^X (was: an odd question

From: Paul Williams <celigne_at_tinyworld.co.uk>
Date: Sun Aug 12 11:31:36 2001

Iggy Drougge wrote:
>
> Paul Williams skrev:
> >
> >F11 only generates raw ESC in VT100 mode, which has other drawbacks.
>
> What was DEC's logic behind the exclusion of a proper escape key?

I'm not sure what you mean by "proper" escape key. If you would like a
key that sends a code that is interpreted by applications as meaning
"cancel" or "exit insert mode", that is fine. However, this key
shouldn't send the control code ESC and expect the application to do
something with this code alone. That usage should have been considered
obsolete by the first issue of ASCII in 1963. ESC alters the meanings of
a certain number of following characters.

I would think that DEC's exclusion of a key that sends ESC alone was a
consequence of placing a number of function keys on the keyboard. If you
put a keys "F1" to "F20" or "Select" or "Help" on the keyboard, you
quickly run out of 7-bit codes that they could send. Using ESC as an
introducer gives you an extensible system, in the spirit of ANSI X3.64.
(I know X3.64 doesn't define keyboard codes, but the mechanism is
identical).

I think Bill Joy developed vi on an ADM-3a, but I don't know anything
about the control codes for that terminal, so I don't know whether it
might have been considered safe to use a lone ESC in that situation.

As for Emacs' use of ^S to mean something other than XOFF, that is plain
broken.
Received on Sun Aug 12 2001 - 11:31:36 BST