Yes, that would be an outrageous claim! That is, of course, not what I said.

Arguing that channels lead to more problems than flakes is a good argument in favour of adoption of flakes. But you can also abandon channels without adopting flakes.

Which is what I said: flakes don’t solve any problems that can’t be solved without flakes.

At this point I don't even quite remember what would be the sane alternative without flakes but I am happy to discuss...

I have a repository with system configurations for some CI infrastructure: a build server, a test runner.

The test runner can either be generated as an SD-card image using nixos-generators or live-updated using a remote `nixos-rebuild switch`. The OS configuration contains stuff about purposing specific motherboard GPIO pins.

Both systems depend on custom software not in nixpkgs; these are just hosted in private git and have a flake that mainly provide a devShell, but also provide a cross-compiled binary.

Flakes handle all of that in a predictable way: OS images, cross-compiled binaries, devShells, cross-repo linking, convenient hash-pinning.

2) Overlays only solve issue of adding your own packages to an existing channel

3) System channels and user channels are two different things.

4) Many times I've updated my home-manager profile and forgot to update system profile and it borked due to channels being out of sync (user error, but flakes remove that foot gun)

5) Very easy to have portable dev-env. If a system has nix installed, just typing `nix develop` in my repo would put you in the exactly same dev environment as me. In most cases it would byte for byte identical. I'm not going to tell you to install 100 of dependencies, not going to bother you with what application is written in, all you have to do to build it locally is to type `nix build .#`. I'm not even going to bother you how to run test because `nix check` will run them.

6) Flakes provide some schema, you know here nixos or home-manager modules would be.

7) Flakes are easy to compose together

8) I can have identical env on CI, production and my local machine without any extra overhead - flake.lock takes care of this.

All of this is extremely predictable: I got a new laptop, using nix-anywhere I've installed nixos on it, that had pretty much identical look and feel of my desktop. It all boils down to - channels suck and hard to use.

Eval caching depends on pure eval and pure eval was previously not possible because channels are by design the most impure and cursed thing and are just a bandaid that lasted way to long.

The scaling with big repos gets worse as some use flakes instead of their normal build system which is not the intended use case. You still do development like normal with the normal build system.

Flakes are still unstable because everyone wants something different and there are still breaking changes planned which then everyone would wine about and maintaining backwards compatibility is a pain if you still change fundamental things.