From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!bonnie.concordia.ca!ccu.umanitoba.ca!zirdum Thu Feb 20 15:20:59 EST 1992
Article 3755 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!bonnie.concordia.ca!ccu.umanitoba.ca!zirdum
>From: zirdum@ccu.umanitoba.ca (Antun Zirdum)
Newsgroups: comp.ai.philosophy
Subject: Re: Humongous table-lookup misapprehensions
Message-ID: <1992Feb15.041109.18230@ccu.umanitoba.ca>
Date: 15 Feb 92 04:11:09 GMT
References: <1992Feb12.002312.19459@ida.liu.se> <1992Feb12.145716.22305@ccu.umanitoba.ca> <1992Feb14.015016.21990@ida.liu.se>
Organization: University of Manitoba, Winnipeg, Manitoba, Canada
Lines: 37

In article <1992Feb14.015016.21990@ida.liu.se> c89ponga@odalix.ida.liu.se (Pontus Gagge) writes:
>zirdum@ccu.umanitoba.ca (Antun Zirdum) writes:
>
>>>Is everybody happy that a DFA exists which passes a Turing Test, and
>>>does so in a completely uninteresting manner? Is the Turing Test still
>>>a good criterion for intelligence?
>
>>If it quacks like a duck!
>>All I would like to argue is that if it was possible to actually construct
>>such a table (which it is not!) what is your objection to calling it
>>intelligent???
>
>I do not deny its intelligence *in the limited context for which it was
>designed*. However, it is singularily uninteresting. To me, it is too
>weak to be described by the intuitive concept of "intelligence".

And Pray tell, just what is "the intuitive concept of intelligence"?
I am prepared to concede that if the table is not able to do everything
that you are capable of on the 'net' then it is not intelligent, and
therefore is uninteresting! (I must repeat that the table must be able 
to DO everything that you do! but not necessarily labour intensive things
like working with its hands, etc.. As I have observed that there are people
that are completely paralized but are never-the-less intelligent.)
If the table just wrote what I did, in answer to your message then
I would have to say that it is intelligent (however, some people might 
have difficulty with this :-)
>
>BTW, in *principle* possible certainly does not mean *practically* possible.

Unless we (humans) are able to map four dimensional space-time into a 3D
(time omitted) table in the future sometime, then I would have to say that
the table is impossible even in principle. However, this does not exclude
other machine constructs that are somehow time-dependent (eg. nearal nets,
symbolic processors, etc...) (oops, I meant neural nets above ^)
>
>However, I am curious whether all AI opponents are comfortable with the
>existence of a DFA that *can* behave intelligently.


