16:01:02 #startmeeting S27 Kick Off 16:01:02 Meeting started Tue Mar 26 16:01:02 2019 UTC. The chair is pili. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:02 Useful Commands: #action #agreed #help #info #idea #link #topic. 16:01:06 who's here? :) 16:01:10 <- 16:01:11 asn is here 16:01:16 <- 16:01:19 * nickm is 16:01:36 has everyone seen the link for the kick off agenda ^ 16:01:40 ? 16:02:12 or rather, has anyone _not_ seen the link? 16:02:23 here it is again for posterity: https://storm.torproject.org/shared/3my7ZfcZt27Hpo7nq_-Cc6vy8g72fJn_2-lPnCkUpB9 16:02:44 here 16:02:53 I'll give a few more minutes for others to join and everyone to digest the agenda :) 16:03:41 * mcs is here 16:03:44 gaba and I wanted to call this meeting to officially kick off this project which is starting in April and will go on for 12 months 16:04:43 great 16:05:13 I have started collecting/tagging some tickets that were listed in the proposal, here: https://trac.torproject.org/projects/tor/query?sponsor=%5ESponsor27 16:05:27 pili how is colombia related with this and why S9 is my sponsor here? 16:06:05 antonela: in the proposal we mentioned we would do some testing in Colombia and Mexico as part of this project 16:06:16 pili: that looks like a reasonable start. 16:06:49 pili i see, but is funding the user research part not the ux/ui part 16:06:51 also, the UX improvements are not actually budgeted under this project but will be covered as part of the S9 work (if that makes sense) 16:07:05 antonela: correct 16:07:18 okey 16:07:43 * asn adding more tickets to the s27 list 16:07:56 not sure if this is what im supposed to be doign right now, but im doing it 16:08:23 going back to the tickets, I have not yet split them all into "can" and "must" but we can tag them accordingly with the sponsor27-can and sponsor27-must tags under the Sponsor field 16:08:32 asn: that's great, please carry on 16:08:39 although you can also do it some other time :D 16:09:25 I have linked the proposal on the kick off document to refresh our minds on it 16:09:46 and gaba very kindly wrote a great project overview here: https://storm.torproject.org/shared/D8lGanTy1p8wPlvqGx_cndsxXBAf5COz95hx_M7Oj1M 16:10:05 very nice 16:10:11 it seems there are some unresolved footnotes? 16:10:15 and finally, there is also a kanban board for this project which is still very sad and mostly empty: https://storm.torproject.org/shared/jRFRaP2xaTogYbU9f-1CYb02BoYOps41zqbhL33h-LC 16:10:18 i guess they point to trac tickets? 16:11:28 GeKo: yup, there is probably a cleaner version on granthub, but that is harder to share 16:12:22 first of all I wanted us to think about the external stakeholders that we will need to keep involved in this project 16:12:57 I started a list but I would appreciate some input and discussion so we can keep in mind who we may want to invite to future meetings and engage as we start working on this project 16:13:12 I'd like us to think about tooling that is based on onion services, like briar. 16:13:36 nickm: ok, I guess that would be an additional section in the kick off document 16:13:47 I meant as an external stakeholder 16:14:03 we dont usually have external stakeholders as a separate entity in our grant procedures, why have them in this one? what does it mean? 16:14:06 another one would be users/operators of smaller onion services; we don't have any of those on the list now afaik 16:14:08 ah, ok 16:14:29 also, it is IMO crucial to have a user in mind for this approver api -- does anybody actually want to use it right now? 16:14:40 nickm: approver api? 16:14:44 asn: well, for the objective on solving the issue with long onion service names we talked about different approaches in mexico 16:15:04 and these approaches seem to require collaboration with partners 16:15:13 ' Write a Tor software change proposal for a “rendezvous approver” API that can be useful for: ' 16:15:20 mentioned on gaba's list of objectives 16:15:21 pili: i think we have kind of decided on the one approach we are gonna take 16:15:43 asn: it's not really part of the grant procedure, it's something that I thought would be useful for this kick off 16:15:46 nickm: ah right. yes not sure if that's something -must. but yes we need to think of consumers 16:16:00 pili: i think the plan is to go with the https-everywhere approach and not with any other one 16:16:06 asn: the httpse approach? 