Diffstat (limited to 'BoardOfDirectors/IrcLogs/2022/06-08.mdwn')
1 files changed, 111 insertions, 0 deletions
diff --git a/BoardOfDirectors/IrcLogs/2022/06-08.mdwn b/BoardOfDirectors/IrcLogs/2022/06-08.mdwn
new file mode 100644
@@ -0,0 +1,111 @@
+[[!format txt """
+[12:04:38] [connected at Wed Jun 8 12:04:38 2022]
+[12:04:39] [I have joined #xf-bod]
+[12:06:06] <Lyude> We've actually got enough people already that we can skip rolecall, XDC2022, XDC2023, GSoC/Outreachy/EVoC, status from FDO admins, rebranding, SFC, hybrid talks
+[12:06:13] <Lyude> oops, *agenda for today:
+[12:07:17] <Lyude> So for XDC2022/2023: I think the only update from me is that I sent out another set of reminders for XDC2022 registration/CFP
+[12:07:31] <Lyude> danvet: also you were asking again about the LPC throw mics, right?
+[12:07:54] <danvet> oops sorry
+[12:07:57] * danvet here too
+[12:08:07] <danvet> yeah I pinged ivyl again, but no reply yet from either ivyl or guy
+[12:09:03] <Lyude> Gotcha. Oh, I also forgot to add the MIPI stuff to the agenda: although there's no updates there really
+[12:09:16] <Lyude> and mfilion - I guess for the branding stuff we're still waiting on SFC, or?
+[12:11:21] <Lyude> (will BRB)
+[12:15:20] <Lyude> Alright back, we'll skip that item for the moment I suppose. Regarding SFC, I sent out the questionaire, and SFC will be taking a look at it during their next meeting in june
+[12:15:54] <alyssa> nya!
+[12:16:07] <Lyude> tlwoerner: you around for GSoC/EVoC stuff?
+[12:18:26] * rg3igalia be back in ~15m
+[12:19:20] <Lyude> I guess we can skip this item for now too
+[12:20:25] <mfilion> yeah not doing anything until we sort out SFC
+[12:20:34] <mfilion> or at least get a sense of where it will go
+[12:21:02] <Lyude> gotcha. So status from FDO admins: I don't know if everyone's looked at the emails yet but tl;dr, we had a bunch of skiddies try to do bitcoin mining on the gitlab instances, bentiss has pretty much put a stop to it for the time being but as a result we currently have to have manual account registration approval (which is flooding my inbox…).
+[12:21:22] <mfilion> eesh
+[12:21:33] <alyssa> f'naaa >.>
+[12:21:50] <Lyude> So what might end up happening is we might have to make it so that people who want to contribute put in a request for an account which, is sub optimal compared to everyone being able to just register willy nilly, but we're not totally sure if there's any way around that
+[12:21:57] <Lyude> especially since it seems like gnome has already had to do that
+[12:22:42] <danvet> yeah I think that takes away a huge benefit of gitlab making it easier to contribute
+[12:22:55] <danvet> bentiss is trying to figure out some miracles meanwhile
+[12:23:25] <danvet> but yeah I guess everyon once in a while someone will figure out how to convert our server budget into bitcon in support of global warming :-(
+[12:23:28] <Lyude> Yeah, I've been trying to think of other things as well. I still wonder if we should look into seeing if cloudflare or friends might be able to help with this a bit
+[12:25:01] <alyssa> how would cloudflare help
+[12:25:14] <alyssa> it's a permission issue isn't it?
+[12:25:30] <Lyude> alyssa: well there's technically two issues, that, and the fact we just have a pretty big influx of spam in general
+[12:25:44] <alyssa> Currently we give permissions to "anyone with an account" to "use large amounts of compute"
+[12:26:04] <Lyude> Like besides just the bitcoin mining spam account registration has definitely been an ongoing issue
+[12:26:08] <alyssa> whackamole aside
+[12:26:13] <alyssa> (is whackamole like guacamole?)
+[12:26:41] <Lyude> But yeah, I think if we could just get permissions on who's allowed to execute runners that would certainly help
+[12:26:41] <alyssa> long term either we need to restrict account registration or tighten access controls for CI ... neither solution is palatable, I admit.
+[12:27:29] <Lyude> I think I'm at least more in favor of just restricting CI runner access
+[12:27:49] <alyssa> sure
+[12:27:53] <alyssa> was that something we can do though
+[12:27:58] <alyssa> i thought that was an EEgated feature
+[12:28:01] <alyssa> unless i misunderstood the emails
+[12:28:12] <Lyude> No, that's automatic resource management - e.g. allowing compute, but not too much of it
+[12:28:19] <alyssa> oh, ok
+[12:28:38] <alyssa> for starters the 24h timeout should be, like, 2 hours tops..
+[12:28:45] <alyssa> and even then like
+[12:28:56] <alyssa> if you're using a runner for that long something has gone SEVERELY wrong
+[12:29:24] <alyssa> although i'm also cognizant of any changes to CI timeouts may result in CI pains for mesa and friends down the line given how much we're already hosing CI
+[12:29:46] <alyssa> though mesa pipelines are torched at 60min as i understand
+[12:30:09] <alyssa> iunno
+[12:30:11] <alyssa> f'naa
+[12:30:36] <Lyude> tbh, I feel like when you get down to it the only time CI really becomes a nessecity is for active development, or once a new contributor has submitted something and it needs to be checked for quality. the first one we can handle pretty easily through requests, the second one doesn't really matter to anyone but project maintainers and the person making the commit
+[12:30:48] <Lyude> so I guess it honestly doesn't seem like hosing it that much to me?
