Newsgroups: rec.arts.int-fiction
Path: gmd.de!Germany.EU.net!mcsun!uknet!doc.ic.ac.uk!agate!howland.reston.ans.net!gatech!usenet.ufl.edu!usenet.cis.ufl.edu!caen!batcomputer!munnari.oz.au!metro!usage!sea.spectrum.cs.unsw.oz.au!cnquinn
From: cnquinn@cse.unsw.edu.au (Clark Quinn)
Subject: Re: A bill of players' rights
Message-ID: <1993May30.050147.28813@usage.csd.unsw.OZ.AU>
Sender: news@usage.csd.unsw.OZ.AU
Nntp-Posting-Host: sea.spectrum.cs.unsw.oz.au
Reply-To: cnquinn@cse.unsw.edu.au (Clark Quinn)
Organization: none
References: <50.1014.679.0N000155@freddy.ampr.ab.ca> <87873@ut-emx.uucp>
Date: Sun, 30 May 1993 05:01:47 GMT
Lines: 72

In article <87873@ut-emx.uucp> silver@metonymy.ots.utexas.edu (silver Harloe) writes:
>In article <50.1014.679.0N000155@freddy.ampr.ab.ca>  
>neil.k#d#.guy@freddy.ampr.ab.ca (Neil K. Guy) writes:
>> gdr11@cl.cam.ac.uk (G.D. Rees) writes:
>> > [...]  But which of these games would you rather play:
>> 
>> >    You are on the south side of a chasm.
>> >    >north
>> >    You plunge to your death.
>> >    *** You have died ***
>> >or:
>> >    You are on the south side of a chasm.
>> >    >north
>> >    You decide not to throw yourself to your death.
>>
>>  The former. The latter makes presumptuous assumptions. Maybe I *do*
>> want to kill myself for some obscure thanatoid reason. Then, if I'm
>> playing a well-written game that uses an UNDO feature like that built
>> into TADS, I simply undo the last move. No problem.
>
>Well, perhaps I'm in the wrong zone here, but I still think:
>
>	You are on the south side of a chasm
>	>north
>
>	Do you really wish to jump off the edge of the chasm?
>
>is the optimal answer... it just kind of makes sense to me.

It makes more than sense.  In interface design, we recognize that people
make mistakes.  A key error (particularly if you have supported your
user by giving them single keystrokes for common commands) has some
finite probability.  That's a mistake, intending to hit one key and
accidentally hitting another.  There's also a mode error, where your user 
thinks they're at some other location.  Or a memory fault, where the
user has forgotten which direction was in the room description as the
way *away* from the !@#$% chasm.  Oops.  There are a variety of
different types of slips that could lead someone to indicate an action
that they don't want.  Now, in most cases there aren't fatal
consequences, so we don't need to prompt for *every* action, just
dangerous ones.  Remember, in the real world we seldom walk close enough
to a chasm edge that one mistake on our part puts us over the edge, we
usually give ourselves a margin of error (why do we put railings on
balconies and people creep on hands and knees to peer over the edge of
unguarded cliffs with unknown drops?).

Some systems also prompt *for dangerous and irreversible actions* (some
do more, and they're considered obnoxious).  So it's not unreasonable for 
there to be a query at a time that someone wants to do something that
would generally be considered foolhardy.  If you *really* want to jump
over the cliff, you just answer in the positive, but if you've been
playing along and made a mistake or a slip that wasn't a bad decision
because of game info and your judgement but was one of the side-effects
of the cognitive architecture we've inherited you'd hate to pay *that*
price.  

A few thoughts based upon _the psychology of everyday things_(1), -- Clark

(1) D. A. Norman "The Psychology of Everyday Things", Basic Books 1988 
or "The Design of Everyday Things" in paperback.

---------------------------------------------------------------------------

 _--_|\    Clark N. Quinn
/      \   School of Computer Science and Engineering	    +61-2-697-4034
\_.--._*   The University of New South Wales            Fax +61-2-313-7987
      v    PO Box 1                                cnquinn@cse.unsw.edu.au
           Kensington, NSW                             C.Quinn@unsw.edu.au
           AUSTRALIA 2033                                      

           "A California Yankee in the Land of Oz"