16:16:07 ok 16:16:08 pili: we took this decision while makin the proposal more specific 16:16:23 okay. So shouldn't read the bullets on gaba's pad as objectives, but rather as possibilities?? 16:16:30 pili: because otf did not want vague stuff. i still stand by this decision, but if we want to be more flexible, we cna discuss more 16:17:23 nickm: the source of truth is the proposal, I believe that gaba took the objectives on that pad from the proposal 16:17:30 but I haven't gone through each of them one by one 16:18:00 nickm: that ticket was footnoted in the original proposal indeed. but if we decide it's too much work for not too much benefit, perhaps we can analyze the tradeoffs and not necesarily build it 16:18:06 is it ok to go back to thinking about stakeholders or have we got everyone we wanted? 16:18:47 (i admit there was a degree of brainstorming when citing tickets while writing the proposal) 16:19:20 So my claim was that we sort of need to know what we are doing before we know about the stakeholders. 16:19:39 If we are doing the rendezvous-approver thing, then we really need to find a potential user for it to count as a stakeholder. 16:19:47 if we aren't, we don't. 16:19:48 IMO 16:20:12 ok, let me look into it quickly 16:20:44 FWIW, I think the rendezvous approver thing is not something anybody is asking for right now? 16:20:51 correct 16:20:51 (any user, that is) 16:21:00 it was something that was deemed useful back when the DoSes were happening 16:21:07 ack 16:21:36 nickm: checking granthub to see the final proposal there 16:21:38 the DoSes are still happening (#29607) and it's also part of the proposal, but not sure if the rendezvous approver is what we should do 16:21:48 because the one linked above says "we will" 16:21:53 for the naming thing, we have a possible category of internal stakeholder we should think of: who owns the PR/reputational aspect of naming? 16:22:40 nickm: i dont think we want anyone to own the reputational aspect of naming 16:22:54 so the plan is to build the functionality in httpse, advertise it, and let the community do what it wants 16:22:55 like, if naming happens, somebody will want to register very-evil-stuff.onion, maybe just to be a troll. It will likely get coverage. 16:23:12 we had disucssed with bill that we want securedrop to make a ruleset with httpse rules for their directory 16:23:35 when it does, who will deal with messaging there? And who will wish that we had involved them in strategy discussions before the media hits us? 16:23:57 also there will not be a central registry of onions 16:24:04 perhaps i can write an email with the plan here 16:24:12 and we can discuss it there? 16:24:15 asn: that might help :) 16:24:16 thanks 16:24:23 ok 16:24:24 granthub is being very slow, sorry... 16:24:36 let's assume we do have to do this rendezvous approver work 16:24:50 who would the external stakeholder be? 16:25:08 some sort of onion service operator(s) that can help us take good UX decisions 16:26:07 great, thanks :) this would be people we need to find I assume 16:26:09 so that the rendezvous approver we build is useful by real onion installations 16:26:16 well we can brainstorm for a few of them 16:26:39 we know a few experienced onion service operators that could help 16:26:46 great 16:27:03 e.g. s7r, alec, sina, micah 16:27:12 ok, I'll add them in 16:27:39 any more or should we move on as there's quite a bit more to cover :) 16:27:44 now i dont know if these people would care to attend every meeting as "external stakeholders" or they would just be interested in helping with that particular deliverable 16:28:05 any more on the external stakeholder thing? 16:28:40 i think bill from httpse is extremely important to the onion naming plan, and we should let him know that the plan is on, and help him help us 16:28:43 asn: I don't think we want to invite them to everything but we need to think about how and when we want to engage them as part of this project 16:29:07 i.e we should keep them in the loop 16:29:11 asn: and ask him whether he is still the driving force from eff or whether someone else took over 16:29:12 (also the securedrop people would be quite useful in terms of experienced service operators) 16:29:34 GeKo: he def was back in mexico, and he didnt seem to give up any time soon, but yes good point 16:29:48 pili: agreed 16:30:21 asn: reading the backlog about the DoS as the reason for this API 16:30:40 I think it's ok if we can do something equivalent and explain why that's a better approach than what we said we'd do 16:30:52 yep 16:31:02 Do we plan to rely on EFF to make changes to HTTPS-E or will we produce patches and ask them to accept and merge? 