[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#48764: [External] : bug#48764: mini-buffer completion
From: |
Drew Adams |
Subject: |
bug#48764: [External] : bug#48764: mini-buffer completion |
Date: |
Tue, 1 Jun 2021 16:00:47 +0000 |
> > But the mini-buffer does not provide the user with any indication that
> > he can use the "up" and "down" keys to navigate a list of settings.
> > It would be beneficial to introduce some kind of glyph or indicator in
> > the mini-buffer to make the user aware that arrow keys can be used,
> > and be described in the manual.
>
> It should not come as a surprise that basic navigation keys allow you to
> navigate basically.
>
> Closing.
I wouldn't argue that the minibuffer itself needs to
provide any such indication. (Some doc should perhaps
mention it, however.) I agree with you about that.
But the reply that users should expect "basic navigation
keys" to "navigate basically" presents things in a
distorted way.
Basic navigation keys generally do work in the basic,
i.e., ordinary way in the minibuffer. But <up> and
<down> do NOT do so. Their normal/basic behavior is
to move up or down a line of text. They _could_ do that
in the minibuffer, but they don't. Likewise, `C-n' and
`C-p'.
So while I tend to agree that the minibuffer itself
need not advertise what <up> and <down> do, the argument
that you gave is specious, IMO. In fact, it argues the
opposite: it argues that <up> and <down> (and `C-[n|p]')
should "allow you to navigate basically", i.e., move up
and down a line of minibuffer text.
Not to mention that _cycling_ candidates is not a basic
navigation behavior. It's not a navigation behavior at
all. It doesn't move point or the mouse pointer, doesn't
move focus, or do any of the things one might think of as
"basic navigation" behavior.
bug#48764: [External] : bug#48764: mini-buffer completion,
Drew Adams <=