| 23 Jun 2023 |
 Dandellion | In reply to @qyliss:fairydust.space
What if we backported the new functions, without a warning? That would maybe work in this case, but as a general policy it can't, it'd be unreasonable to always make secondary functions just to to keep the old name clean for 6 months | 10:43:51 |
 Alyssa Ross | I think we all do :) | 10:44:17 |
 Robert Hensing (roberth) | For trivial changes we can automate it with lib.isInOldestSupportedRelease (better name suggestions welcome) | 10:44:46 |
| Robert Hensing (roberth) | Only for trivial changes though because when we "flip the EOL switch" it creates a bunch of change that should "just work" | 10:45:23 |
| Robert Hensing (roberth) | Grossly underdocumented btw, which is another reason to have a policy, as that would be a good place to have or link such docs | 10:46:08 |
| Alyssa Ross | yeah, this works for when we want to rename a function, not change its behaviour. | 10:46:23 |
| Alyssa Ross | I think this would be a good general policy to adopt for renames | 10:46:39 |
| Alyssa Ross | I need to move the functions under rust into lib.systems soon, so that they're available in make-derivation.nix. Previously I would have done it the same way this PR currently proposes, and not even really been aware of the issues that would cause stable. As a result of this discussion, I now know a better way. :) | 10:48:16 |
| Robert Hensing (roberth) | btw I think we'll want to make pkgs accessible from make-derivation.nix at some point | 10:49:14 |
| Robert Hensing (roberth) | it'd help with things like https://github.com/NixOS/nixpkgs/pull/206728 | 10:49:48 |
| Robert Hensing (roberth) | or the inputDerivation thing, which maybe shouldn't be an attribute, but also has a problem that it's unable to create multiple files because it can't have any sort of dependencies because of the lack of pkgs | 10:51:10 |
| Alyssa Ross | that'd be nice | 10:51:36 |
| Robert Hensing (roberth) | why should make-derivation.nix know about rust though? | 10:51:52 |
| Alyssa Ross | Because, unfortunately, it knows about Meson | 10:52:48 |
| Alyssa Ross | and my attempt to stop it knowing about meson and cmake was met with resistance | 10:53:10 |
| Robert Hensing (roberth) | :'( | 10:53:22 |
| Robert Hensing (roberth) | needs more pkgs-modules ;) | 10:53:41 |
| Alyssa Ross | and if it's make-derivation.nix that sets up the meson cross file, make-derivation.nix has to be able to tell meson what rust target to use | 10:53:52 |
| Robert Hensing (roberth) | fair enough; cause and effect. They'll be complicit ;) | 10:55:58 |
| Robert Hensing (roberth) | I have a meeting now. ttyl :) | 10:57:15 |
| 27 Jun 2023 |
 John Ericson | did we have meeting today? | 14:51:27 |
 infinisil | In reply to @Ericson2314:matrix.org
did we have meeting today? I asked Robert Hensing (roberth) to lead it while I'm on vacation, did it not happen after all today? | 15:13:40 |
| Robert Hensing (roberth) | oh no, I didn't get a notification from my calendar | 16:30:00 |
| Robert Hensing (roberth) | I am so sorry :( | 16:30:18 |
| Robert Hensing (roberth) | I'd even put an extra item for preparation. wth | 16:37:19 |
| 28 Jun 2023 |
 tomberek | i've been out on vacation (and have more coming up in mid-July). apologies for the sporadic attendance | 13:18:20 |
| 4 Jul 2023 |
|  jlesquembre joined the room. | 09:15:16 |
| 5 Jul 2023 |
|  AtnNn joined the room. | 18:22:56 |
| 10 Jul 2023 |
| infinisil | This is the second time that I tried to pass leadership of these meetings to somebody else but it failed for various reasons. Combined with the fact that every meeting it's just me monologuing anyways makes me really wonder how such a team should be run | 13:58:00 |
 Growpotkin | I appreciate you bro! <3 | 13:58:31 |
