Path: utzoo!utgpu!news-server.csri.toronto.edu!dgp.toronto.edu!kevins
Newsgroups: comp.sys.amiga
From: kevins@dgp.toronto.edu (Kevin Schlueter)
Subject: Re: Reality check: Amiga coverage is not a right, but a privilege
Message-ID: <1990Dec19.192642.17584@jarvis.csri.toronto.edu>
Organization: CSRI, University of Toronto
References: <1990Dec13.155848.8152@maytag.waterloo.edu>
Date: 20 Dec 90 00:26:42 GMT
Lines: 40

In article <1990Dec13.155848.8152@maytag.waterloo.edu> giguere@csg.uwaterloo.ca (Eric Giguere) writes:
>Don't flame me for this, but.... most people DON'T need multitasking on
>their computer.  What they do need is a way to switch rapidly between
>applications and/or share data.  Very few people are TRULY doing two
>concurrent things.  (Print spooling is about the only thing I can think of
>that the average user will want to do.)  This is why MultiFinder on the
>Mac, the ultimate kludge, is successful.  It works!  (Well, usually.)  It's
>why Windows could get away with "co-operative multitasking".  It's why
>the IBM world proliferates with TSRs.
>
>Eric Giguere                                       giguere@csg.UWaterloo.CA

Don't worry, this isn't a flame :-).  I'm not sure about `most people',
but I sometimes need multitasking on a computer.  Just a few minutes ago,
I was waiting for a somewhat involved Prolog program to finish a computation
and decided to catch up on my email.  If I had been using multifinder,
my Prolog program would most likely have stopped computing, which would have
been pretty undesirable.   Cooperative multitasking would probably have not
been any better, as how many Prolog interpreter writers surrender control
of the CPU in their computation loops (doing this in a completely general
way requires fairly intimate knowledge of program execution, which often
no one programmer in a large project has).

My other argument in favour of multitasking is that it simply isn't intuitive
for a program to stop when its window isn't active or its screen is not visible.
Granted, the word `intuitive' isn't quantifiable, but I suspect that those
who only do one thing at a time on a computer do so only because they learned
this behaviour when they learned about computers (mental set for using a
computer).

Personally, this is the kind of topic I'd like to see more discussion of
in this news group.  I'm a PHD student in Human - Computer Interaction,
and would be interested in examples of situations where multitasking
clearly leads to a better interface (or even arguments as to why it is
unnecessary).