#tor-meeting log

14:59:38 <sysrqb> #startmeeting Tor Browser 10.5 Roadmapping
14:59:38 <MeetBot> Meeting started Thu Nov 12 14:59:38 2020 UTC.  The chair is sysrqb. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:59:38 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:00:07 <sysrqb> oh, i wonder if meetbot's clock is slow, or if mine is fast
15:00:07 <gaba> who else is here?
15:00:16 <sysrqb> (Thu Nov 12 14:59:38 2020 UTC)
15:00:22 <acat> o/
15:00:54 <sysrqb> Pad: http://kfahv6wfkbezjyg4r6mlhpmieydbebr5vkok5r34ya464gqz6c44bnyd.onion/p/TB-roadmap-planning-keep
15:01:02 <GeKo> o/
15:01:20 <gaba> please look at the pad and add anything you think may be needed so we can discuss
15:01:23 <gaba> bonus for tickets
15:01:27 <gaba> bonus point*
15:02:59 <GeKo> so, originally i think this exercise was for the next like 2-3 months
15:03:11 <GeKo> but now it seems to be the whole way down to 10.5?
15:03:28 <GeKo> or what is the scope here?
15:05:03 <antonela> o/
15:05:05 <sysrqb> i think we should plan as far into the future as we can, with priority for items we should complete within the next 2-3 months
15:06:21 <sysrqb> we did begin with only thinking about the 2-3 months time frame
15:06:57 <sysrqb> but i think knowing which items should go into that is difficult without a larger picture of what we want to accomplish
15:07:14 <gaba> +1 what sysrqb is saying
15:07:31 <sysrqb> "Per-site security settings" heh :)
15:07:37 <sysrqb> antonela: you are ambitious :)
15:07:43 <gaba> we can put anything for 10.5 but focusing on being more granular for the next few months
15:07:49 <GeKo> reminder of the glorious past
15:08:10 <gaba> if you can be as specific as possible in the issues then it would be great
15:08:44 <antonela> :) well, im just adding wishes, we can plan some steps
15:08:53 <sysrqb> si :)
15:09:42 <sysrqb> gaba: do you have a procedure for discussing these items you used previously?
15:09:57 <sysrqb> or should I/you begin going through them?
15:10:53 <gaba> when doing this online i have been doing it in all kinds of ways... and i miss face to face..
15:10:59 <gaba> We should go through the needs
15:11:03 <gaba> discussing them and let's estimate work
15:11:08 <gaba> so we know how much we can accomplish
15:11:24 <sysrqb> okay, sounds good
15:11:51 <gaba> for sponsor 30 we need to implement antonela's proposal
15:12:41 <sysrqb> acat: did you read the proposal yet?
15:12:59 <acat> yes
15:13:31 <sysrqb> do you feel comfortable estimating the time needed for implementing any (or all) of it?
15:13:41 <gaba> https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/31284 has linked the bootstrap ticket but I do not think is included in the proposal
15:13:49 <acat> i can go through the tickets and to that
15:14:47 <acat> gaba: no, and i think we still need to decide the final UX solution for that
15:15:21 <sysrqb> yes, that is an open question
15:15:36 <sysrqb> i think it needs some user research/study, as well
15:15:46 <acat> should that ticket not be there? i think it's related to s30 even if not in the proposal
15:16:00 <sysrqb> yes, i consider it required for S30
15:16:23 <antonela> yes, the plan is to run user research with a nightly and iterate for the stable release
15:16:30 <antonela> maybe in March/April?
15:16:41 <antonela> one part that needs works is the integration with Salmon
15:16:55 <antonela> but im not sure how much do you think is doable for 10.5 and what for 11
15:17:18 <gaba> antonela: what specifically is that TB needs to implement now in the next few months?
15:17:34 <antonela> what devs consider doable
15:17:39 <sysrqb> heh
15:17:51 <GeKo> all the things
15:17:57 <antonela> all the things, now
15:18:07 <sysrqb> as much as possible, and no less :)
15:18:14 <sysrqb> okay
15:18:18 <antonela> we discussed starting with an static file for censorship detection
15:18:19 <gaba> mmm
15:18:26 <antonela> that needs a ticket or maybe is an ac ticket somewhere
15:18:36 <acat> is this tor-launcher#24527?
