Hi all !

I'm going into unchartered OT land, for a while...

Wirt correctly (AFAIK) stressed that the human mind works in parallel,
whereas a computer CPU is more serialized.

I'm not sure we should talk about "algorithms" used by the human mind. I'm
not even sure what to call our inner thought "processes". What I'm
convinced, though, is that the human mind uses analog "signals", instead of
the computer's digital signals. That means it's very easy for the human mind
to tell, say, a rose from a tulip, or instantly identify my daughter's
speech on the phone. As a comparison, a computer would need a still
unavailable amount of CPU power to achieve the same feat in a probably
less-than-perfect manner.

My feeling is that analog systems are very, very appropriate to compare
things, situations, emotions, whatever. In contrast, digital systems are
very, very appropriate to perform number crunching. That's why I'm not
nearly as fast as a computer, and this by a handful of magnitude orders, to,
say, determine a prime number, but the computer will have a hard time to
recognize my five-year-old on the phone, and tell from the almost
imerceptible nuances in her voice, whether she's had a good day or not. Even
with 2 GHz.

Hey, after all, one of the most digitally-reproduceable game of all, chess,
has only recently been won by machines instead of humans. I'm referring to
Deep Blue, a few months (years, already ?) back.

And then, we can add another dimension : the soul. French philosopher
Descartes (and he may or may not have been the first one) once wrote "cogito
ergo sum", which loosely translates as "I think, therefore I am". Do
computers think ? No - They crunch numbers. Do we think ? Yes. Definitely.
Artificial intelligencee, with all the hype around it, is just that -
artificial. At this point, I ought to emphasize that my definition of
intelligence has to include one's conscience. In other words, do computers
know they exist ? Surely not. You can backup your hard disk and restore the
contents to another machine, and all the data and applications and operating
system and who knows what else will be there unaltered. Do we know we exist
? Sure we do. You will be reading this message (that's if you've been
patient enough to read down to this point), and then build your own
interpretation, reaction, thought, criticism, analysis, whatever.

Now what does the 2 GHz figure prove ? Apart from the fact that a 2 GHz CPU
will oscillate in one second about as many times as your heart beats in your
entire life, nothing. What's mind-defying here is not so much the 2 GHz
figure (it will be beaten sooner or later, more likely sooner), but what the
human using the computer can achieve with such power. Here, I voluntarily
and gleefully emphasize the role of the human being at the end of the chain,
which is you and I. Apart from that, we're stuck with numbers. What will
really puzzle me is when the computer really can tell why Bruce Springsteen
was so good when he performed at the MSG last year, or why Mona Lisa's smile
looks so enigmatic, or whether the human life is a superior life form. 2 Ghz
is just another manifestation of Moore's law. It does not defy my mind. It
just makes me think Moore was damn right !

The most exciting part of my job (I'm a part-time systems engineer,
part-time consultant, part-time salesperson, part-time manager) is also the
most unpredictable one. It's the human mind. And that really does defy the
human mind. Well, it defies mine, at least.

Of course, reactions and comments are welcome.

Christian Lheureux
Responsable du Département Systèmes et Réseaux / Head of Systems and
Networks Department
APPIC R.H.
business partner hp invent
Tel : +33-1-69-80-97-22   /   Fax : +33-1-69-80-97-14 / e-mail :
[log in to unmask]
"Le Groupe APPIC recrute, contactez nous !"

Frank wrote after Wirt lots of stuff :

> -----Message d'origine-----
> De : HP-3000 Systems Discussion [mailto:[log in to unmask]]De la
> part de Frank Gribbin
> Envoyé : mardi 28 août 2001 16:29
> À : [log in to unmask]
> Objet : Re: [HP3000-L] OT: You don't have to wait very long any more
>
>
> Definitely amazing.  Yet HP and MIT's work in nanotech
> threatens to eclipse
> these speeds.  The human mind is doing something different, seemingly
> demonstrating an ability to dynamically "rewire" itself
> creating a plethora
> of parallel "processors" and then wading through the results (again in
> parallel.)  This explains our ability to entertain
> conflicting points of
> view at the same time and make a decision.
>
> Frank Gribbin
>
>
> On Tue, 28 Aug 2001 09:55:49 EDT, Wirt Atmar
> <[log in to unmask]> wrote:
>
> >Who would have thunk it? This is an enormous amount of power in a
> serialized
> >processor. Somebody is going to have to think up some
> applications that
> can
> >really take advantage of these extraordinary speeds. Clock
> timings are now
> >well under half a nanosecond.
> >
> >Perhaps this last sentence is the greatest indicator of how
> differently
> >implemented animal neuronal informational processing is from
> what we're
> doing
> >in computers. PCs are now running almost precisely one million times
> faster
> >than the time it takes for a single neuron to fire, and yet
> we struggle to
> >make a PC understand speech, or even more simply produce high-quality
> speech.
> >We clearly don't understand the natural algorithms yet or
> the depth of
> their
> >complexities.
> >
> >Wirt Atmar
> >
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
>

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *