16:01:18 #startmeeting network-health 08/16/2021 16:01:18 Meeting started Mon Aug 16 16:01:18 2021 UTC. The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:18 Useful Commands: #action #agreed #help #info #idea #link #topic. 16:01:25 alright, let's get started 16:01:36 here is the pad to enter the updates and items for discussion 16:01:42 http://kfahv6wfkbezjyg4r6mlhpmieydbebr5vkok5r34ya464gqz6c44bnyd.onion/p/tor-nethealthteam-2021.1-keep 16:01:57 o/ 16:02:06 this contains all the network-health/metrics related parts which are not s61 related 16:02:13 o/ 16:02:40 we have split that part out to the network-team meetings/monthly syncs 16:03:08 hi! 16:03:16 hello, hello 16:04:51 irl[m]: in case you are around, feel free to add your updates, too 16:06:21 okay, let's get started 16:06:45 so far i don't see any items marked in bold 16:07:01 so we might not have much to discuss today 16:07:08 yep 16:07:53 hiro: aha, you are using snippets for that, interesting 16:08:03 does that help to organize stuff for you? 16:08:09 GeKo: trying it out so I don't forget what I have been doing 16:08:21 yeah 16:08:41 in google it was custom to submit your snippets on friday... trying to resurrect that habit 16:08:48 i've been trying to update the meeting pad at the end of my working day 16:08:58 that way i document what i did 16:09:09 and had all the notes for the meeting ready at the meeting 16:09:17 and did not start frantically add things ;) 16:09:52 but, yes, maybe snippets is a good/better way to do things... 16:10:02 whatever works! 16:10:30 you board should also reflect this :) 16:11:05 well, yes 16:11:14 but it might not reflect the things actually touched 16:11:20 right 16:11:44 If I take notes I also record some thoughts too... which sometimes doesn't necessary go into one single ticket 16:11:51 right 16:12:25 okay, are we still good with the work for this week or do we need to discuss anything? 16:12:36 I am groot 16:12:41 hiro: one things comes to mind right now. 16:12:52 I hope to be able to merge and deploy the overload stuff this week I am not sure GeKo if you could review so irl isn't overloaded 16:12:55 sure tell me 16:12:59 before you came back we thought about what to do with the metrics-team mailing list 16:13:12 and thought it can go or merged to the network-health one 16:13:22 given that metrics is not an own team anymore 16:13:22 yes we discussed on the call 16:13:40 aha, it was on the call, right 16:13:49 I think that's ok as most of the issues would be of interest to the network health list members too 16:13:53 so, you are still okay with the idea 16:13:56 yeah 16:14:12 we just have to make sure people know and subscribe 16:14:24 right 16:14:54 i wonder whether should subscribe the folks on metrics-team@ automatically to network-health@ 16:15:15 so they don't need to do anything 16:15:26 not sure what the expectations are here 16:15:30 that could be a solution 16:15:38 and then, of course, announce the change on tor-project@ etc. 16:15:39 I was thinking as an automate response too 16:16:04 I think the only reason someone could get upset is if network-health has a lot of emails 16:16:16 yeah 16:16:39 we used to have a period where we used it for contacting eol relay ops 16:16:50 and folks were not happy about that as that involved a ton of mails 16:16:50 but I don't think that's the case? 16:16:57 but then i stopped 16:17:04 and things got pretty quite again 16:17:06 GeKo: hiro: if we don't any other item on the agenda, can we discuss roger's email after this topic? 16:17:28 if you tell me about roger's email, yes :) 16:17:32 which one? 16:17:56 I think is the one about spikes 16:18:00 *that* one 16:18:26 hiro: okay i'll file a ticket in the team project for the metrics-team mailing list and we can get the ball rolling 16:18:31 aha, yes 16:18:36 ggus: go ahead :) 16:20:26 so i was checking metrics data the other day, and sometimes the data doesn't look like a censorship circumvention or motivated by other social/political factors 16:20:36 like weird data 16:20:45 i wonder what to do in these cases 16:21:32 it's relevant to net-health 16:21:39 but not so relevant to community team 16:21:41 hm. i guess we could try to detect "weird data" 16:21:50 and then someone should take a look at it 16:22:05 that's probably related to what i wrote about who should own the scanning 16:22:31 I think with the right tooling scanning and alerting can be "easy" to maintain 16:22:47 yep, i agree 16:23:01 I have started playing with grafana and prometheus and it's promising 16:23:38 I think it is a question of allocating resources with a specific objective so that it can fit into one of our sponsor work present or future 16:24:25 ggus: we could put the task to ticketify things onto our plate 16:24:27 as i wrote 16:24:41 and that should cover cases you have in mind, too 16:24:52 GeKo: ok, i agree 16:25:23 the tricky part will be to get this sufficiently high onto our plate to get it actually done :( 16:26:09 i guess i could start with ticket creation this week at least 16:26:19 and we can start adding meat to those tickets 16:26:23 and flesh things out 16:26:30 if we have some time 16:26:37 uhm so maybe the relay side of the "weird data" is actually very relevant... some of the reports about bad relays from last year mentioned lack of tools... the client side could be less so atm 16:27:04 GeKo: gaba: we will have a new call for projects in Outreachy. I wonder if we want to submit a new project. and if "metrics data analysis" would be an interesting one. 16:27:15 hiro: yes, it could be 16:27:31 but even bad-relay work is a topic which is an endless sink 16:27:45 yes true 16:27:51 because there is so many ways for potentially weird data you can scan for 16:28:00 which means prioritization 16:28:28 but, yes, if we are smart we can make progress on both fronts at once 16:28:44 and we likely should, good point 16:29:22 ggus: well, maybe. i am not sure what the usual target audience for outreachy is 16:30:00 but i'd be happy seeing someone we get that way working on this topic 16:30:18 GeKo: new free software developers. and we can define the required skills for the project. 16:30:32 i see 16:30:41 do we have a deadline here? 16:30:44 at the moment we're doing user feedback reports and user support, but that will change with the new user support person (i guess). 16:30:45 I have tried to sell this topic to some of the ms students at the uni were I supervise thesis work... but no luck 16:31:01 nice try, though :) 16:31:06 GeKo: i believe we need to submit a new project at the beginnig of september 16:31:38 okay. so, what about we file tickets for different aspects of the whole project 16:31:48 and then pick some promising area for the outreach part? 16:31:54 *outreachy 16:32:17 sounds good. let's use network-health/team repo for this? 16:32:22 yes 16:32:46 what i am not sure about, though, is how this ranks to other needs we might satisfy with an outreachy person 16:33:00 like is it *the* project we should try to get someone for? 16:33:10 or do we have even more important things to get help with? 16:33:23 good question 16:33:32 how many projects can we submit to outreachy? and how does selection work? 16:34:10 like: is it about someone liking project A over project B? 16:34:11 we can submit 1 or 2 projects. in the last time, we only submitted one because we only had 1 mentor 16:35:17 in the previous years we had projects to update the trac documentation, user feedback+user support, and anon-gitlab. 16:36:17 i guess we can create the tickets and we can try to find out in parallel whether we want to have parts of the project for outreachy 16:36:42 not sure who is involved here or should 16:36:48 i guess you, ggus 16:36:54 maybe gaba as well 16:37:12 i let that for you to figure out, ggus :) 16:37:36 ok, the next project will happen during december 2021 - march 2022 16:38:38 i will ask to ux team too 16:38:56 sounds good, thanks 16:40:03 anything else for today? 16:40:27 i'm good 16:41:01 alright, hiro was already happy, so, thanks everyone and have a nice week 16:41:05 #endmeeting