Subject: Re: wskbd change
To: Lennart Augustsson <lennart@augustsson.net>
From: Greywolf <greywolf@starwolf.com>
List: current-users
Date: 05/04/2005 13:13:45
[Thus spake Lennart Augustsson ("LA: ") 12:01pm...]

LA: From: Lennart Augustsson <lennart@augustsson.net>
LA: To: Martin Husemann <martin@duskware.de>
LA: Cc: Valeriy E. Ushakov <uwe@ptc.spbu.ru>,
LA:      NetBSD current <current-users@NetBSD.org>
LA: Subject: Re: wskbd change
LA: Date: Wed, 04 May 2005 12:01:39 +0200
LA: X-Spam-Level:
LA:
LA: Martin Husemann wrote:
LA: > On Wed, May 04, 2005 at 03:48:57AM +0200, Lennart Augustsson wrote:
LA: >
LA: >>I will make it a compile time option and the default will be off.
LA: >
LA: >
LA: > I still do not understand what it should mean - you can't realy repeat
LA: > "pressed" events, since that would just be a lie.
LA: >
LA: It might be a lie, but it works in e.g. the X server. :)

To my understanding, the way the X server handles it is, indeed,
event-based, but not directly as an autorepeat.  It's a key-up,
key-down thing.

Whatever is reading said keyboard is what should be handling the
autorepeat, as obviously the hardware itself is not usually generating
repeated keystrokes.  It's a subjective opinion, on the part of the
software, as to whether a "this-key-held-down-for-this-long-or-longer"
should translate to "repeat-the-keystroke-at-this-rate."

Along the lines of keystroke rates, I seem to remember some "dumb terminals"
which would start out repeating slowly (about 4cps) and then accelerate
to their max (about 30 cps) over about a 3 second interval.

I wonder how useful this would actually be.  No, I'm not volunteering nor
am I asking it to be implemented.  This is just a casual toss of the idea
into the ether to see if anyone else thinks it useful.

				--*greywolf;
--
Bought the planet (paid cash).