Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!ima!haddock!karl
From: karl@haddock.UUCP
Newsgroups: comp.emacs
Subject: Re: gnumacs bindings
Message-ID: <1238@haddock.ISC.COM>
Date: Thu, 24-Sep-87 21:12:41 EDT
Article-I.D.: haddock.1238
Posted: Thu Sep 24 21:12:41 1987
Date-Received: Sat, 26-Sep-87 13:37:42 EDT
References: <3720007@hpsemc.UUCP> <28500018@ccvaxa>
Reply-To: karl@haddock.ima.isc.com (Karl Heuer)
Organization: Interactive Systems, Boston
Lines: 38

In article <28500018@ccvaxa> aglew@ccvaxa.UUCP writes:
>[haddock!karl writes:]
>>If you remove the language-dependent mnemonics, what sort of mnemonics do you
>>have left?
>
>Question mark, "?", is pretty universal.

True.  But I believe RMS stated that he wanted a single-character HELP key,
which rules out your suggested C-x ? and ESC ?.  C-? would be great, if such a
character really existed.  (Some terminals do generate C-_ when you hit
CTRL-?, but not all -- and emacs shouldn't depend on it%.  The convention of
denoting DEL by ^? is irrelevant, even if DEL were not otherwise taken, since
the mnemonic value is lost on someone who doesn't know that convention.)

I'd have no objections to ESC ?, except that it's a bit hard to type%% and
it's annoying if you have to use it very often.  There's also the problem that
help-char has to be a single byte; this is part of a more general problem%%%.

>Unfortunately, for EMACS, "\e?" is a prefix for many function keys.

If you're worried about function keys, you shouldn't use ESC at all!

Karl W. Z. Heuer (ima!haddock!karl or karl@haddock.isc.com), The Walking Lint
________
%Actually, emacs should also quit assuming that CTRL-SPC generates C-@.
________
%%I was once part of a project that used that sequence for HELP -- we
discovered that since "?" had to be shifted, users would often shift
prematurely and get a shifted ESC, which on the HP terminals would generate
DEL, halting the program.
________
%%%One of the main reasons I didn't bind a function key to isearch-forward,
back when I was having trouble getting C-s to pass through, was that I
couldn't use the same function key for search-repeat-char.  This suggests that
either emacs should use strings instead of chars for certain objects, or that
emacs should read function keys as out-of-band integer values (a la curses).
I prefer the second approach, which would also let me bind my ANSI function
keys without disturbing backward-paragraph.