16:31:17 (maybe I should wait for asn’s email) 16:31:22 mcs: yes we need to figure that out. bill was flexible and willing to work on this back in mexico. 16:31:27 i think eff 16:31:41 eff would be easiest but they'd need to be willing :) 16:32:06 I think they would be 16:32:18 I mean we are helping with their redesign and they have been very appreciative 16:32:30 Alexis would implement the new design 16:32:54 ok, any more on stakeholders? :) 16:32:56 So I could assume they would welcome suggestions with reasonable expectations 16:33:05 agreed 16:33:07 let's proceed i'd say 16:34:15 ok, so next I wanted to talk about how we're going to organise the work and whether we have some sort of idea which activity/objective we want to start working on 16:34:39 yes i think that's the meat here 16:34:40 I also wanted to highlight that the community team will be going to Colombia in the next month and it may be a good opportunity to gather use cases 16:34:55 so we need to think about whether there is something we could prototype quickly for testing there 16:35:05 or whether we just want to gather user needs information 16:35:22 oook 16:35:25 let's talk about organizing work first? 16:35:32 we need to talk about it, ggus cannot do everything alone and not sure who can travel next month (two weeks) to colombia with him 16:35:44 ye 16:35:48 antonela: I could go if needed, but we need to decide that between us :) 16:35:54 i think we should first triage the open tickets and split things into -must and -can 16:36:07 asn: sounds like a good plan to me 16:36:23 we can also estimate points, and add subtickets for anything that doesn't have enough granularity 16:36:28 ye 16:36:40 I also wanted to ask whether SponsorR leftover work was relevant here as I saw that mentioned in relation to this elsewhere I think 16:36:52 or maybe I just made it up, but would be good to confirm :) 16:36:54 i think we might want to grab some sponsorr leftovers. some stuff wer egood. 16:37:03 not quite sure yet tho. i just skimmed them today. 16:38:04 so im really interested in triaging the tasks (little-t-tor ones but also ux and tbb if people want me to help), but im currently quite hosed in roadmap work 16:38:21 in particular, i imagine that the current roadmap work will also drag a bit into april 16:38:34 and with dgoulet out of the game until the 16th it's basically just me 16:38:35 asn: the triage is possibly something that gaba and I could do with input 16:38:42 yes 16:38:51 possibly as a separate meeting 16:38:54 agreed 16:39:07 i think we can triage the whole project in 1-2 meetings 16:39:24 sounds good to me 16:39:29 where 'meetings' could be replcaed with 'hours' 16:39:42 will we put tickets under objectives? 16:39:48 yes 16:39:59 geko, who will work on it on apps side? 16:40:07 antonela: there are already some listed in the proposal iirc :) 16:40:12 but i would need help from the ux and tbb team 16:40:16 depends 16:40:27 there is no core team yet 16:40:28 yes, but we need to organize it in trac, somehow 16:40:37 but i guess mcs/brade could be part of it 16:40:46 geko, got it 16:40:55 antonela: I tagged most of them with the Sponsor27 label 16:40:56 it mostly depends on what we want to prioritize after 8.5 got out 16:41:11 maybe we could label under objectives further 16:41:16 pili yes but s27 label doesnt fit under objectives 16:41:18 exactly 16:41:42 so suggestion: 16:41:51 we can have a triaging meeting next week when also gaba comes back 16:42:00 where we figure out the process of how to roadmap on trac keywords etc. 16:42:07 GeKo: that is a good point and brings me to my point about what do we want to start working on 16:42:08 maybe we don't know yet and that's fine 16:42:09 and also do some of the core-tor triaging 16:42:16 and then we do another meeting for the tbb and ux stuff 16:42:19 but if we do, it would be good to know, especially for the Colombia visit 16:43:02 asn: +1 on the triaging meeting and also I wanted to discuss with everyone a meeting schedule for this project 16:43:23 as in, do we want weekly meetings? monthly? somewhere in between? 