Newsgroups: comp.robotics
Path: brunix!uunet!cs.utexas.edu!utgpu!watserv1!watmath!mwtilden
From: mwtilden@watmath.waterloo.edu (Mark W. Tilden)
Subject: Re: This anthropomorphism thing...
Message-ID: <1992Feb22.200218.23621@watmath.waterloo.edu>
Keywords: ha!, anthropomorphism, bugs, neurons
Organization: University of Waterloo
References: <42542@dime.cs.umass.edu> <1992Feb3.232008.26712@watmath.waterloo.edu> <kpoc04INNf0g@pollux.usc.edu>
Date: Sat, 22 Feb 1992 20:02:18 GMT
Lines: 71


In article <kpoc04INNf0g@pollux.usc.edu> bucur@pollux.usc.edu (alexander bucur) writes:
>In article <1992Feb3.232008.26712@watmath.waterloo.edu> mwtilden@watmath.waterloo.edu (Mark W. Tilden) writes:
>>...Robots with
>>a strong anthropomorphic base are, frustratingly enough, usually
>>seen as 'toys'.  Yearg.
>
>We are not trying to generate show-stoppers, are we?  For that kind 
>of thing, Hollywood has it all.  Nobody needs to invent it!

Hollywood also has space movies, but we don't see a thriving space 
industry from it (not directly, anyway).  Regarding both space and 
robotics, Hollywood has only served to make a difficult standard to 
live up to.  Se la vie.  

Here's a thot though, could this explain the failure of industrial
robotics in the 70s and the success of the PC today?  Anyone can see 
that industrial robots do not behave like they should, whereas 
many believe the salesman when he sez 'the PC can do Everything!'.
I postulate that Hollywood has reinforced the robot stereotype to 
such a degree that many a promising robotics implementation has
been severely compromised, while reinforcing the belief of sentient 
omnipotence amongst computers in general.

I can cite numerous examples.  Anybody else got horror stories
that would support/refute this?

>>...It's not essential, but it is 
>>much easier to build a robot with a 'run away!' instinct than a 'I 
>>must do my duty to my master' concept.
>
>...I can just see a robot
>who analyzed that it has to save his/her energy, therefore it stops working
>in the middle of a process since it found out that it takes away from its
>energy.  Isn't this a possible scenario for your idea of autonomous robot?

Yes, but no more so than realizing if you don't get some water, you're
going to fall right on your honker.  Such behavioral responses are typically
the 'last straw on the camel' type, so we can assume that our metal 
worker will keep going to the last second it can.  If a machine ever got
smart enough to think "why the hell am I doing this?", then I suggest
we start worrying, but that's not likely for a while.

>Have you looked yet at how hard is to make autonomous kids to follow certain
>working rules? Do you want that in a robot?  And if you obtained it, then 
>what is the purpose of constructing such a new being?  Do we need more
>beings on welfare? ...

Here's the dream for such devices:

A sea-urchin overpopulation is quelled by a small army of robots
which act as a replacement for their natural enemys.  A farmers
field has a thirty percent increase in crop yield without pesticides
because small robotic mechanisms protect each plant against intruders.
A house owner realizing his place needs a clean, reminds himself to drop
by the Hardware store to pick up a handful of credit-card sized robots
which will invisibly keep his house clean and his cockroaches nervous.

There won't be any welfare problems, just permutations of useful
robot species which can preform slow but persistant chores.

That's the plan anyway, military applications notwithstanding (ouch).

Is all.


-- 
Mark Tilden: _-_-_-__--__--_      /(glitch!)  M.F.C.F Hardware Design Lab.
-_-___       |              \  /\/            U of Waterloo. Ont. Can, N2L-3G1
     |__-_-_-|               \/               (519)885-1211 ext. 2454 
"MY OPINIONS, YOU HEAR!? MINE! MINE! MINE! MINE! MINE! AH HAHAHAHAHAHAHAHAHA!!"
