| 29 Dec 2024 |
 emily | I always heard that there's no real way to test ofborg changes other than running it in prod, is that still true? | 20:32:40 |
 Janne Heß | Yep, one of the pain points :D | 20:32:50 |
| Janne Heß | It's alwys nvim && cargo c && nix flake lock && colmena apply | 20:33:03 |
| emily | I guess it shares that with GHA to some degree tbf :) | 20:33:03 |
| emily | yuck | 20:33:10 |
| emily | I admire your persistence | 20:33:15 |
| Janne Heß | Yes unfortunately | 20:33:17 |
| Janne Heß | I regret it :D | 20:33:23 |
| emily | GHA at least has the docs / tooling / approachability advantage, I suppose, in terms of future directions for what we want to drive CI | 20:34:05 |
| emily | too bad about the lock-in | 20:34:10 |
| Janne Heß | Yeah, that's my exact opinion. But if we are being honest, how likely are we to ever break out of that lock-in? | 20:35:25 |
| emily | well, 0 likely if we use that reasoning at all key turning points :) | 20:37:34 |
| emily | the GitHub UX causes me pain every single day I do Nixpkgs stuff, so I would personally be very happy if we got Gerrit or something. but yeah, I don't know if it's a lost cause or not. | 20:37:59 |
| Janne Heß | It is imo | 20:38:29 |
 7c6f434c | >0% of just eventually getting forced out for one reason or another… | 20:38:41 |
| emily | eh, it's not hard for me to imagine a world where we have {gitlab,forgejo}.nixos.org with GH login. I think at least one or two people on the steering committee indicated they were interested in pushing in that direction | 20:40:06 |
| 7c6f434c | (like, some CI limit gets restructured) | 20:39:11 |
| emily | unfortunately it's precisely non-GH-cloning code review UI that I'd like. but even GitLab/Forgejo would require some CI re-engineering. | 20:40:26 |
| emily | ideally we'd have relatively thin GHA wrappers around more agnostic tooling | 20:40:39 |
| Janne Heß | That's what the Lix people are doing with ofborg afaik. We'll see how that goes | 20:41:36 |
| emily | I do sometimes wonder how much Nixpkgs costs GitHub to host | 20:42:41 |
| emily | I guess they get some prestige from hosting a big repository in return, but I feel like our name recognition is not so high that that could never become an issue. | 20:43:06 |
| Janne Heß | Yeah, they probably care about vscode a lot more | 20:45:02 |
| Janne Heß | here we go: https://github.com/NixOS/ofborg/pull/701 | 20:45:06 |
| Janne Heß | my broken Gist code still needs to be pushed though… | 20:45:22 |
| 7c6f434c | Well, you don't need name recognition if there is a metric that allows to justify a «one of the top open-source projects on GitHub forced out» headline, so we might get some slack from nobody wanting to go on record that the savings are worth it. | 20:47:51 |
 raitobezarius | If by Lix people, you meant me, you are free to ping me :p | 21:46:25 |
| raitobezarius | (it's not made for Lix only fwiw, it's a side effect.) | 21:46:47 |
| Janne Heß | Well, at the point where I'm willing to put additional work into it, yes | 21:53:14 |
| Janne Heß | It'd be great if we combined our efforts at some point and the sooner the better but I'm really annoyed by the current state of ofborg (code and evaluator, not people) | 21:54:12 |
