| 21 Jul 2022 |
John Ericson | they are still doing things I understand, but yes this should be driven by the foundation | 14:18:46 |
John Ericson | it is some technological work but also more importantly believing that multiple approaches is good, vs everyone must use exprs or flakes or whatever | 14:19:21 |
John Ericson | I do think nickle or hnix or something will give us types | 14:20:02 |
John Ericson | not C++ nix | 14:20:09 |
John Ericson | and then we'll really be able to refactor nixpkgs | 14:20:17 |
infinisil | I think this is approaching off-topic territory :) | 14:21:07 |
j-k | I think there are still some easier wins within our grasp for improving nixpkgs itself but I do generally agree with the changes you want for nix itself | 14:21:27 |
infinisil | John Ericson: What do you think of having a call to just pick your brains on how the current stdenv needs to be redesigned and how to go about that? | 14:22:46 |
John Ericson | Sure | 14:23:05 |
infinisil | Because I think you know a lot of things, but not many other people do | 14:23:12 |
infinisil | It might be good to start with the premise of "We have Nix but no nixpkgs, how to compile a C program on various platforms?" | 14:23:42 |
John Ericson | In reply to @j-k:matrix.org I think there are still some easier wins within our grasp for improving nixpkgs itself but I do generally agree with the changes you want for nix itself Yes I do agree, just whenever we reach some sort of sticking point like "are setup hooks good?" I immediately find myself distracted :) and longing for the better future | 14:24:08 |
John Ericson | In reply to @j-k:matrix.org I think there are still some easier wins within our grasp for improving nixpkgs itself but I do generally agree with the changes you want for nix itself * Yes I do agree, just whenever we reach some sort of sticking point like "are setup hooks good?" I immediately find myself distracted (I'll admit it! :)) and longing for the better future | 14:24:23 |
infinisil | In reply to @infinisil:matrix.org It might be good to start with the premise of "We have Nix but no nixpkgs, how to compile a C program on various platforms?" John Ericson: Is there maybe already some document that explains stdenv from the ground up or is your brain the main source for it right now? | 14:25:36 |
Robert Hensing (roberth) | In reply to @yorik.sar:matrix.org I think we should optimise the evaluation process, not try to avoid it... Not to pick out a single comment, but we're discussing rather many suggestions to prioritize even larger projects. Ambition is nice, but this is also how we get stuck in ivory towers. | 14:25:53 |
John Ericson | infinisil: probably not yet | 14:26:46 |
John Ericson | but happy to get more stuff written | 14:26:54 |
infinisil | Yeah let's try to keep the discussions here limited to "What can we do with the current Nix" | 14:26:58 |
j-k | It was mentioned before about bootstrapping and hopefully using stage0 and kaem etc like guix team will/already(?) do. I'm happy to help in that effort & I think this intro to stdenv would help me understand what already exists and what needs swapping out.
I am also very interested in the avoiding-stageN-nonsense stuff | 14:27:28 |
Alyssa Ross | Does that mean not the all-packages.nix restructuring from the other day, where a new builtin was being designed? | 14:27:31 |
infinisil | John Ericson: Sounds good to have a call then, I'd love to dig and experiment with this myself, maybe also write some stuff | 14:27:31 |
infinisil | Alyssa Ross: Hehe yes | 14:27:51 |
Alyssa Ross | j-k: I'm very interested in that too. I would really like it for Nixpkgs to be a lot more committed to bootstrappibility. | 14:28:04 |
infinisil | Caught me there 😅 | 14:28:06 |
infinisil | John Ericson: When are you available? | 14:28:51 |
infinisil | Maybe let's have a when2meet.com | 14:29:04 |
John Ericson | I can do right now if you want | 14:29:54 |
John Ericson | or when2meet | 14:29:56 |
infinisil | Oh neat, yeah let's do it now then | 14:30:26 |
infinisil | I'm in https://meet.jit.si/nixpkgs-architecture | 14:31:26 |