Sunday, September 7, 2014

What Part Of Distributed Tuning Of Some But Not All Complex Systems Do People Have A Mental Block About?




From CEO 'Takers' To CEO 'Makers': The Great Transformation

This article suggests that capitalism should transition it's key management metric from shareholder value to customer satisfaction. ..... Uhhh, be still my beating heart?

In an email, Casey Haskins states the obvious response by pointing out that for such complex tasks, "no SINGLE measure will work."

You'd think that would be obvious for everyone, but it's not! What is it about tuning of some but not all complex systems do people have a mental block about?

What happened to the concept of distributed solution sets for complex tasks? Don't stop doing nested polynomial N(i), wherever appropriate? That's what we do all the time, but just not consistently.

Ignorant citizens all over the country are quite comfortable using multiple parameters to tune engines, or card games or to "solve" video games. Then they turn around and can't yet apply the same logic to tempering their ideology, politics and policies.

Isn't diversity that what makes democracy so resilient? Cultural recombination is as useful as sexual recombination. What's missing? Just practice? Just practice at managing enough outcomes to know how many variables have to be juggled?

Sure, all people get experience handling a wide spectrum of processes which they presume depend upon one to many control or feedback parameters.

It seems overwhelmingly clear, however, that many of the presumptions people make about MANY of the processes they utilize are, in fact, grossly erroneous. The bulk of humans in a crowd - or mob - are remarkably cavalier about monitoring the variables which co-effect their personal-+-group outcomes.

The concepts of central or distributed CONTROL of experimental variables is ostensibly taught as a fundamental axiom for use of the scientific method ..... but I can tell you by experience that remarkably few supposed "scientists," ever actually learn that axiom in undergrad courses, graduate schools or other training programs. Even fewer citizens, scientists or not, ever get enough practical experience at managing massively parallel "combinatorial" experiments, of the sort faced daily by individuals, electorates, economies, nations and cultures. In combinatorial evolution, there is no control, only accelerating outcomes to surf. And only the agile aggregates survive.
It's ironic that attempted utilization of the scientific method itself has become largely formulized, as more of a ritual than an honest act of exploratory logic.

If formulization of the scientific method itself has become ritualized, why is anyone surprised that faux disciplines like our various policy ideologies are steadily drifting further from common sense or relevant reality?

The big question is what to do about this recognized problem.

Where are the key places to intervene?
What are the key methods for gracefully intervening?
What type of key people are able to use these methods to gracefully intervene at key places?
How do we get the right key people with key skills into key places in key institutions?

For get the Dem and GOP parties. Maybe we need a MACS party - for Multivariate Adaptive Common Sense? Jane and Joe Sixpack might sign up, if the party platform presented a combination of engine tuning, card games and computer games as their political platform. :(