Subject: Re: hmm
From: "Benjamin J. Tilly " <ben_tilly@operamail.com>
Date: Thu, 27 Feb 2003 02:14:17 +0500

Rich Bodo <rsb@ostel.com> wrote:
> On Wed, 26 Feb 2003, Benjamin J. Tilly  wrote:
> 
> > Any engineer who designed a mechanical system that
> > people's jobs depend on which only continues to work as
> > long as a pencil remains balanced is being silly.
> > Anyone whose job depends on that engineer's design
> > working is being stupid.  Either figure out how to
> > clamp that pencil so that it is stable in the
> > onfiguration that you want, or else go back to the
> > drawing board because it ain't going to work.
> 
> I'm probably getting a little OT, but your that comment has promted a
> runaway train of thought that I would like to share.  It's sometimes
> necessary to include inherent instability in your design criteria.
> Many robots and aircraft, like pencils on their tips, hold themselves
> in unstable positions during operation.  They remain unstable, but
> their inherent instability is handled by fast feedback systems.

Note that the feedback systems convert the overall system
to one which satisfies the description of Le Chatelier's
Principle.  Therefore it appears unstable, but is not.

I will come back to the pencil example shortly...

> When he talks about feedback loops and the paradox of logical
> indeterminacy I interpret Soros as describing a sort of uncertainty
> principle that makes it difficult to view any economic system as
> stable.  What he basically says is this: Exchange rates affect
> fundamentals of economies.  Fundamentals affect exchange rates.  This
> can get into a runaway feedback loop, one way or the other.  If you
> predict a direction to the feedback loop, you influence it, if you're
> a player.  Even if you don't tell anyone, you have realized it's
> existence, and therefore affect your own behaviour, which has some
> bearing on the situation, or the potential of the situation, small or
> large.  When everyone recognizes this, markets become inherently
> unstable.  He therefore argues for strong market authorities.  He
> wants control systems in place.  It's not a totally absurd theory.

I agree, and I have the same major qualm about
macroeconomics.  There are fundamental questions about
what solutions are stable, and my feeling is that
well-meant attempts to induce stability can itself be
the cause of future problems.

However I think that it is a mistake to conclude that
the right solution is a comprehensive control regime.
Our knowledge of dynamic systems says that sometimes a
surprisingly dumb answer works very well, while detailed
control fails miserably.  For instance with the balanced
pencil, an exercise that I had in an advanced DE course
in college was to demonstrate that a pencil balanced on
end with an oscillating bottom was actually stable as
long as the magnitude and frequency of the oscillation
passed a basic limit.  As a practical illustration take
a look at someone riding a unicycle.  They move the base
in a 3 point pattern, giving rise to oscillations
left-right and front-back that make their sitting upright
a stable system.  Anyone who thinks that it is practical
to minutely adjust to every little fall is welcome to
try that theory out and compare results...

How does this apply to macro-economics?  Well my
observation is that humans and human organizations model
risk based on observed experience.  In an environment
where nothing has gone wrong for a long time, risk is
overly discounted.  Therefore I wonder whether the
government should deliberately attempt to oscillate
interest rates - to try to create periodic mild amounts
of risk so that businesses avoid things like the dot
bomb excess.

> Maybe the Soros Uncertainty Principle has some tricky workarounds like
> quantam teleportation does.  There are plenty of spooky phenomenon in
> sociology to research.  Wouldn't it be a blast?  My guess is that we
> won't penetrate entagled minds so readily as we have the entangled
> photons.  It's a wonderful thought to me, though.

Unfortunately the social sciences generally fail of the
most basic requirement for scientific progress.  Which
is the availability of examinable systems where useful
simplifying approximations can be applied.  Without that
you may have interesting speculations, but with little
chance of making demonstrable progress towards useful
conclusions. (Note: Science is not the pursuit of Truth,
it is a way of finding useful approximations to reality.)
The extent to which specific social sciences
(economics comes to mind) manage to overcome this issue
is strongly correlated with their ability to focus on
problems that don't involve the full complexity of human
behaviour.

Note that I am not denigrating the people who try to
study the social sciences.  I would do no better in
their position.  Without somewhere to start and some
easily agreed on measure of progress, it is hard to get
anywhere useful.

Cheers,
Ben

-- 
_______________________________________________
http://www.operamail.com
Now with OperaMail Premium for only US$29.99/yr

Powered by Outblaze