1.0 Roadmap

[nrdxp]

Is your feature request related to a problem? Please describe.
Obviously, if we are providing a template, we want to give users a solid idea of where they stand, and a clear upgrade path. Of course, this isn't urgent as flakes and the nix-command still haven't reached a stable release, but by the time they do, I'd like to have reached 1.0.0 (or close).

There are several usability improvements I'd like to see, so no need to rush. I went ahead and tagged 0.7.0.

Third draft goals for 1.0:

• A stable api and optimized structure Simplify devos #137
• Refactor and reduce lib
• stop pushing directly to core (my fault)
• modules api, convertable to os, devshell, nm etc (in Dhall?). rexport profiles as integration modules #127 Incorporate ./shells #38
• generalize hosts
  • Add Nixos integration tests
  • The current, base override model is incredibly useful, but shouldn't be the only option. Generalize hosts/default.nix to allow for hosts with different versions. Allow setting of channel host-wide #117
• generate lib documentation from the comments — shouldn't be terribly difficult to genaralize this example
• Something solid for Home-manager external modules from flakes #106
• The remaining docmentation work finished revamp docs #111
• make defaultTemplate a derivation that builds the mdbook for local reference and removes the source doc files
• A solution for semi automatic update for /pkgs #118 — linked PR pkgs: automatic source management #135
• One line boostrap, even without nix — nix-user-chroot already has that goal, so maybe we could help there.
• Some tolerable solution to secrets Encrypt with (r)age #37, make pass state part of this repo's structure #56
• Solve #91 — upgrading without the git history would be fantastic.
• Definitely need to implement incomplete substituters file #98 — this will probably wait until the stable release of nix-command.
• Trustix — when it's ready.

---

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

[tgunnoe]

I was just about to suggest the same thing.

I cannot follow daily changes, when I'm changing my own, and at the same time figuring out how to upstream the useful parts to the template. I didn't even realize there was a new name for this project.

One suggestion would be to not try to abstract flakes so much. I havent used the shell with flk related commands, because nixos-rebuild switch/test/boot --flake '.#hostname' is more direct relationship of understanding on a NixOS system.

'"If you try to hide the complexity of the system, you'll end up with a more complex system". Layers of abstraction that serve to hide internals are never a good thing. Instead, the internals should be designed in a way such that they NEED no hiding.'

... I think flakes and nixos-rebuild fit that description pretty well.

there's so many great ideas in this template and its been very helpful. I hope to be caught up with the updates and contribute back soon!

[nrdxp]

Totally agree, flk is there mostly as a band-aid I always hoped to peel off as the ui moved toward stability.

Thanks for the insightful quote 👍

[codygman]

I cannot follow daily changes, when I'm changing my own, and at the same time figuring out how to upstream the useful parts to the template.

Yeah, a few hours ago I tried to resolve conflicts between my branch that diverged at e67cd37 with master (currently 2cb1ec6) and I had to put it down for the weekend.

I'm not sure if there's a good way to make that easier though... maybe once I move to suites/etc that will make it easier.

[Pacman99]

I think it will generally be easier once the api is stabilized and the format of the repo changes less often. And switching to suites will probably help.

[nrdxp]

For anyone subscribed to this, I have went ahead and tagged 0.7.0 so you can have a stable upgrade target. I'll probably release a few 0.7.x release shortly containing a lock file update and some doc fixes.

[Pacman99]

Closing this as most of it is done or no longer applicable. Anything important can get its own issue.