Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!rutgers!sri-spam!ames!sdcsvax!sdics!norman
From: norman@sdics.ucsd.EDU (Donald A. Norman)
Newsgroups: comp.sys.mac
Subject: Re: A Design Philosophy(& coffee)
Message-ID: <401@sdics.ucsd.EDU>
Date: Mon, 12-Oct-87 11:05:04 EDT
Article-I.D.: sdics.401
Posted: Mon Oct 12 11:05:04 1987
Date-Received: Tue, 13-Oct-87 05:39:27 EDT
References: <4670@oberon.USC.EDU> <21248@ucbvax.BERKELEY.EDU>
Reply-To: norman@sdics.UUCP (Donald A. Norman)
Organization: UC San Diego Institute for Cognitive Science
Lines: 44
Keywords: coffee
Summary: Error can be designed for

In an earlier note I argued that correct design philosophy would solve
the errors of incorrect disk insertion (rather well done now, by the
way), of not knowing why the apple menubar symbol was blinking, and
similar issues.  (The notification manager sounds like the correct approach to
the alarm clock, mail, and other notification problems.)

Now to take care of coffee spills. A respndent complained that trying
to avoid all human frailty would require a warning about not spilling
coffee into the top of the minitor.

How do you design against that kind of error?  There are several
answers.  (Why do you care?  Because the same philosophy and approach
will also lead to better interfaces and program interaction with the user.)

First, any sensible task analysis would indicate that people really do
drink things while working, that these things spill, and that
therefore they had better be guarded against.

One solution: forbid drinking in the workplace (a common approach.  I
fordbid it myself.  I violate my own rule.)

Another solution: design the workplace so that there is a place to put
the coffee cup, coke can, ...      Any airline cockpit or Macintosh
workplace that doesn't leave room for the drink container is asking
for trouble.

Finally, design the top of the monitor so that it won't support a
cofee cup or coke can.  Make it rounded, or tilted or pointed.  (That
would also stop people from placing their manuals on top of the
monitor, thus blocking the ventilation and cooking the electronics.)

This note really belongs in cog.engineering, but that would require
establishing the entire context of the discusion in that newsgroup.  A
problem of design, again. 


Donald A. Norman
Institute for Cognitive Science C-015
University of California, San Diego
La Jolla, California 92093
(619) 534-6770 (UCSD)
(619) 481-9191 (home)
norman@nprdc.arpa    	{decvax,ucbvax,ihnp4}!sdcsvax!ics!norman
norman@sdics.ucsd.edu	norman%sdics.ucsd.edu@RELAY.CS.NET