| 7 Jun 2021 |
 maralorn | Just need someone who is motivated to do this. | 12:00:13 |
 pwmosquito | 100% | 12:06:35 |
 sterni (he/him) | I think we should definitely talk to Domen Kožar on this, as far as I am aware they have a stake in all three things concerned: Haskell, macOS and aarch64-darwin specifically | 12:12:17 |
| maralorn | Well, I mean we are not even sure that it’s a huge amount of work, right? I mean theoretically it could just build, right? | 12:13:58 |
| sterni (he/him) | I mean there are bound to be unexpected problems, like me having to spend an entire morning packaging apple's xattr “fork” and stuff like that | 12:20:22 |
| sterni (he/him) | also darwin is not in a perfect state by any means as evidenced by tests.writers and GHC outside of stdenv being broken | 12:20:57 |
| sterni (he/him) | I am getting increasingly frustrated by the situation since I personally have little stake in darwin and am not really well equip to do stuff for darwin as the darwin machine I have is 10 years old now | 12:22:00 |
| sterni (he/him) | but it is not like I can stop caring since if darwin GHC / haskellPackages breaks, the nixpkgs channel stops advancing | 12:22:46 |
 cdepillabout | I don't think it would be crazy to stop maintaining it if none of us want to use it. We'd have to remove it from blocking the nixpkgs channel though. | 12:25:16 |
| cdepillabout | I mean, ideally what we should do is come up with a list of packages we'd like help maintaining, and create an announcement asking for help. I'm imagining mainly darwin and aarch64 stuff. | 12:26:11 |
| sterni (he/him) | cdepillabout: no we can't realistically say that cachix will stop working in the channel | 12:26:23 |
| cdepillabout | We can totally say that cachix will stop working in the channel if no one is willing to maintain it. | 12:26:51 |
| sterni (he/him) | I guess we have to | 12:27:13 |
| sterni (he/him) | but given the current situation and the experience from the last two days, I am pretty sceptical that it is a good idea to add aarch64-darwin support at all if no one says they will maintain it | 12:27:58 |
| sterni (he/him) | because I don't want to get forced into a situation were I have to keep it working | 12:28:12 |
| sterni (he/him) | what is the platform support tier of aarch64-darwin now anyways? | 12:28:35 |
| maralorn | sterni (he/him): peti never even checked if anything was building on aarch64 or darwin. He didn‘t even notice. | 12:28:47 |
| cdepillabout | In reply to @sternenseemann:systemli.org
but given the current situation and the experience from the last two days, I am pretty sceptical that it is a good idea to add aarch64-darwin support at all if no one says they will maintain it Yeah, that is completely understandable. | 12:28:56 |
| maralorn | I think us ignoring this and saying that people who care need to care is totally okay. | 12:29:29 |
| sterni (he/him) | maralorn: I mean I don't want to go back to that attitude, but I can definitely relate more to that nowadays | 12:29:37 |
| maralorn | I mean we shouldn‘t actively break stuff and if something is easy to fix we can do it. | 12:30:05 |
| maralorn | But this is a volunteer project. And a lot of darwin users are kinda commercial. | 12:30:40 |
| sterni (he/him) | well I still don't want to introduce a channel blocking situation by pressing merge | 12:30:49 |
| maralorn | If you can‘t let it rest, then you need a way to find a healthy balance there. But I certainly can. | 12:31:05 |
| sterni (he/him) | but if there is no way forwards, I think we'd have to remove darwin from hydra for all things haskell | 12:31:25 |
| maralorn | In reply to @sternenseemann:systemli.org
well I still don't want to introduce a channel blocking situation by pressing merge Okay, fair. | 12:31:26 |
| maralorn | In reply to @sternenseemann:systemli.org
but if there is no way forwards, I think we'd have to remove darwin from hydra for all things haskell Why? Can‘t we just have failing jobs there? | 12:32:04 |
| maralorn | Or rather just mark everything broken, which is broken? | 12:32:30 |
| sterni (he/him) | nixpkgs-unstable would block on cachix.x86_64-darwin | 12:32:39 |
| sterni (he/him) | I guess we'd have to mark it broken yeah | 12:32:49 |
