Re: AI over the Internet (was Re: making microsingularities)

From: James Rogers (jamesr@best.com)
Date: Mon May 28 2001 - 13:12:34 MDT


At 10:36 AM 5/28/2001 -0700, John Marlow wrote:
>All of this seems to assume that we're talking about today's landline
>internet--which could indeed be faster. And also today's computers--
>which could indeed be better.

As I stated in the part of my post you quoted, this is only a limitation
today and to a lesser extent tomorrow. There is no reason to believe this
will not disappear as a problem in a relatively short period of
time. However, while the time may be vanishingly small on a historical
scale, it will seem like a long time to people living right now waiting for
this capability to grow.

>But what of tomorrow--when it is, and they are? And who says you
>can't like a bunch of basement clusters together, even today?

I think the question was whether or not the Internet could be exploited in
its natural state, not whether a human engineered solution would
work. Obviously a system explicitly engineered for the problem will do
just fine as these things go.

-James Rogers
  jamesr@best.com



This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:07:49 MST