3 ms·
The backwards compatibility users require (myself included) holds back innovation, in so many ways. The problem with any non-standard keyboard is it being non-
by Fnoord 8y ago
The backwards compatibility users require (myself included) holds back innovation, in so many ways.
The problem with any non-standard keyboard is it being non-standard. The biggest problem is a different layout than Qwerty (though its a bit more complex; e.g. Azerty is standard in Belgium, Colemak is only marginally different from Qwerty, whereas Dvorak is completely different).
If the layout sticks to the defacto standard (Qwerty) then you can argue less is more. However, there are some minimum needs and once keys have different positions compared to a dumbed down Qwerty keyboard (without keypad or keypad + arrows + the other [6 + 3] keys on the right). It also depends on what you're used to. Having used the keyboard on Nokia E71 and Nokia N900 and Nokia N810 I lost a lot of CLI speed just from factors such as: non-standard keybinds (e.g. requiring combos for standard Bash and Emacs keybinds), non-standard amount of keys, small keys, with not much travel.
There are many examples mentioned in this thread. Some membrane, some mechanical, but generally full size keys just without keys such as spacebar, enter, tab, caps, alt, etc. The only keys which can be removed from a keyboard without compromising UNIX CLI usage are caps lock and "function" (Windows key etc). With the notion that caps lock is already rebinded to escape by Vi(m) users.