Newsgroups: comp.ai.philosophy
From: Lupton@luptonpj.demon.co.uk (Peter Lupton)
Path: cantaloupe.srv.cs.cmu.edu!nntp.club.cc.cmu.edu!godot.cc.duq.edu!hudson.lm.com!news.pop.psu.edu!news.cac.psu.edu!howland.reston.ans.net!swrinde!pipex!demon!luptonpj.demon.co.uk!Lupton
Subject: Re: Strong AI and consciousness
References: <3c7dvj$ab5@news1.shell> <420521987wnr@luptonpj.demon.co.uk>
Distribution: world
Organization: No Organisation
Reply-To: Lupton@luptonpj.demon.co.uk
X-Newsreader: Newswin Alpha 0.6
Lines:  51
X-Posting-Host: luptonpj.demon.co.uk
Date: Sun, 11 Dec 1994 00:01:50 +0000
Message-ID: <555383080wnr@luptonpj.demon.co.uk>
Sender: usenet@demon.co.uk

In article: <3c7dvj$ab5@news1.shell>  hfinney@shell.portal.com (Hal) writes:
> 
> Lupton@luptonpj.demon.co.uk (Peter Lupton) writes:
> 
> >The classifications we make will tend to be made on
> >the basis of data compression. It is most unlikely 
> >that such classifications will be eliminable in the 
> >strong sense required by iff definitions.
> 
> >Most of our classifications will be flaky at the edges -
> >and for good reason. The classifications we make are 
> >related to our ability to compress data - the sort of 
> >cases which expose the flakiness of our classifications 
> >are just *not relevant* to the data compression problem. 
> 
> While this is true, you don't want to go too far in the other direction and
> say, therefore, that there is no way to tell whether something is a tree
> or not, or a computer or not, or a particular program or not.  

Absolutely. There is no doubt in my mind that the machine I am 
currently typing into is a computer. There is also no doubt that
the rock I kicked earlier was not.

> Granted
> there are gray areas, but there are non-gray areas as well.  A tree is
> not running an AI program, and a computer running Lotus is not a tree (or
> a flower :-).

Right.

> 
> Our classifications do have power even though they are imprecise and
> imperfect.

I don't accept this wording, althought I think I agree with the intent.
Our definitions may be as precise as they can be. No more, no less. That our 
classifications are flaky at the edges is not a sign that they are 
imperfect (although they very often will be). We cannot improve our
classifications to avoid marginal cases - the attempt will just shift 
the margin.

Our classifications do, indeed, have power - the point I am making
is that the AC approach to classification owes nothing to naive realism 
and its bed-fellow, logicism. There is no danger, in the context of AC,
of being accused of saying nothing just because one is unable to give
'iff' definitions. Why was that classification made? - Because it 
compresses. That's explanation enough.

Cheers,
Pete Lupton
