bug#62207: 29.0.60; Trying to remove non-existent key binding instead adds a binding

[Stefan Monnier]

>> I think the removal code is "good enough" for uses in a user's init
>> code, but I'd rather add some recommendation to avoid on it in
>> ELisp packages.
>
> If we want to recommend that, we need to come up with a rationale and
> a couple of alternative ways of achieving goals that commonly lead to
> using this option.  Can you tell what to say to that effect?

I don't know of "goals that commonly lead to using this option", sorry.
AFAIK it's not used commonly (after all, it's brand new in Emacs-30).

Most of the requests for it I have seen were either of the "make the API
complete" kind of argument, or examples like Jonas' (which are
inherently tightly tied to a particular keymap and a particular binding
in that map, so the chance of bumping into an odd corner case is
significantly reduced).


        Stefan