18:00:19 <olasd> #startmeeting
18:00:19 <MeetBot> Meeting started Thu Apr 29 18:00:19 2021 UTC.  The chair is olasd. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:19 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:00:21 <olasd> #chair pollo
18:00:21 <MeetBot> Current chairs: olasd pollo
18:00:32 <pollo> #topic Roll Call
18:00:41 <pollo> please say hi if you're here for the meeting!
18:00:41 * nattie waves
18:00:45 <olasd> ohai
18:00:47 <wouter> o/
18:00:52 <pollo> proposed agenda: https://deb.li/oNCD
18:01:34 <tumbleweed> o/
18:01:42 <olasd> sometimes I wonder why we don't pingall at the beginning of the meeting
18:01:52 <olasd> pretty sure we used to
18:02:11 * h01ger waves :)
18:02:14 <pollo> We could, I just felt that would be trolling people
18:02:23 <pollo> if people are ok with it I'll do it :)
18:02:45 <paddatrapper> \o
18:02:53 <tumbleweed> it'd probably help with attendance, yes
18:02:58 <wouter> pollo: please use #topic, not /topic
18:03:05 <pollo> oops
18:03:10 <wouter> ;)
18:03:21 <pollo> #topic Follow up on DC21 dates & tracks
18:03:32 <pollo> #info The 21st Debian Conference is being held Online, due to COVID-19, from August 22 to August 29, 2021. It will also sport a DebCamp from August 15 to August 21, 2021
18:03:48 <pollo> IIUC, atm we have 2 tracks (regular + pt_BR) ?
18:03:50 <olasd> (the schedule may be contracted a little, but that's the gist)
18:04:12 <pollo> + an ad-hoc one?
18:04:13 <wouter> I think we should assume there's going to be a few more tracks
18:04:16 <wouter> it's still early on
18:04:50 <pollo> wouter: well, it affects us quite a bit, as we need to add machines for each additonal tracks :(
18:05:10 <olasd> we need to add machines for each additional track running in parallel
18:05:20 <olasd> I don't expect all language tracks to run on all days?
18:05:23 <pollo> ah, that's what I meant by "tracks"
18:05:39 <wouter> my point is that we should probably assume there's a deadline a few weeks or months from now
18:05:42 <tumbleweed> probably depends on their turnout
18:05:55 <wouter> and not that we have all the numbers already
18:06:16 <wouter> so "two" for now, but likely something like 3 or 4 in the end?
18:06:20 <pollo> ok, so too soon to decide on that, but if someone could remind the content team each parallel track means a bunch of more work for us, that'd be nice :)
18:06:30 <wouter> yes, that too
18:06:45 <pollo> #topic Matrix integration
18:07:11 <olasd> I haven't had time to follow up on that yet; wouter suggested talking to the element people, which I will do, getting in touch with the sponsors team
18:07:16 <wouter> are people interested in my FOSDEM experience on that subject?
18:07:18 <olasd> I'm on vacation next week
18:07:23 <pollo> wouter: sure
18:07:25 <olasd> so should have time to do so
18:07:30 <olasd> and yeah, absolutely
18:07:43 <wouter> so, basically, for FOSDEM we talked to the element people, and they did all the hard lifting for matrix
18:07:49 <wouter> including hosting everything on their cluster
18:08:10 <wouter> that included a bot which would start things on schedule and hard cut when time ran out
18:08:18 <wouter> (which I honestly thought was a terrible idea)
18:08:23 <olasd> that was pretty awful
18:08:26 <wouter> yeah
18:08:44 <wouter> anyway, so there was another bot which would invite people to a "speakers" room when their talk was about to start,
18:08:56 <pollo> hmm, I'm somehow not super down with sub-contracting a very large part of our setup :(
18:09:06 <wouter> and you could send it commands from a special "backroom" channel that only FOSDEM orga had access to
18:09:23 <wouter> we don't have to, all the work they did is open source
18:09:31 <tumbleweed> a bot for coordinating speakers does sound useful
18:09:34 <wouter> we would be able to talk to them but run it ourselves in the end
18:09:35 <pollo> ok, nice
18:09:58 <wouter> they did a writeup of how everything worked on their blog
18:10:01 <pollo> some of the matrix dev is closed source for business reasons, that's why I ask
18:10:02 <wouter> let me look it up
18:10:17 <wouter> yeah, I don't think that happened here
18:10:21 <wouter> not 100% sure, but like 99%
18:11:00 <wouter> (still looking for blog post, sec)
18:11:01 <olasd> thing is, we really mostly want to look at the widget for streaming + chat integration, and maybe the jitsi / backroom stuff for managing (recorded) Q&A in a smoother way than we did
18:11:13 <olasd> you pasted it on the task
18:11:51 <wouter> https://matrix.org/blog/2021/02/15/how-we-hosted-fosdem-2021-on-matrix
18:11:57 <wouter> that's the blog post/writeup
18:11:57 <olasd> right, just got it too :D
18:11:59 <olasd> #link https://matrix.org/blog/2021/02/15/how-we-hosted-fosdem-2021-on-matrix/
18:12:18 <wouter> hehe :)
18:12:29 <pollo> ok, so I guess the next step is to look at how disruptive that change would be and how hard it would be to use the FOSS thing they did
18:12:46 <pollo> anything else?
18:12:48 <wouter> yeah, I think so too
18:12:57 <olasd> so I'll follow up on that next week: talk to the debian.social people to see how our matrix works, look at what element did, and maybe contact them
18:13:22 <wouter> should we action that?
