Personally I have a complicated relationship with the finalisation…
Flakes as they are right now have their problems, some of them are obvious, some are not.
My current biggest concerns are how the system is dealt with in flakes, and how hard it is with the current interface to make cross compilation work, without either violating the flake output schema or using legacyPackages.
They way how nixosSystems is treated specially by nix flake show/nix flake check but darwinSystems/homeConfigurations or whatever else spits out warnings.
The error reporting when nix can’t find files and reports “random” store paths, rather than refering to the “flake and location within”. There is ongoing work on this.
The way how inputs only used internally can explode consumers lockfiles (recursively). Which is also semi actively worked on.
On the other hand side, a lot of necessary changes can’t be done, as flake became a quasi standard with too much adoption. We are in a situation where breaking changes are a necessity, but impossible, as consumers treat flakes production ready…
So I am continuing to ask to either break or release within 2023. There is no inbetween.