Experimental backends and tasks project status

[jdx]

I'd like to set some expectations around experimental backends (meaning anything non-asdf or core plugins). These are experimental features: they have bugs, they have regressions too, they will go through breaking changes. I know people love them, and I'm excited for them to eventually come out of experimental but they aren't because I haven't had time to complete the work around them. Tasks are in a similar state but even further out since I definitely plan to finish backends before really even thinking seriously about getting tasks out of experimental. I don't anticipate tasks shipping out of experimental until 2025 at the earliest.

Specifically, when it comes to tools that rely on other tools, this has been a hard problem since mise needs to shift from being a simple manager of tools to working more like a bonafide package manager. This has meant a lot of refactoring has needed to take place to handle what might look like simple use-cases. That refactoring alone is still a WIP.

Unfortunately my summer is looking very busy. I don't think I'll be able to get it out of experimental within the next couple of months at least. I also don't think anyone else could manage to ship this and even if they did, I'd still have to code review it.

So just be patient—we'll certainly get there one day. At some point it will work like asdf plugins do now where it handles pretty much all the use-cases we throw at it and we don't need to make anymore breaking changes but that day is not today.

Should you use them? That just depends on your appetite for encountering potential problems. Most users don't have too many problems and the ones they do they can work around.