Tim Pope

FWIW `[]` and `][` override built-ins.

The interpreter needs carriage returns and the interpreter gets carriage returns. Not seeing a problem there. Is it the removal of leading whitespace that is tripping you up?

Feels to me like 2 would be the more appropriate choice? That's the value that most accurately implements the syntax highlighter's intentions.

I could see making `coe` toggle `2` but `[oe` and `]oe` decrement and increment.

The limitation here is we're running low on maps. `]h` isn't great because the linewise map would be `]hh`, which is ambiguous with "decode one character to the left".

Could you test if `virtualedit^=all` to prepend it fairs any better?

I do kinda like the `;`/`,` idea but that's a pretty invasive thing to experiment with this late in the game.

I find this wording confusing. Would changing "unchanging" to "globally unique" do the trick?

These are cool but I'm not sure where they would fit. It's tempting to swap them out for `:cnext` and `:cprevious` but I think it's a bit too soon for...

If it's scoped to the current file then yeah, we probably shouldn't replace the existing mappings. Linter warnings is also my use case, but that means `:labove` and `:lbelow`, which...