From mboxrd@z Thu Jan  1 00:00:00 1970
Date: Thu, 11 Sep 1997 13:49:32 +0200
From: elliott Elliott.Hughes@genedata.com
Subject: [9fans] Questions
Topicbox-Message-UUID: 630caaba-eac8-11e9-9e20-41e7f4b1d025
Message-ID: <19970911114932.HEaBb6i98QMxecTBHYxS9kM2DONnCEL6FmgVbsAjmyo@z>

Steve_Kilbane@cegelecproj.co.uk wrote:
> There are two aspects to that. Firstly, much of Plan 9 hangs together
> because agreed conventions are obeyed (namespace construction, for
> example). Secondly, the mouse is used to move the pointer, which is
> just a convention, but it's one that's assumed at a pretty low level
> of many systems. Principle of Least Surprise.

you're bound to have surprises as long as you make mistakes. it seems to
me that the Plan 9ers consider there to be a difference between
"character keys" and "control keys", but they fail to make this
distinction in the software. just because Unicode has all those
hang-over bits of ASCII that fit many of their needs (Tab, Esc,
Backspace, Del, Return) they thought they could get away with their
dualism. the place where they fall over is the right-hand side of the
keyboard. they end up _feeling_ that the arrow keys are control, but
having left themselves no way to say this. but this _feeling_ makes them
incapable of using the keys to produce the Unicode arrow characters
(say).

but hey, even if they _did_ do that, where would they be with Insert or
Home or End?

[I remember seeing an IBM experiment where they projected a keyboard
onto the desk. Great for switching between Latin and Greek character
sets, but I can see a lot of uses for a comfortable dynamic keyboard. If
only one existed.]

> (Incidentally, this is also a reasonable argument for Del being a
> delete key; I much prefer the way libXg just accepts either backspace
> or delete as the delete key.)

The "delete key"? Confusion reigns ;-)

-- 
Elliott Hughes - GeneData AG, Postfach 254, CH-4016 Basel, Switzerland
mailto:elliott.hughes@genedata.com  http://users.ch.genedata.com/~enh/