Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84 +2.11; site dcl-cs.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!think!harvard!seismo!mcvax!ukc!icdoc!dcl-cs!de
From: de@dcl-cs.UUCP (David England)
Newsgroups: net.cog-eng
Subject: System aided input
Message-ID: <267@dcl-cs.UUCP>
Date: Tue, 25-Jun-85 04:03:38 EDT
Article-I.D.: dcl-cs.267
Posted: Tue Jun 25 04:03:38 1985
Date-Received: Fri, 28-Jun-85 00:11:30 EDT
Reply-To: de@dcl-cs.UUCP (David England)
Organization: Department of Computing at Lancaster University.
Lines: 23
Xpath: icdoc ivax

Eat me.
	We're doing some work here on improving the user interface to
existing unix tools. One of the problems we were discussing is how far you
can take the do-what-i-mean approach in an environment for non-casual
users. For example you could prompt for or correct missing or invalid
parameters but how far do you want to take this ? I come down on the 
user-centred view that the system should not try to second guess the user 
but only do what was clearly intended, others here disagree slightly saying 
that you should build more 'knowledge' of the system into commands to aid 
the user.  Ideally the system should be able to adapt to the requirements of
different levels of user-skill but that's another story ...

Does anyone out there in net.land who has been working on predictive
systems or syntax-directed editors have any views or some user reaction to 
some system control of user interfaces?  
Any replies (via E-mail) gratefully received.

-- 
UUCP:	...!seismo!mcvax!ukc!icdoc!dcl-cs!de
DARPA:	de%lancs.comp@ucl-cs		| Post: University of Lancaster,
JANET:	de@uk.ac.lancs.comp		|	Department of Computing,
Phone:	+44 524 65201 Ext. 4586		|	Bailrigg, Lancaster, UK.
Project:Alvey ECLIPSE User Interface	|	LA1 4YR