16:43:35 something between weekly and biweekly i'd say 16:43:45 possibly more at the start while we iron out the details and then drop to longer intervals 16:43:52 i'd say weekly for the beginning so that we straightne things out 16:43:55 and then move to biweekly 16:44:00 pili: yes 16:44:07 ok, is everyone ok with that? 16:44:18 yep 16:44:19 does this time work for people for, let's say, April> 16:44:21 ? 16:44:34 you mean 1600UTC? 16:44:45 it does for me 16:44:51 wfm 16:44:54 same here 16:44:55 it could 16:45:07 just to clarify this will be at 15:00 UTC (starting from next week) 16:45:20 aha, even better :) 16:45:27 also doable 16:45:37 15:00 UTC is better for brade and me 16:45:50 mcs: perfect :) 16:45:56 1500 utc or 1600 utc? 16:46:07 1500UTC 16:46:08 1500 16:46:19 ok 16:46:26 great 16:46:29 does that still work for everyone? :) 16:46:43 sure 16:47:02 we may have to reschedule once dgoulet and gaba are back, of course 16:47:17 true :) 16:48:06 ok, let's take that back to them when they are back 16:48:39 we have ~10 minutes left, should we discuss starting points for this work? 16:48:55 any user needs we want to discover on the field? 16:49:10 or does anyone else have any other points they'd like to bring up? 16:49:27 i'm really disconnected from this grant right now to come up with things we want to figure out in colombia :/ 16:49:42 i dont even quite remember the UX stuff we promised them 16:50:29 asn: that's fine, I can write an email summarising this if that would help 16:50:32 i dont think we can do anything related with this for colombia, if that happens in 2 weeks 16:50:43 antonela: ye i also doubt we can formulate a good plan for this 16:51:02 I think that the most important thing to do is to identify the stuff that takes the hardest upfront decision making 16:51:05 we can do that as we triage 16:51:06 antonela: I agree it's very ambitious, I wanted to bring it up so we were aware there was this opportunity 16:51:20 but it's ok to let this opportunity go and concentrate on other things instead 16:51:38 so when do we start triaging? next tuesday this time? 16:52:06 sounds good 16:52:07 asn: yes, as long as gaba is also able to join 16:52:19 will we triage all together? or per team? 16:52:20 i think i can lok over the ticket list earlier as well 16:52:34 to be better prepared for the meeting triage 16:52:47 *look 16:53:02 antonela: we can do it all together so everyone is aware of what the others are working on 16:53:07 and we can identify dependencies 16:53:08 * antonela added ux-team s27 tickets to this meeting pad 16:53:20 ok 16:53:25 ok 16:53:40 I will also do another run through of the proposal and tickets to tag potential ones that are missing 16:53:49 and everyone please feel free to do the same if you have a chance 16:53:52 we should create tickets, im sure 16:53:55 yes 16:54:04 ideally we should have a good idea of missing tickets before triaging 16:54:11 or roadmapping, if we want to call it that 16:55:04 ok, can I give everyone that as an action point before next meeting? 16:55:13 ok 16:55:18 i.e to look over the existing S27 list and think about missing tickets 16:55:32 and create them and tag them before the meeting next tuesday 16:55:49 btw given that we havent even roadmapped the thing, i doubt we will start working on this in 4 days when april starts, right? 16:56:00 is it ok to consider the actual start time to be mid-april? 16:56:04 is there an actual start time? 16:56:20 only for the purposes of when it will end 16:56:26 well we do work on it 16:56:34 triaging and meeting is part of it 16:56:41 as in, we have said we'll start in April, so the project will finish end of March 2020 16:57:00 also what GeKo said :) 16:57:04 sounds good :) 16:57:36 ok, anything else from anyone? 16:57:48 also, I would love some feedback on the format of this meeting 16:58:07 (you don't have to do it now) 16:58:08 what worked, what didn't, what you wish we had discussed 16:58:17 that way we can come up with a good template for these kinds of projects 16:58:34 and any other comments you may have 17:00:06 ok... is everyone good then? :) 17:00:10 yep :) 17:00:30 ok, great, thanks everyone for joining! 17:00:35 #endmeeting