Re: $ collision

[Richard Riley <rileyrg@googlemail.com>]

Russ Allbery <rra@stanford.edu> writes:

> Richard Riley <rileyrg@googlemail.com> writes:
>
>> I guess this is where I am getting confused. Esc is the meta here or I
>> have certainly always thought of it as that .. but that to one side for
>> a moment : the rest of the emacs driving keys use M so do you redefine all
>> those or just use Esc as the Meta?
>
> I use ESC as meta, but most meta sequences in Emacs don't have to be
> repeated multiple times.  The only other one I ever remember noticing this
> with is M-v, and I don't use that anywhere near as much as marking
> messages as spam.  In order to have M-d be a single shifted keystroke,
> there has to be a shift key on the keyboard that can successfully act as
> the meta shift key (not the ESC prefix workaround), which is true in X but
> not (for me at least) in an xterm through ssh to emacs -nw on the other
> side.
>
> If one is marking multiple messages as spam, it's much more annoying to
> have to alternate between two keys (and ESC is often in an awkward
> location) than to just repeatedly use the same key, possibly shifted.
>
>> Don't misunderstand me, I'm not objecting per-se to anything I am just
>> trying to understand why existing keys that people use for Mairix are
>> remapped to add a facility key to something already mapped.
>
> I've never even heard if nnmairix before the last month and certainly
> wasn't aware of it when I proposed the additional key binding, and I
> suspect Lars wasn't aware either.
>
> I don't use it and therefore don't particularly care; I can continue to
> locally map $ to the right thing since I don't use nnmairix.  But it
> seemed like it would make mark handling more consistent.

That would explain why the change wouldn't concern you I guess. I did
use nnmairix for a long time until I discovered (recently) the beauty of
fts_squat in local dovecot server and now use nnir search.