+[12:31:10] <Lyude> I don't think having to wait a little longer to hit a button to start CI will really push anyone away from contributing imo
+[12:31:29] <Lyude> as long as people actively using it regularly can get a pass from that
+[12:32:39] <Lyude> I guess if other folks could comment on bentiss's email that'd be good, so they know how we want to move forward
+[12:32:48] <alyssa> i might not be understanding then
+[12:32:49] <alyssa> my bad
+[12:33:09] <Lyude> alyssa: it's np, anything in particular you're confused about?
+[12:33:27] <Lyude> I'm waiting for rg3igalia before the next item anyway so we have time
+[12:34:52] <rg3igalia> Back, regarding CI
+[12:35:40] <rg3igalia> I know nothing about the topic but the possibility of not making it available until someone approves sounds good to me
+[12:36:05] <rg3igalia> That doesn't prevent a new contributor from sending patches, right?
+[12:36:13] <Lyude> yeah, my thinking exactly
+[12:36:57] <rg3igalia> And cuts like 90% is the problems, I guess
+[12:37:07] <rg3igalia> Of the*
+[12:37:18] <rg3igalia> (On mobile)
+[12:37:20] <Lyude> mhm, there's still spam accounts but I feel like if we have the CI issues under control that one is less pressing to deal with
+[12:37:22] <Lyude> no problem
+[12:37:58] <Lyude> I guess we'll continue the discussion over email then, I'll just poke people occasionally to make sure everyone's at least seen it and forward the input so far
+[12:38:48] <Lyude> anyway rg3igalia not sure how much feedback we'll get since the meeting seems kind of quiet today despite reaching quorum, did you want to talk about your questions w/r/t hybrid presentations?
+[12:39:42] <rg3igalia> Yeah, basically the submission page says that precoded or virtual talks can be sent, but in-person talks will be strongly preferred
+[12:40:34] <rg3igalia> However, there's nothing in the form to indicate the type of talk you're submitting, which will or should be important for y'all selection after the deadline
+[12:41:41] <Lyude> rg3igalia: I think we might just have to ask folks, but it would be nice if we could get something like that addex
+[12:41:55] <Lyude> ( also cc: mupuf ^ any idea if we might be able to do something like that?)
+[12:41:57] <rg3igalia> I just wanted to know if we could add a checkbook or similar, yes
+[12:42:54] <rg3igalia> Because asking to 30+ people doesn't sound fun
+[12:43:11] <Lyude> yeah, I think we might need to poke mupuf tbh
+[12:44:07] <danvet> +1 on checkbox
+[12:44:15] <Lyude> rg3igalia: after the meeting I will send an email out to mupuf and cc samuel on it just in case, I think that's probalby going to be the easiest way to get an answer
+[12:44:21] <danvet> iirc lpc did it like that in the first year of the pandemic
+[12:44:35] <rg3igalia> Thanks 👍
+[12:44:44] <Lyude> I do also have the indico admin info now so I should be able to look into it possibly as well
+[12:44:48] <danvet> well drop list is what lpc did, with 3 options "remote only" "physical presentation only" "either is fine"
+[12:45:01] <danvet> rg3igalia, you're volunteering to add this?
+[12:45:24] <danvet> maybe should also have a note that we're ok with a best guess, since the situation is fluid
+[12:45:37] <Lyude> danvet: I was just going to ask around for them, I don't think they have the privileges to - although I probably could add them to the admin list on indico
+[12:45:38] <danvet> so if someone who plans to attend but might cancel last minute
+[12:45:57] <danvet> Lyude, I mean if you want to do that, but it's a pile of work usually
+[12:46:08] <danvet> like just getting the info out of folks for travel grants is a pile of work
+[12:46:18] <danvet> or the confirmation mails for all accepted talks
+[12:46:25] <Lyude> danvet: I meant asking mupuf/samuel about adding a checkbox or option
+[12:46:39] <danvet> ah I'd have volunteered rg3igalia after the access stuff is sorted :-)
+[12:46:47] <danvet> it's pretty easy to add to the cfp
+[12:46:51] <Lyude> ahh, I can just do that then :P
+[12:46:59] <rg3igalia> I know nothing about indico/web programming, if it can be added easily I can do it, yes
+[12:47:24] <danvet> since samuel has some newborn to handle and mupuf seems permanently busy :-)
+[12:47:29] <Lyude> rg3igalia: I will set you up an admin account today if I can (otherwise I'll poke mupuf to see how to add you)
+[12:47:38] <Lyude> danvet: yeah good point
+[12:47:52] <danvet> Lyude, you don't need indico admin, only added as an organizer for the current event
+[12:47:57] <Lyude> ahh gotcha
+[12:48:01] <danvet> indico admin is only needed to create a new event
+[12:49:38] <Lyude> anyway - that's it for the meeting everyone, thanks!
+[12:50:01] <rg3igalia> 👋
+[13:54:36] <alyssa> i should strongly consider eating
+[13:54:39] <alyssa> er
+[13:54:40] <alyssa> wrong buffer
+[15:33:58] <anholt> Lyude: probably 1-2 more meetings I'll miss at the current time, then hopefully be freed up. though I'll usually be a little late because 9am is the earliest possible time I can be home.
+[15:34:53] [disconnected at Wed Jun 8 15:34:53 2022]