Miss Dada  🏳️‍⚧️

right, from my POV 'owned has its own problems here. I kinda like the symmetry of 'static, but I feel like the fixes here are in how we teach this, not in introducing a new thing

I am open to alternative suggestions but I don't know how one would go about teaching people this better than we already do without making changes to the language.

We could add material everywhere but in the end I feel like the easiest way to get ppl to learn this is to make it distinct and require that they look it up. The problem with 'static is that people assume they understand what it means but they really don't.

what we really need is a distinct lifetime that means "until the end of the program" that doesn't make it seem like the memory MUST live until the end of the program, which is the main problem with static. 'forever, 'none, 'end-days, whatever

As long as its distinct from 'static, because to me a 'static lifetime really feels like the one and only concrete named lifetime in rust, to me it means data that just exists forever. Where as a 'static lifetime bound is much more abstract in nature.

It's cool how we looked at the thirteen different interpretations of that keyword in C and said "that sucks" and did it anyway anyway I feel like 'top would be okay for "outlives, but is not in the data segment"

still strongly believe we're using `'static` and `static` to mean the same thing everywhere, it's just that the discovery path for this mental model isn't clear the actual confusing thing is that `'a` and `: 'a` don't work the same way

static the declaration very firmly means "has exactly one address allocated by the linker" while static the lifetime just means "you'll die before it does", which is the set of things allocated by the linker and also by mem::forget