bug#44341: 27.1; define-minor-mode generates inaccurate docstring

[Drew Adams <drew.adams@oracle.com>]

> We should definitely signal an error here. A form such as (my-mode
> 'enable) actually disabling the mode is very confusing. The mode
> function needs to check for the various cases anyway, it might as well
> use `cond' and signal an error in the non-matching case.

Not to argue, but this kind of thing is all over
Emacs Lisp.

The ability to use an unspecified non-nil value,
to mean/do something that might work against the
natural-language "meaning" of the name of a
symbol argument, is just one example of the vast
amounts of rope that Lisp gives its users to hang
themselves with.

Do you really think `define-minor-mode' should
be fiddled with specially here, to prevent use
of an unfortunately named symbol arg to disable
the mode?

`define-minor-mode' and its doc, and the doc of
minor modes, are already complex/confusing enough.
Do you think fiddling to eliminate confusion over
poorly named symbol values won't actually add to
that confusion?