!VyoUhyWvlhSpFWWxHL:matrix.org

Zulip setup coordination

92 Members
Coordination to setup https://nixpkgs.zulipchat.com/, see https://github.com/NixOS/foundation/issues/14329 Servers

Load older messages


SenderMessageTime
25 Feb 2024
@raitobezarius:matrix.orgraitobezariusRegarding contention, I'd like any "deep dive" regarding this to be backed with an analysis of the Fastly logs and the effective popularity of store paths, this would make the discussion much morei nteresting IMHO20:01:53
@jade_:matrix.org@jade_:matrix.org
In reply to @raitobezarius:matrix.org
But some proprietary redistributable software can have some consequences on those resources, e.g. GPU ecosystem with 38 copies of BLAS, one per CPU manufacturer/family/weird new instruction arch, etc.
re this, this is due to the habit of antisocial and silly packaging practices in proprietary software being more common, i guess?
20:08:31
@raitobezarius:matrix.orgraitobezarius
In reply to @jade_:matrix.org
re this, this is due to the habit of antisocial and silly packaging practices in proprietary software being more common, i guess?
That's true, but it has a clear consequence on us unfortunately
20:09:48
@jade_:matrix.org@jade_:matrix.orgyes, and we would like to limit such consequences of course20:10:10
@jade_:matrix.org@jade_:matrix.orgit sucks that CUDA is 1gb ;p20:10:33
@raitobezarius:matrix.orgraitobezariusWe are basically in a computational social choice theory land20:10:33
@raitobezarius:matrix.orgraitobezariusThis is a fair resource allocation problem of a common good20:10:42
@raitobezarius:matrix.orgraitobezariusAnd what we are trying to attempt is something like Pareto-efficiency20:10:49
@raitobezarius:matrix.orgraitobezariusThat's the "theory setting", now there's all the discussion and feelings and what we believe in to settle too20:11:04
@jade_:matrix.org@jade_:matrix.orgtbh we maybe should have ofborg say how large packages are in prs, related to this20:11:55
26 Feb 2024
@janik0:matrix.org@janik0:matrix.org * Hi, so I want to self nominate as a board observer.
I'm already quite involved with the NixOS foundation, attending all the governance workshops at lasts NixCon. I'm one of the Organizers for this years NixCon EU and organized the NixOS Foundations participation in this years Google Summer of Code and also being interested in other more adminstrative tasks like or the transparency on affiliations or the general sponsorship situation. The reason why I'm asking to become a board observr is that I dm'd raitobezarius a bunch of times in the last half a year or so to get more insight knowledge about some topics then the meeting notes offer and I'm very keen on making process and decisions surrounding the foundation more transparent.
Also I currently do not have any financial ties to Nix/OS or the foundation and I'm aware that the foundation meeting is every two weeks Tuesday evening (at least in my local time) and would also have time to attend those.
Couldn't find a documented process or similar for this, this Is why I'm asking here, it would be awesome if you could consider my nomination at the next board meeting.
00:42:18
@adis:blad.is@adis:blad.isReplying in a thread because it's semi off topic but https://github.com/NixOS/nixpkgs/pull/272179 could fix that by delaying the python composition 03:14:51
@me:linj.techlinjCould you provide a reference link? My search yields an opposite opinion. https://lists.gnu.org/archive/html/guix-devel/2018-04/msg00054.html03:51:06
@delroth:delroth.net@delroth:delroth.netGeneral meta comment: I feel like these RFC-style discussions would be significantly more productive if we let stating concerns about a particular topic to those who actually have some amount of ownership about said topic. If you're not working on the cache / the build infra and you think the unfree change would have an impact on those things, ask (in private or on their team channel) the people involved to have a look and potentially comment, but it's very useless to get rabbit-holed in discussions like the one above - it's just noise and uncertainty.07:08:13
@jade_:matrix.org@jade_:matrix.orghttps://issues.guix.gnu.org/48924#1407:14:10
@zimbatm:numtide.com@zimbatm:numtide.com
In reply to @janik0:matrix.org
Hi, so I want to self nominate as a board observer.
I'm already quite involved with the NixOS foundation, attending all the governance workshops at lasts NixCon. I'm one of the Organizers for this years NixCon EU and organized the NixOS Foundations participation in this years Google Summer of Code and also being interested in other more adminstrative tasks like or the transparency on affiliations or the general sponsorship situation. The reason why I'm asking to become a board observr is that I dm'd raitobezarius a bunch of times in the last half a year or so to get more insight knowledge about some topics then the meeting notes offer and I'm very keen on making process and decisions surrounding the foundation more transparent.
Also I currently do not have any financial ties to Nix/OS or the foundation and I'm aware that the foundation meeting is every two weeks Tuesday evening (at least in my local time) and would also have time to attend those.
Couldn't find a documented process or similar for this, this Is why I'm asking here, it would be awesome if you could consider my nomination at the next board meeting.
Sounds good! I added your nomination to tomorrow's foundation agenda and will get back to you afterwards. Typically we setup a call for nominees so all the board members get to know you and understand where you are coming from, what are you interested in, how we could work together, ...
08:50:41
@aloisw:kde.org@aloisw:kde.orgDoesn't really sound like rejecting systemd itself to me, more like rejecting a systemd package from someone who explicitly stated that they will only use it for proprietary software.08:55:55
@raitobezarius:matrix.orgraitobezariusneat13:21:25
@janik0:matrix.org@janik0:matrix.orgRedacted or Malformed Event13:31:37
@raitobezarius:matrix.orgraitobezariusI'd like to add for the next team representative meeting the request to sunset and archive the Hydra repository20:17:03
@raitobezarius:matrix.orgraitobezariusThere's no real ownership on that repository and it is effectively mostly abandoned like some other nixos/ org repositories20:17:18
@raitobezarius:matrix.orgraitobezariusIt does receive fixes as part of some people usage of it, but there's no coherent development team behind it anymore20:17:33
@raitobezarius:matrix.orgraitobezariusIt generates confusion for some newcomers as to "what is the CI solution in the Nix ecosystem" and when they encounter Hydra, it's always a bit, meh20:17:49
@raitobezarius:matrix.orgraitobezariusThis comes off as a recent remark from the infrastructure team in https://matrix.to/#/!RROtHmAaQIkiJzJZZE:nixos.org/$cvAZDLrJZVW6-u99gS-tkdyZJIMxL7jKTNfxqnh_AWk?via=nixos.org&via=matrix.org&via=nixos.dev which mentioned the possibility to take hydra.nixos.org to a fork or another branch for their needs20:18:19
@raitobezarius:matrix.orgraitobezariusBeyond hydra.nixos.org and other one-off private deployments, it is unclear to me whether it's worth to continue to have Hydra acting as some sort of official (effectively unmaintained) solution for CI in Nix ecosystem20:18:49
@joepie91:pixie.town@joepie91:pixie.townTIL that it is unmaintained20:19:38
@patka_123:matrix.orgpatka joined the room.20:20:08
@joepie91:pixie.town@joepie91:pixie.townI do feel like there should probably be some sort of 'blessed' or at least canonical solution to "produce a binary cache for a given set of expressions" but I don't see a reason why that would need to be Hydra specifically20:20:33
@raitobezarius:matrix.orgraitobezariusThe alternative is also to do a call to form a new team to maintain Hydra20:23:26
@raitobezarius:matrix.orgraitobezariusor at least clearly identified people who would like to take up the maintenance20:23:42

Show newer messages


Back to Room ListRoom Version: 10