18:13:22 <pollo> #action olasd follow up on matrix integration next week: talk to the debian.social people to see how our matrix works, look at what element did, and maybe contact them
18:13:33 <wouter> I guess ;)
18:13:36 <pollo> #topic Q&A webapp
18:13:52 <wouter> so, if we are going to use matrix, then this may be superfluous?
18:13:54 <pollo> We talked a bit on gitlab and decided we'd go with a django module
18:13:57 <pollo> maaaybe
18:14:06 <pollo> I still think it would be worth it for in person confs
18:14:19 <wouter> (not saying it's useless, just that the matrix stuff includes some bits there)
18:14:43 <pollo> It's a good point, but "real work" on this hasn't started yet, so we'll see :)
18:15:03 <wouter> sure
18:15:04 <olasd> I suspect a Q&A standalone thing would be worth it integrated as a widget in the matrix too
18:15:07 * fil sneaks in late
18:15:10 <wouter> and at any rate, it wouldn't be wasted effort
18:15:23 <pollo> anyway, not much more to add :)
18:15:32 <pollo> #topic Streaming
18:15:40 <pollo> not sure I remember why I added this
18:15:53 <pollo> are we OK with our current streaming setup?
18:16:07 <pollo> Ah, it was the Cloudflare CDN thing
18:16:16 <olasd> there was the cloudflare thing, but cloudflare also did another cloudflare in the last two weeks
18:16:16 <tumbleweed> yeah, we should probably try it
18:16:34 <olasd> (don't remember which cloudflare they did /this/ time)
18:16:35 <wouter> olasd: EPARSE, can you clarify?
18:17:22 <olasd> wouter: they did something stupid ethically/politically, again. I don't remember what
18:17:31 <wouter> oh, right
18:17:45 <olasd> ah, yes, the nft thing https://blog.cloudflare.com/cloudflare-stream-now-supports-nfts/
18:17:50 <tumbleweed> ah, that
18:17:53 * olasd headdesks again
18:18:03 <tumbleweed> they're also trying to shut down another patent troll, +1 there
18:18:09 <wouter> ooooohh, my goodness
18:18:21 <pollo> anyway, I don't think any progress has been done
18:18:25 <olasd> nope
18:18:39 <pollo> #topic Ansible & migration to Bullseye
18:18:44 <pollo> nothing on that side either
18:18:54 <wouter> it's not urgent though?
18:19:01 <pollo> it appears ansible will make it in bullseye though
18:19:02 <wouter> I mean, buster still works
18:19:09 <olasd> I think there was movement in the "getting up to date ansible in bullseye" front?
18:19:13 <tumbleweed> it appears it will, yes
18:19:21 <pollo> wouter: sure, but we said we'd like to migrate if possible
18:19:22 <tumbleweed> it's waiting for 20-day aging
18:19:28 <pollo> #info it appears ansible will make it in bullseye
18:19:29 <olasd> nice
18:19:35 <wouter> pollo: fairy nuf
18:19:57 <pollo> #topic Video Archive Documentation
18:20:00 <olasd> also, built-in mitogen, nice
18:20:03 <tumbleweed> day 5 of 20 (on mitogen)
18:20:39 <wouter> should I make a start on the docs?
18:20:40 <pollo> nothing new on the archive thing, my (small) appartement was under massive renos last week and kinda messed up my plans
18:21:00 <wouter> something like "videos get copied to X, they get rsynced to video.Debian.net from there"
18:21:03 <pollo> wouter: if you want, but I was planning on learning about the archive setup this way too
18:21:20 <wouter> pollo: okay, if you expect you'll have the time soon, I'll leave it to you then
18:21:29 * wouter <- not very good at writing good docs anyway
18:21:31 <pollo> it's not critical work, I'm just leaving it in the schedule not to forget
18:21:39 <wouter> right
18:21:43 <pollo> I skipped review!
18:21:46 <pollo> #topic Review system
18:21:47 <wouter> yes!
18:21:49 <wouter> I was going to say :)
18:21:59 <wouter> so I have made a start on the final review thing
18:22:04 <pollo> \0/
18:22:18 <tumbleweed> thanks!
18:22:22 <wouter> basically it'll be a form that says "is it good" with a yes/no comntrol
18:22:24 <wouter> control, even
18:22:29 <wouter> and a comment field that will turn into a log
18:22:41 <pollo> #info work has started on integrating the final review step in Sreview
18:22:44 <wouter> (the regular review comment field will also be a log going forward)
18:22:59 <wouter> so there's a few minor things changing there, nothing crazy though
18:23:27 <pollo> #topic Any Other Business
18:23:44 <wouter> oh, right
18:23:58 <wouter> I was also thinking of replacing the bs1770gain thing by ffmpeg audio normalization
18:24:07 <wouter> because bs1770gain *changes timing*
18:24:09 <wouter> horrible
18:24:20 <wouter> anyway, I've done some research there, but it's not ready yet
18:24:34 <wouter> I should be able to get that ready by debconf though
18:25:00 <pollo> #info audio normalisation should be ready for DC21
18:25:05 <pollo> anything else?
18:25:47 <pollo> #topic Next Meeting<
18:25:52 <pollo> Next Meeting Thursday May 13th from 18:00 to 19:00 UTC ?
18:25:59 <olasd> sounds good to me
18:26:03 <wouter> sgtm
18:26:03 <paddatrapper> sounds good to me
18:26:08 <pollo> #agreed Next Meeting Thursday May 13th from 18:00 to 19:00 UTC
18:26:12 <pollo> #endmeeting