Hi John
Sorry it's taken so long to get back to you - computers, kids and
birthdays (not mine) and other crap getting in the way. Anyway, where
was I? Oh yeah....
I can't find my copy of Penroses book so I've been trying to remember
exactly why I thought it was a pretty poor read and I think it was
partly because I was expecting a book mainly to do with AI and instead
ended up with a book on popular science! Which was a bit disappointing.
I also found it very odd that he spent most of the book talking about
almost everything except AI. His explanations of various aspects of
science - quantum computers, mandelbrot set, turing machines etc. - were
very interesting and informative (also relevant to some parts of the
degree course I was doing at the time) but I don't think they were
entirely relevant to AI and refuting strong AI.
What Penrose was trying to specifically disprove or refute was John
Searle's idea of strong AI - and if I remember correctly, used the
chinese room example to do this. I won't go into it here, as you can
find much better analyses on the net, but this combined with his ideas
about chess playing computers, was where he managed to completely miss
the point of a lot of AI research and get some of his refutation
completely wrong.
I'm not saying that he didn't bring up a whole load of very relevant
points because he did, but the general approach, to my mind, seemed a
bit long-winded and not quite relevant.
To be honest though, I would probably agree with him that the top-down
approach to AI (including the sort of reductionist approach of early AI)
is unlikely to produce something like a human intelligence in the near
future. The bottom-up approach of people like Rodney Brooks, Patti Maes,
Luc Steels and others is where, in my opinion the smart money should be.
If you're interested then check out Rodney Brooks and the subsumption
architecture - it's fascinating and producing some amazing results. This
was a whole area that Penrose didn't address, although, to be fair, at
the time he wrote The Emperors New Mind it was only just getting started
so it's understandable.
Unfortunately I know very little of Pryor and Q'm but from what you say
you may well find what Brooks et.al. have to say is relevant.
Anyway, I hope that answers a few of your questions and if I haven't
answered what you want to know then fire something back
Cheers
Horse
On 19/04/2010 06:10, John Carl wrote:
Horse,
You said,
Penrose's arguments (especially those in "The Emperors New Mind" - the one
that specifically targets AI) were unravelled and dismissed a long time ago
- I read it when it came out and thought it was a pretty poor read.
and it seemed that I threw in the towel myself with:
I can see that this argument is not going to go my way at all. :-)
But I'm pretty interested in the gist of why you think Penrose is wrong.
But then I got to think while I was driving, and I changed my mind. Or
rather, in spite of the unlikelihood of the argument "going my way", I wish
to pursue it anyway. Despite your superior expertise, I think you're wrong.
Of course, we have to define exactly what we're talking about Of course AI
is possible. It's more than possible, it's actual. Artificial
intelligence is all around us. I often wear a wrist calculator watch
because they're very handy on the job. Especially if you can get one with a
sq root function!
The ability to calculate quickly that we term "intelligence", is not only
possible, its factual.
Artificial intellect, however, is another story entirely. And here I need
make no reference to Penrose, it is Pryor's logical argument derived from
the Q'm" with which you must contend.
For thereby, there is only one way to attain intellectual status, and that
is by encompassing the totality of the inorganic, biological and most
importantly, social, ways of being.
For even as biological life comes to naught with a dearth of necessary
inorganic "patterns" upon which to feed, and Social patterns die soon when
their biological substance subsists, and all intellectual patterns die when
no social support is given them, so it is and always must be,
epistemologically speaking that you can't create intellectual patterning
from inorganic matter and pure intellectual ideas.
For intellect is born of self / other dichotomy nurtured into existence with
maternal attention and biological support. There's no way any increase of
mere intelligence is going to re-create that process which is the heart of
autonomous self-dom, self-realization or value-cognizance. However you
wanna spell it, call it consciousness, call it "I and I" for all "I" care.
It comes only through this social creation called infant nurture, coupled
with biological analogues of felt pain and desire. The whole body is a
brain, when it comes down to our deepest understandings, and a brain hooked
up to an environmental matrix that can't be copied by algorithm, program or
artificial replication.
And all the expertise in the world will not obviate this fundamental truth.
So there.
Cheers,
John
Moq_Discuss mailing list
Listinfo, Unsubscribing etc.
http://lists.moqtalk.org/listinfo.cgi/moq_discuss-moqtalk.org
Archives:
http://lists.moqtalk.org/pipermail/moq_discuss-moqtalk.org/
http://moq.org/md/archives.html
--
Life should NOT be a journey to the grave with the intention of arriving safely in an
attractive and well preserved body, but to skid in sideways, chocolate in one hand, wine
in the other, body thoroughly used up, totally worn out and screaming "WOO HOO what
a ride!"... Hunter S Thompson
Moq_Discuss mailing list
Listinfo, Unsubscribing etc.
http://lists.moqtalk.org/listinfo.cgi/moq_discuss-moqtalk.org
Archives:
http://lists.moqtalk.org/pipermail/moq_discuss-moqtalk.org/
http://moq.org/md/archives.html