15:18:37 <antonela> (i saw a phw comment if im not wrong)
15:18:50 <antonela> good one acat
15:19:00 <acat> well, for some new definition of tor-launcher
15:19:22 <antonela> right
15:19:37 * antonela still need to reply to the list and incorporate that feedback in the proposal
15:19:42 <sysrqb> https://gitlab.torproject.org/tpo/community/outreach/-/issues/28531
15:19:47 <antonela> acat can do it with me and we can add some technical bits of it :)
15:20:30 <sysrqb> okay
15:20:33 <antonela> the second part is offering bridges
15:20:37 <antonela> and salmon should cover it
15:20:47 <antonela> https://gitlab.torproject.org/tpo/anti-censorship/rdsys/-/issues/7
15:21:42 <antonela> dismissing all the social graph part is plausible, as we discussed
15:22:13 <sysrqb> yeah
15:22:34 <sysrqb> limiting that information is important
15:23:05 <antonela> and there we have a good scope for s30. If you think we are not ready for removing tor launcher ui yet, then we plan on using tor launcher for showing all this user feedback
15:23:07 <sysrqb> okay, we don't really know the amount of time needed for these pices of S30
15:23:14 <antonela> but, you know
15:23:16 <antonela> ...
15:23:49 <sysrqb> *pieces
15:24:00 <antonela> (we still need to  define how we will launch tor browser offline, im aware)
15:24:25 <acat> antonel: do you mean keeping tor-launcher and implementing salmon UI there?
15:24:36 <acat> *antonela
15:24:49 <antonela> well, if you want to keep tor launcher in the scene
15:24:52 <antonela> i really hope we don't
15:25:01 * antonela like really
15:25:15 <sysrqb> but we have 1) quick-start, 2) salmon integration, 3) salmon UI, 4) country/region selection menu, 5) automated bridge configuration based on country
15:25:28 <acat> i think quick-start should be the first step
15:25:41 <antonela> good
15:25:51 <acat> because it will be more work trying to implement new UI in current tor-launcher
15:26:02 <acat> and then migrating it to the quick-start flow
15:26:08 <antonela> i also think that
15:26:32 <antonela> sysrqb: 4 is not necessary as we discussed before
15:27:31 <sysrqb> yep. but i included it just-in-case we decide that is the better method
15:28:23 <sysrqb> but we'll need some mechanism for deciding where the user is localted)(
15:28:35 <sysrqb> *located
15:28:59 <antonela> if im not wrong, we talked about doing an ip lookup
15:29:09 <antonela> GeKo: could you refresh that idea?
15:29:19 <antonela> quickstart can ask for consent, if is needed
15:29:25 <sysrqb> in any case, acat, do you think 4-5 months is a reasonable time for this?
15:29:43 <GeKo> i thought firefox has some internal means helping with location determination
15:29:46 <sysrqb> yes, i think we can look up the location by ip address
15:30:09 <GeKo> so, i was thinking using that
15:30:13 <gaba> 4-5 months for all those pieces and everything to be included in 10.5?
15:30:28 <GeKo> but we should re-investigate whether that actually fits our needs
15:30:30 <gaba> antonela: which of those pieces requiere more work in UX? is duncan will be involved?
15:30:44 <acat> the part i'm less sure of is the salmon integration, so maybe we could estimate 1, 4, 5 first and then 2, 3
15:30:45 <GeKo> i remember that it flew by when i audited code a while back
15:30:48 <acat> not sure if that makes sense
15:31:09 <sysrqb> acat: that's fair, sure
15:31:15 <antonela> gaba: will be, and is aware about the iterations it could take
15:31:41 <acat> but i think 4-5 months are reasonable
15:31:50 <sysrqb> i think those are easier than the others, because we have control over most of those pieces (at least 2))
15:32:26 <sysrqb> and then UI, querying rdsys, and plus modifying tor's config with bridges
15:32:36 <sysrqb> that seems more difficult to estimate
15:33:02 <sysrqb> and we may re-use the moat code, or re-implement it in firefox
15:33:03 <antonela> well, we can have 10.5 providing built-in bridges per region and plan 11 for rdsys
15:33:54 <gaba> tbh it seems a lot for the time on s30. dow e really need quick-start for it?
