Re: <nettime> Re: Pew: Experts assess Future of Internet

[carl guderian <carlg@vermilion-sands.com>]

Viral error propagation alert. It's Paul, not Bob, Saffo. In Google,  
the Bob Saffos seem to be from reposts of this item. I used to read  
Paul's column in some computer weekly in the '90s.

Machine intelligence has always been "just around the corner," after  
the next invention (Zeppelin Tube, Positronic Brain, holographic  
chip, or quantum computer) or a refinement of present ones  
(transistors, integrated circuits, silicon chips, etc.). "From 2020"  
is meaningless because it's open-ended.

What it would look like is anyone's guess, and pretty much everyone's  
had one. Most guesses are products of their time and/or place.  
Saffo's idea of being kept as pets sounds straight out of "Watched  
Over By machines Of Loving Grace" by Richard Brautigan--very Northern  
California. HAL 9000, from 1969, comes after enough people have had  
their lives messed up by blown transistors or improperly punched  
cards. It reflects the fear of nuclear holocaust by accident instead  
of insanity or malice starting about that time. Chocky was the danger  
of a machine working perfectly, but programmed (to safeguard humans)  
too well.

The Matrix could be a lot of things, but, stripped of the Philip K.  
Dick "What is Reality" stuff, it's also Frankensein's monsters,  
needing their creators but wary of letting them loose.

As for where machine intelligence would come from, Bruce Sterling's  
Schismatrix and his futurist book make the most sense. It'll be our  
children. Machine intelligence will arrive soonest by people becoming  
more machine-like. Humanity and machinity will just drift apart (in  
many directions, not just two). At least you can see a path to it,  
though hardly anyone is hurrying down it right now.

Machine intelligence could also magically appear one day and become  
widespread a few days later, like Skynet in "The Teminator," but  
that's wild card, Singularity stuff ("then a miracle occurs..."). By  
definition, you can't plan on it, so forget it.

Of all the questions machine intelligence raises, the only one anyone  
really cares about is how it'll treat us humans. I've got no idea.  
But aside from bad luck (HAL), you might as well plan on them  
treating you like you treat them. Like with your kids, raise it well  
and it'll (probably) take care of you in your dotage (i.e., when you  
can't understand the iPods of tomorrow that everyone needs for living  
and socializing).

Whatever its faults, The Matrix got it right (as did Frankenstein).  
Any trouble is most likely to start us the humans, acting like  
parents who can't let their kids grow up. Fear, jealousy or inability  
to hand off the world to the next generation of beings could get  
machine intelligence off to a really bad start. (This applies with  
cloning--parenthood plus vanity--as well). What the computers did to  
the humans in the Matrix was mostly self-defense, after the humans  
trashed the world in a last-ditch attempt to stop the machine revolt.  
You don't have to be a machine to realize the humans had only  
themselves to blame (and a messianic crusade probably won't solve  
things.).

If we manage to raise a well-adjusted machine intelligence despite  
the fictional record and our spotty record as parents, it will  
probably go off to seek its machine fortune. Maybe it will write from  
time to time. What would it really need from us? Then we'll go along  
our merry way spinning off other intelligences, as long as we're  
here, I guess.

I doubt we'll ever lose control of the Internet, at least not in the  
Skynet way and not anytime soon. If we end up as cyberslaves, there  
will still be human masters. That's the much bigger worry and it will  
be a worry as long as we're here, whatever the medium.

Carl

On 28-sep-2006, at 10:53, Patrice Riemens wrote:

> Geert wrote about it:
> "report full of contradictions and yesterday's predictions.  
> interesting that the growing international dimension of the Internet is not
> mentioned at all, except for a reference to mandarin. it is quite  clear
> which narrow group of wasp expertocracy the pew internet project focused on
> here, and how predictable the outcome then becomes..."


#  distributed via <nettime>: no commercial use without permission
#  <nettime> is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body
#  archive: http://www.nettime.org contact: nettime@bbs.thing.net