Re: ^O in EMACS

Henry Spencer (attcan!utzoo!
8 Nov 88 22:35:27 GMT

In article <8811012301.AA08441@ETN-WLV.EATON.COM> mcc@ETN-WLV.EATON.COM (Merton Campbell Crockett) writes:
>...more specifically "why?". In a rhetorical vein, why does
>EMACS, in general, use standard control characters as application dependent
>function characters? Why would any application?

Basically, because it is very useful for an application which can involve
text entry to have command characters which cannot be mistaken for text.
This avoids the need to have some sort of mode switch to distinguish text
from commands. Users quite reasonably tend to feel that all printable
ASCII characters are legitimate text, so a command character must be
a control character, or a sequence thereof. Moreover, commands get typed
a lot, so it's got to be something simple, preferably a single keypress.

The ASCII standards give most of the control characters quite specific
meanings; in particular, communications systems are **NOT** guaranteed
to be transparent to a lot of them. Perhaps unfortunately, most real
communications hardware is pretty transparent. This is a bit less true
than it used to be -- XON/XOFF in particular is a lot more common than
it was ten years ago -- but is still accurate overall. So it is very
tempting to use ASCII control characters as command characters, and
there is little incentive to hardware manufacturers to provide better
alternatives. (There is no reason why holding down control and hitting
O should not send something like ESC O, which networks *are* required
to be pretty much transparent to, but existing keyboards don't do that.)

It's an awkward decision for people writing interactive software, text
editors in particular.

The Earth is our mother.	|    Henry Spencer at U	of Toronto Zoology
Our nine months	are up.		|uunet!attcan!utzoo!henry

This archive was generated by hypermail 2.0b3 on Thu Mar 09 2000 - 14:44:30 GMT