From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Mon Sep 26 2005 - 16:43:27 CST
From: "Wulfy" <wulfmann@tiscali.co.uk>
> Since the accent key would follow the <compose>key, it would only be
> "dead" in the sense that it produced the combining character, all other
> times, it would produce what is on the key. Someone mentioned the case
> when typing a URL <http://www.example.com/~elaine> (or some such) and the
> dead "~" produced an e-tilde, rather than a "~e". This couldn't happen on
> my proposed map... to get e-tilde, you'd have to *want* to do it...
> <compose>e~.
Bad idea! Almost all time, users of a native keyboard for their language DO
WANT the combination to occur. It's then best to have eventually a more
omplex input for entering the characters when they DO NOT WANT the
combination.
That's why the combination of DEAD KEY + SPACE is mapped and used this way.
But I can accept the fact that a keyboard driver could implement two modes
with either preferences. But the simplest then is to have two keyboard maps
and to switch between them (for example with the language/keyboard bar in
Windows where alternate keyboards can be switched easily with the mouse or
with a user-designated special key combination like Ctrl+Alt+3, made
distinct from AltrGr+3 when dual keyboard maps are active).
This archive was generated by hypermail 2.1.5 : Mon Sep 26 2005 - 16:47:10 CST