From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!csd.unb.ca!morgan.ucs.mun.ca!nstn.ns.ca!aunro!ukma!asuvax!cs.utexas.edu!uunet!psgrain!quagga!g89a2582 Tue Mar 24 09:56:14 EST 1992
Article 4499 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!csd.unb.ca!morgan.ucs.mun.ca!nstn.ns.ca!aunro!ukma!asuvax!cs.utexas.edu!uunet!psgrain!quagga!g89a2582
>From: g89a2582@alpha.ru.ac.za (Gary Allemann)
Newsgroups: comp.ai.philosophy
Subject: Re: Intelligence and Understanding
Message-ID: <g89a2582.700815762@alpha.ru.ac.za>
Date: 17 Mar 92 07:02:42 GMT
References: <1992Mar4.022416.11169@a.cs.okstate.edu> <1992Mar6.011131.4146@ccu.umanitoba.ca> <1992Mar6.051607.13266@a.cs.okstate.edu> <1992Mar6.171417.15561@ccu.umanitoba.ca>
Sender: usenet@quagga.ru.ac.za (Rhodes University NNTP server)
Organization: Rhodes University, Grahamstown, South Africa
Lines: 32

In <1992Mar6.171417.15561@ccu.umanitoba.ca> zirdum@ccu.umanitoba.ca (Antun Zirdum) writes:

>I should have said that it has not been done to any complex
>extent, but it has been done! The mars probe that NASA was
>working on is a good example, it could roam around the
>terrain without getting stuck, falling off cliffs, getting
>out of dead ends, etc... It could also decide where to take
>samples of soil and air, *when it decided*, all in all it
>has the intelligence of a beatle! (not much - aggreed -
>but a start, there is no reason that it cannot be
>upscaled from there!)
 work on mobots at MIT is/was going this way
   i.e. working up from insect level intelligence

>I see now where you are coming from, you really do fear that people
>can create something that is better than they are! Do you seriously
>beleive that people are better at adding numbers than computers,
>how about searching information, how about precision control.
>Computers exceed the capability of people in many areas, I do not
>believe, and no one has shown, that there is any reason that
>this progression cannot continue to other areas of human skill.

Although computers can solve any problem which can be formalized better
than humans it is extremely debatible as to wether they can solve problems
which cannot be easily formalized - eg. natural language recognition - as
efficiently as humans can. Software - by virtue of being a formal system -
is not easily nonformalized.
--
***     Gary Allemann, Rhodes University, STANDARD DISCLAIMERS APPLY         ***
***                                                                          ***
*** There are three types of people in the world, those that can count and   ***
*** those that can't.                                                        ***


