14:59:35 #startmeeting Tor Browser Weekly Meeting 2024-01-08 14:59:35 Meeting started Mon Jan 8 14:59:35 2024 UTC. The chair is richard. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:59:35 Useful Commands: #action #agreed #help #info #idea #link #topic. 15:00:09 I hope you all have had a restful break 15:00:26 o/ 15:00:41 I don't have anything specific to plan or announce today beyond what you already likely know 15:01:16 s96 is still our top priority with a deadline of something we can show the sponsor at the end of this month 15:01:27 specific date is probably in gitlab somewhere one moment 15:01:52 Hi! 15:02:09 if you haven't had a chance to play with the latest alpha yet, we hit a pretty big milestone with the latest alpha release: we're now using the desktop boot experience and TorProvider system on Android (alpha)! 15:02:11 \o/ 15:02:22 * Jeremy_Rand_36C3[m] is joining the meeting for a little bit, but might vanish soon due to having been visited by the COVID Fairy 15:02:50 there's still a ways to go but it seems we're in the home stretch on Android as far as sponsor obligations go 15:04:40 of course the other main pillar of this s96 deliverable is LOX on desktop which henry-x, cohosh and jagtalon are/have been working on 15:05:18 hi 15:05:35 anyway i know we're only 1 day back, but can we get a brief summary of where we're at with each of these deliverables, any major issues or blockers we need to plan around/throw people at? 15:05:51 startinw th Android and connect assist I suppose? 15:06:57 I'm looking at firefox-android tor settings and wiring the settings UI to use TorSettings when appropriate 15:07:50 it looks like the date in Gitlab for s96 handoff for security review is the 18th 15:07:59 So, for Android we have the HTML UI out 15:08:17 With some preliminary backend work. Some additional backend work will be needed 15:08:38 Dan has worked on the plumbing for the settings and as he wrote he's now exposing them in a UI 15:08:58 I've found a thing I didn't expect while working on the defaultSettings thing, but I can fix it and it isn't a big deal 15:09:36 And it's that setSettings only sets bridges. But we can easily extend it to the firewall, and in a slightly less easy way to the proxy 15:09:59 However, currently we don't have firewall and proxy exposed on Android, so no big deal 15:10:46 I've also quickly exposed the bootstrap events, we should be able to use them with the native UI, but we will likely need to change the plumbing code to pass also additional details to Android 15:11:01 To avoid round trips if possible 15:12:10 (to avoid callback hells, if possible, since round trips are async; I'm not too worried about delays, since it's basically Java <-> C++ <-> JS stuff that happens in the event loop, if I understand correctly) 15:13:05 I think we'll need more plumbing code to expose TorConnect methods to the native UI, but I don't remember what we've already implemented (and in any case, it hasn't been tested) 15:13:22 I'd say settings are the most important piece at the moment 15:13:40 Otherwise users don't have a way of noticing/changing what the connection assist has done 15:14:31 The HTML UI could use some additional polishing, but if I understand correctly it isn't needed for the deliverable of the next week 15:14:57 not unless clairehurst's native UX MR is delayed 15:16:11 ok cohosh+henry-x how goes the Lox portion? 15:16:40 we're still working on implementing the backend part for henry-x 15:17:00 we're currently targeting using the HTML bootstrap for the s96 deliverable shipped this month and claire's native UI for after if I recall? 15:17:13 if that's blocking you we could push some stub functions that return values that of the right type/size/appearance at least 15:17:24 /o 15:18:08 dan_b: if the native bootstrap is ready and works then we'll send them that, otherwise we use the html as a fallback 15:18:12 regarding lox, for me i wasn't able to finish the bridge line validation errors before the holidays so i'll continue to work on that https://www.figma.com/file/rWgMwiiFTDFp4ujuP3PKbq/Tor-Browser-13.5?type=design&node-id=1364-1447&mode=design&t=kVa9Ox3N1XqUtpGH-4 15:18:14 From my side, I'm still working on the non-Lox changes (see https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/42036#note_2980161). I'm doing the CSS styling right now 15:18:30 > I hope you all have had a restful break < .. wait .. no one told me, I worked thru, 24/7 15:18:40 but it's not a hard requirement so if it turns out to be harder or there are gotchas its fine we can wait until the next major alpha release before shipping it 15:19:28 richard: the native UI ui parts are coming along well from what I've seen but the wiring and plumbing needs a redo as its lightly sitting on top of the integration to tor-android-service. the HTML bootstrap skipped ahead cus it allready calls all the GV bootstrap stuff and just needed exposing 15:19:51 cohosh/henry-x: is there any extra tasks you could use a PieroV for? 15:19:59 but ok so once I'm done hooking up settings in the settings pane that plumbing is my next priority then 15:20:04 or is this a mythical man month situtation 15:20:27 do not forget TorConnect events, eventually :) 15:20:30 Yeah it doesn't currently load web pages 15:20:38 (the native UI android) 15:21:07 Something with how its connecting not being properly implemented 15:21:35 But its a major rework of how we used to do it, so I'll need to go through more carefully and try to figure it out 15:22:20 clairehurst: ok :) 15:23:07 richard: no extra for PieroV on my side 15:23:36 kk 15:23:44 yeah TorController.kt needs to be ported to calling GV instead of tor-android-service 15:24:04 cohosh (and others): what if we merged the patch to separate the domain fronting from Moat.sys.mjs to a file on its own already? 15:24:46 I think that separation would be good for us, but I don't know if it'd be a problem for your current Lox backend work 15:25:21 (I'm talking about this commit, except for the changes in Lox.sys.mjs of course: https://gitlab.torproject.org/pierov/tor-browser/-/commit/6b0edbba62c49c9d1ea3d30f43204674cdd89fcd) 15:29:28 oops sorry got distracted by another chat 15:29:41 the domain fronting patch is definitely great 15:30:34 i'll reach out if we need anything else, but for now i think it's very specific to the lox library so better for onyinyang[m] and i to work on it and then reach out for advice on gluing it together once we're farther along 15:30:55 ok 15:31:18 ok finally catching up on mail from the break, I see there were a couple of gitlab issues that saw some attention, is there anything there that needs discussion or is worth mentioning? 15:32:54 I can't think of anything right now 15:33:21 I can ask on GitLab if it needs to stay on the record or in #tor-browser-dev 15:33:57 ok then 15:34:25 ok do you all have any topics you wish to discuss? otherwise we can end this 15:35:18 nothing on my end 15:36:36 alright then 15:36:41 have a good week everyone 15:36:49 see you all on the internet o/ 15:36:52 #endmeeting