At 7:43 -1000 10/4/98, Jay Hanson wrote:
>-----Original Message-----
>>>From a long time lurker:
>>
>>I was so impressed by the Roberto Verzola article that I forwarded it to a
>>number of friends for comment. Then I started thinking about the world.
>>
>>I then realized that Verzola's concept of society as a well-constructed
>>computer program was completely inadequate, simply because the "world" is a
>>complex adaptive system, a "chaotic" system, with countless global,
>>semi-global and local variables. It is virtually the entire biosphere. A
>>better analogue is a living organism. Try to imagine a computer model of
>>the human body. Think of all the global, semi-global and local variables
>>you would need. Body temperature is a global variable ~ global temperature.
>>Seritonin is very local (in my understanding) ~ telephone call (from one
>>neuron to another).
>
>It really is complicated, but so is Windows98.
The world is complicated in a different way from Windows98. The latter
isn't a complex adaptive system (CAS) but merely a complicated machine with
some sloppy code that makes it crash every now and then. Conceptually, it
merely speeds up Babbage's analytical engine by several orders of
magnitude. Win98 is intended to accept a defined range of inputs and
produce precisely predictable outputs. Modularity helps the programming
team achieve that objective.
>Obviously, the best we can
>do is approximate reality. But were we have uncertainty, we would adopt the
>"precautionary principle" -- it's the opposite of our present system.
An understatement. With a CAS the size of the biosphere, we simply don't
know how far we can push the system from its current state before it begins
to move towards a different attractor. Historically, a powerful attractor
is a range of temperatures we call an ice age. CASs also tend to change
states very quickly. I've enjoyed trolling through your web pages, Jay, and
learned a lot from them, but I have a deep suspicion of smooth curves on
time-related graphs. It's entirely possible, even likely, that at some
point the systems become stretched to the extent that critical negative
feedback loops turn into positive feedback loops and almost complete
discontinuity results. In other words, things could turn out a lot worse
than the models predict, and sooner. Those curves could be a lot steeper
and go a lot deeper.
But my original point was regarding Verzola's argument in favor of
modularization and the elimination of global variables. It seems to me that
such a project would have about the same probability of success as
attempting to modularize the biosphere or the human body. Enzymes that act
throughout the human body often, and perhaps always, are paired with an
antagonist enzyme that regulates the action of the first one. It's
impossible to eliminate many global variables such as the temperature of
the Earth. In order to keep the system reasonably stable, we will have to
find something to arrest its rise, and the answer must be global.
Similarly, global capital flows are not self-limiting, and I seriously
doubt that modularization in the form of nation-based limitation on capital
flows will be effective -- not when international trade is the lifeblood of
so many developing nations. I'm not so wise as to know the answer on this
one. Perhaps a Tobin tax. But the world economy clearly exhibits the
characteristics of a CAS, and we will be making gigantic blunders if we
treat it like a linear machine.
Regards,
Tom
_____________________________________________________________________
Tom Lowe Judge a moth
Jackson, Mississippi by the beauty of its candle
-Rumi
