| 21 Apr 2025 |
 Tristan Ross | I thought Ofborg had the Ampere boxes from Hetzner | 23:35:16 |
 emily | those are for Hydra | 23:35:30 |
| Tristan Ross | I thought we had some for Hydra and some for Ofborg | 23:35:59 |
| emily | oh hm | 23:36:10 |
| emily | yeah I guess that changed | 23:36:13 |
| emily | originally it was just x86_64-linux + aarch64-darwin | 23:36:22 |
| emily | I thought ofborg wasn't using any Foundation machines though | 23:36:38 |
| emily | maybe that changed | 23:36:40 |
| Tristan Ross | Yeah, I have seen aarch64-linux jobs showing up for Ofborg | 23:36:51 |
| Tristan Ross | In reply to @emilazy:matrix.org
I thought ofborg wasn't using any Foundation machines though I'm not sure | 23:36:57 |
 Randy Eckenrode | I agree it’s not fair, but that’s what I see from Darwin. When there’s a conflict between platforms, x86_64-linux gets the priority. | 23:56:23 |
| Tristan Ross | In reply to @reckenrode:matrix.org
I agree it’s not fair, but that’s what I see from Darwin. When there’s a conflict between platforms, x86_64-linux gets the priority. Yeah, ideally they should get equal treatment | 23:57:45 |
| 22 Apr 2025 |
| emily | if they had equal userbases, that would make sense | 00:00:56 |
| Tristan Ross | Hmm, maybe we should see if the surveys the marketing team puts out could ask people what kind of hardware they run Nix or NixOS on | 01:25:53 |
| Tristan Ross | I've sort of drafted this https://pad.lassul.us/9yYTVp73QBul7dCm6tCWsw?view | 01:41:29 |
| Tristan Ross | The left side is like lower tiers which means better changes of things working. | 01:42:26 |
| Tristan Ross | So like riscv32-linux-gnu is sort of considered not really supported | 01:42:46 |
| Tristan Ross | While x86_64-linux-gnu has full support | 01:42:57 |
| emily | surely aarch64-darwin should have user adoption rather than x86_64-darwin? | 01:45:41 |
| emily | they haven't produced an x86 Mac in years | 01:45:48 |
| Tristan Ross | 🤷 | 01:46:00 |
| emily | (also I don't think cross-compilation is a criterion? maybe I'm misunderstanding) | 01:46:25 |
| Tristan Ross | I consider being able to do cross a sort of base criteria | 01:48:31 |
| Tristan Ross | Some systems may be exempt (like Darwin) for technical reasons | 01:48:50 |
| Tristan Ross | The first two criteria I consider to be a baseline support for it in nixpkgs | 01:49:13 |
| Tristan Ross | It means you can get a system but support is limited | 01:49:28 |
| Tristan Ross | I think we can make an RFC that replaces RFC 46 and make a more formal set of criteria for things. | 01:56:02 |
| emily | I think the RFC 46 criteria are pretty good, esp. in that they primarily define expectations on contributions and support rather than technical proxies for those | 02:00:17 |
| emily | though the ofborg part goes against that a bit :) | 02:00:22 |
| Tristan Ross | Hmm | 02:01:42 |
