ucwcs on Debian testing/unstable
Very long time ago, several UCW sages congregated on the ultimate Czech programmer layout – they ended up with a US keyboard just like usual, except that capslock doesn’t lock case (good in itself!) but acts as a second shift, adding diacritic mark on the letter. So caps+s produces Å¡, caps+e produces é and caps+w produces Ä›. I have been using that ever since, but when moving to another system, setting this up always was a bit of a challenge.
But currently, the X11 Czech ucw layout is easy to set up on squeeze/sid (I’m currently using squeeze with xorg from sid – works fine). Xorg keyboard configuration is HAL-driven on sid currently, what you need to do is put this to /etc/default/console-setup:
XKBLAYOUT="us,cz" XKBVARIANT=",ucw" XKBOPTIONS="grp:caps_switch"
(Though, the way this works, using two groups, I’m not quite sure how to add another layout with reasonably seamless switching. Not my problem, but some other users might want that.)
Probably silly question, but anyway: Why doesn’t capslock just make a switch to czech layout?
Because czech layout has actually quite a small subset of czech characters available directly, since it has to coexist with all the other characters. If you use a special shift state, you can have many more characters available, even actually not Czech ones – in the past I used caps+v for cé cédille quite a lot. Besides, it’s not like the ucw mapping would be hard to remember at all.
For people that are dumb like me, the commands for non-debian systems is:
‘setxkbmap us,cz -variant ,ucw -option grp:caps_switch’
Also, in the most recent X11 servers, there’s a dvorak UCW present, too! Use:
‘setxkbmap cz -variant dvorak-ucw’ for it. Right Alt switches to accents.