15:34:19 <gaba> mostly thinking about the uncertainity of how things will be with salmon
15:34:42 <antonela> quickstart and salmon are orthogonal
15:35:04 <antonela> we are implementing quickstart doing 1, 4 and 5
15:35:14 <gaba> i see, ok
15:35:19 <antonela> from sysrqb's
15:35:47 <sysrqb> okay, that's fine for me
15:36:21 <sysrqb> i think quick-start is a central part of S30
15:36:50 <sysrqb> but if salmon is not really a requirement, then it can be a -can featrure but not a -must feature
15:37:02 <antonela> right, and flexible enough to have the tech you want behind that ux
15:37:38 <gaba> salmon is the main thing phw is working on right now
15:37:40 <gaba> for s30
15:37:58 <antonela> we may have some help from the ac team on implementing salmon in TB also, could be great to collab
15:38:24 <sysrqb> right
15:38:37 <sysrqb> i think this may not be a topic we need to completely answer in this meeting
15:38:44 <gaba> we can discuss salmon with phw today at the s30 meeting
15:38:45 <sysrqb> maybe a better topic for the s30 meeting
15:38:50 <sysrqb> yeah
15:38:52 <antonela> yes
15:39:07 <gaba> tickets for TB s30 work
15:39:36 <gaba> there is a bunch in the pad but not sure they are specifically what needs to be done
15:41:19 <sysrqb> yeah, acat and I can look at those tickets
15:42:07 <gaba> ok
15:43:37 <sysrqb> i think we can assume we'll work on S30 items until March
15:44:13 <sysrqb> and we can squeeze other work into that time
15:44:56 <gaba> ok. Next for the remaining stuff from sponsor 58
15:45:37 <sysrqb> i think we can finish the remaining "audit tickets" by the end of november
15:45:47 <sysrqb> acat began looking at the android tests
15:46:16 <GeKo> we don't do O1.2?
15:46:50 <GeKo> at least it's not mentioned on the pad
15:46:52 <sysrqb> GeKo: i'm still debating what that means and how much we should do
15:47:17 <sysrqb> and that is disappointing
15:47:34 <sysrqb> but my experience with rebasing is that there are always mege conflicts
15:47:41 <sysrqb> *merge
15:48:45 <GeKo> well, yes
15:48:49 <gaba> geko: they are not mentioned in the pad because they are in the main ticket
15:48:56 <gaba> im writing it down now
15:49:02 <GeKo> nobody thought auto-rebasing would solve all rebasing issues
15:49:19 <gaba> 1.2, 2.3 and 2.4 are what is missing from what im understanding
15:49:40 <GeKo> i think we are done with 2.4, no?
15:49:48 <GeKo> i mean we have fenix-based thing now :)
15:50:10 <gaba> yes but there are a bunch of tickets not closed there
15:50:21 <GeKo> yeah, those need to get moved out
15:50:22 <gaba> or we remove them from s58 or we close them
15:50:27 <sysrqb> we're missing a few audit tickets
15:50:38 <sysrqb> i'd like to close those instead of moving them out
15:50:43 <gaba> so all the unchecked items in https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/33661 needs to be removed?
15:50:58 <GeKo> well, there are lots of missing audit tickets in tor-browser
15:51:26 <sysrqb> yes, i'm aware :)
15:51:54 <gaba> who can go through all those tickets and see which ones can get closed? im having a hard time figuring out if people are forgetting to close tickets or there is something missing
15:51:56 <sysrqb> but i think some of the remaining fenix audit tickets are important
15:52:26 <sysrqb> i'll clean up the S58 tickets
15:52:39 <sysrqb> i started working on them this morning
15:53:33 <gaba> thanks
15:53:34 <sysrqb> okay, well, we now have 5 minutes remaining until the hour
15:53:56 <sysrqb> should we schedule another meeting for finishing this?
15:54:01 <gaba> yes, 1 hour may not be enough :(
15:54:13 <gaba> can you all meet next week at the same time?
15:54:37 <acat> i can
15:54:38 <antonela> i can
15:54:40 <gaba> or we can go over the hour and discuss the maintanence stuff
15:54:42 <gaba> ok
15:54:42 <GeKo> i think so
15:55:00 <sysrqb> yes
15:55:16 <sysrqb> i have another meeting at 1600 UTC
15:55:39 <sysrqb> oh, no, that's tomorrow :)
15:55:47 <gaba> actually I do!
15:56:01 <sysrqb> yeah, i think the anti-censorship meeting is next, too
15:56:28 <sysrqb> or maybe that's in another hour
15:56:37 <cohosh> no it's starting soon
15:56:54 <sysrqb> okay, let's plan on next thu at the same time
15:56:55 <gaba> ok, let's meet again next week same time
15:56:57 <gaba> yes
15:57:11 <sysrqb> okay
15:57:50 <sysrqb> thanks everyone
15:57:57 <sysrqb> see you then (again)
15:57:57 <antonela> thanks!
15:58:01 <sysrqb> #endmeeting