🚀 go-pugleaf

In article <26600@castle.ed.ac.uk> cam@castle.ed.ac.uk (Chris Malcolm) writes:
>In article <1992Oct5.190709.17067@guinness.idbsu.edu> holmes@opal.idbsu.edu (Randall Holmes) writes:
>>In article <26534@castle.ed.ac.uk> cam@castle.ed.ac.uk (Chris Malcolm) writes:
>
>>>I take it that you would have insisted that the program's dispelling
>>>of the illusion was the trickery, and that I had actually succeeded in
>>>writing a conscious computer program in 60Kbytes of BASIC?
>
>>No.  But ELIZA does not in fact have internal representations which it
>>really uses to explain what it is doing to a questioner; its
>>apparently intelligent responses _are_ a trick.  The expert systems
>>referred to by the original posting do, and so, in my view, they are
>>conscious.
>
>The original Weizenbaum ELIZA did have some primitive internal
>representations to help it make sense, such as recently used keywords
>which it could fall back on when it could make no sense of a response.
>
>My BASIC program had more -- It kept track of such things as topics of
>discussion, type of discussion, and length. It used these to do such
>things as try to provoke taciturn responders, to return to good topics
>when a responder seemed to be getting irritated, etc..
>
>So I presume that you think my program was conscious?

Maybe slightly.  Did it use information about what it itself was
doing?

>
>>I think that _everything_ has "qualia", in
>>Cahlmers's sense of having a "bare experiential feel".  Consciousness,
>>for me, has something to do with "knowing oneself" to some extent.  I
>>would recognize any system which maintains an internal model of its
>>own activity and can provide answers to queries about same as
>>(minimally) conscious.  Note that consciousness does not necessarily
>>go with being terribly bright.
>
>It seems that quite a small and simple device would satisfy your
>criterion of consciousness. For example, would a thermostat which had
>a "what state?" enquiry button and two coloured lamps with which to
>reply to the query be (minimally) conscious?

>
>How about if we omitted the button and always made the lamps track the
>internal state? Conscious?
>
>How about if we omitted the lamps and simply took the lid off an
>ordinary thermostat so that enquirers could tell the internal state by
>inspection? Conscious?

Does a thermostat manipulate information about its own internal state
(what it itself is doing, not the ambient temperature?)  If so I would
conclude that it was _minimally_ conscious.  The consciousness or lack
thereof has to do with the existence of internal representations of
its own state (a model of its own activity) not with the ease of
access of engineers to that model.  Having an internal state is not
the same thing; everything has an internal state.

>-- 
>Chris Malcolm    cam@uk.ac.ed.aifh          +44 (0)31 650 3085
>Department of Artificial Intelligence,    Edinburgh University
>5 Forrest Hill, Edinburgh, EH1 2QL, UK                DoD #205


-- 
The opinions expressed		|     --Sincerely,
above are not the "official"	|     M. Randall Holmes
opinions of any person		|     Math. Dept., Boise State Univ.
or institution.			|     holmes@opal.idbsu.edu