:45:09 <GeKo> what we have right now is basically the content of our roadmap session + what we discussed in the mobile session
18:45:37 <GeKo> we might need to take into account two or three additional items
18:45:41 <GeKo> not sure yet
18:46:03 <GeKo> they would be related to the "make uploads work from china" thing arma was talking about
18:46:14 <isabela> yep
18:46:20 <GeKo> but the important pieces are already there
18:46:45 <GeKo> if anyone has immediate feedback, please share
18:46:55 <GeKo> otherwise by whatever means you feel comfortable
18:47:05 <GeKo> (no phone calls, though:
18:47:08 <GeKo> )
18:47:31 <sysrqb> :) *noted*
18:48:23 <isabela> i have one thing to note
18:48:36 <isabela> Review of play store stats (install/user stats/usability feedback) (O1.1)
18:48:48 <isabela> ^^ i have been doing this since at leasts Q3 2017
18:48:53 <isabela> i will share with folks on ml
18:48:56 <GeKo> great
18:49:00 <isabela> i am using this as a baseline
18:49:07 <sysrqb> oh, great, yes pleasse
18:49:07 <isabela> for when we ship stuff to see if comments will change
18:49:40 <isabela> spoiler alert - everything so far are things we know
18:49:41 <isabela> hehehe
18:49:47 <isabela> but is a good documentation of baseline
18:49:50 <isabela> *for
18:50:35 <GeKo> okay, do we have anything else for the meeting today?
18:50:43 <sysrqb> i looked through some of the orfox comments last week
18:50:49 <sysrqb> it'll be good to track them
18:50:55 <mcs> When we ship Tor Browser for Android will Orfox users get it automatically? As in, we are renaming the app? Or shipping a new app?
18:51:06 <sysrqb> at this point, no
18:51:10 <sysrqb> we'll change signing keys
18:51:14 <sysrqb> (key)
18:51:18 <sysrqb> so it'll be a new app
18:51:42 <mcs> Okay; I was just curious (I don’t know a lot about Android app delivery things).
18:51:43 <sysrqb> but we can include a dialog prompt in orfox that directs the users directly to the google play installation page for TBA
18:52:00 <isabela> sysrqb: at the playstore analytics they actually filter them with keywords
18:52:11 <GeKo> yeah, something like that sounds good
18:52:11 <isabela> and give you some stats on performance etc
18:52:14 <mcs> Prompting or doing something to migrate people sounds like a good plan.
18:52:25 <sysrqb> good :)
18:52:27 <isabela> hmm
18:52:36 <sysrqb> isabela: yeah, they do that in the google play app, too
18:52:37 <isabela> we wont take over the same page?
18:52:39 <isabela> i am confuse
18:52:48 <sysrqb> "usefulness", "easy to use", etc
18:53:22 * isabela has access to the dev analytics painel
18:53:27 <sysrqb> isabela: when i talked with n8fr8 in Valecia, we thought it would be easier starting with a new app
18:53:44 <isabela> hmm
18:53:45 <sysrqb> but we can discuss and decide when we get closer to TBA release
18:54:05 <sysrqb> basically this depends on how Guardian Project are signing Orfox right now
18:54:06 <igt0> +1 for a new app, otherwise the users can be puzzled about the names, e.g And the id will be different: https://play.google.com/store/apps/details?id=info.guardianproject.orfox (iinfo.guardianproject.org)
18:54:09 <isabela> i would prefer to stay where we are in order to keep users
18:54:15 <sysrqb> if they are using the same signing key as their other android apps
18:54:20 <sysrqb> then they don't want to give us that key
18:54:25 <sysrqb> so we'll need to create a new one
18:54:36 <isabela> ah
18:54:40 <sysrqb> if they are currently using a uniue key, then they could give that to us
18:54:56 <isabela> ok this info is very important for ux+communication to prepare for alpha and education campaign
18:55:14 <sysrqb> true
18:55:17 <isabela> question - will we shut down orfox page completely once alpha is out?
18:55:26 <isabela> or will there be overlap ?
18:55:27 <sysrqb> i'll follow up with n8fr8 and confirm what we should do
18:55:37 <GeKo> i think overlap makes snese
18:55:39 <sysrqb> i think short overlap
18:55:39 <GeKo> *sense
18:55:40 <sysrqb> yeah
18:55:49 <isabela> yeah
18:55:53 <isabela> we need to define all that
18:56:05 <isabela> because we must plan for a good campaign to educate people
18:56:08 <isabela> before during and after
18:56:13 <GeKo> we don't want to switch all stable users to some alpha product at once
18:56:13 <sysrqb> yup, totally agree
18:56:28 <sysrqb> heh, yeah
18:56:46 <sysrqb> one nice thing in google play, is we can rollout to only a percentage of users
18:56:52 <sysrqb> it's bad for fingerpriting
18:56:55 <isabela> i am ok with overlapping - lets make sure communication and ux is in the loop on this discussion too
18:56:57 <sysrqb> fingerprinting
18:57:07 <sysrqb> but good for testing alpha
18:57:22 <GeKo> that's fine with me
18:57:33 <isabela> sysrqb: what you mean to only a percentage of users?
18:57:42 <isabela> sysrqb: wouldnt that be a new page/build etc?
18:57:56 <sysrqb> we can say: "only tell 5% of users that there is an update for this app"
18:58:08 <isabela> but that will be no users right?
18:58:08 <sysrqb> and then we can see what feedback we get from them
18:58:12 <isabela> if the app is all new
18:58:22 <isabela> and we are not updating orfox
18:58:34 <sysrqb> ah, yes, but we can roll out an update with orfox at the same time
18:58:43 <sysrqb> and the orfox app tells users to install the new TBA app
18:58:44 <isabela> hmm
18:59:11 <GeKo> okay, folks. time for moving to a post meeting discussion i guess
18:59:12 * isabela would like more time to think about this stuff
18:59:22 <sysrqb> okay, we can discuss later
18:59:27 <sysrqb> i can start a thread on the ml, too
18:59:29 <isabela> thankyou
18:59:37 <GeKo> thanks for coming and happy tor browser improving *baf*
18:59:39 <GeKo> #endmeeting