Methods of Prediction: Systems theory can be more frightening than illuminating. If the simple act of linking one event to another can result (more . . .)

Using Intuition: Decision Theory uses the mathematics of probability to guide decisions. But setting the odds and making sense of the numbers (more . . .)

Fear of Emergence from Complex Systems: The most fascinating aspect of the behavior of units once connected is that they behave in ways not predicted by (more . . .)

Essays: This is my how-to site. The essays are notes about the experience of becoming familiar with the tools of my (more . . .)

AppleScript for Weblogs: Once upon a time there was Manila Express. It was the tool that drove my early weblogging experience. Once installed, (more . . .)

Uncertainty and Decision Theory: To start, I recommend two books which were my start, Making Hard Decisions by Clemens and Thinking and Deciding by (more . . .)

About this site

Essays

Current Working Notes

My personal website is at ODBlog . I write about decision theory at my site On Deciding . . . Better . This site is a hyperlinked set of essays using Tinderbox.

Syndication available

Most simply, a system is a collection of parts or pieces that interact, affecting each other over time. Each piece obeys its own rules of behavior. The interaction is set by its own set of rules. It would seem that understanding each piece and each interaction adds up to an understanding of the system.

In Decision Theory, understanding a system means predicting its future behavior. In order to choose the best among alternatives, I have to know the result of each choice. Knowing the result is prediction. Some systems, at least in certain states are predictable.

My watch is an Omega Speedmaster Professional that my parents gave me as a Bar Mitzvah present in 1970. Every morning I wind it and it acts predictably to show me the time. Over the years, its predictably grown to lose about a minute a day. I pretty predictably adjust the time every two or three days. As a collection of parts that interact, my watch is an example of a complex predictable system.

On the other hand, sometimes I find that the watch has stopped. I guess I forget to wind it everyday, because it will run for more than 24 hours when fully wound. I have no idea exactly how long it will run because I am unpredictable in my winding and the watch is unpredictable in how long it takes to run down. Of course, I can define the behavior a little more precisely, since I know it has a 100% probability of running for 24 hours and 0% probability of running for a week without being wound. My watch and I, as a system, are more complex than simple clockwork.

Still given the laws of physics and knowledge of the parts of my watch, one could predict that it would tell time. One could predict that it would require winding and probably calculate about how long it would run without winding. Of course, just like my empirical experience, the estimate of runtime would be approximate. Knowing the parts and their rules of interaction, I can estimate some chances of events occurring in the future, which is how Decision Theory defines probability.

As a system, I exhibit a bigger prediction and control problem. My body, my brain, are made of individual cells, connected and interacting in billions of combinations. Understanding the nature of my neurons and their patterns of connectivity would never lead to a prediction of how frequently I would wind a watch. The concept of a watch would be unlikely to emerge from a study of neurons and synaptic connections. Somehow my emotions, motivations, thoughts and actions emerge from the gooey substance of my brain in ways that can't be predicted from a knowledge of the system's parts and its connections.

In Systems Theory, this is well-trod ground. Simple mathematics leads to completely unpredictable behavior. Calculating successive digits of pi is not hard, but is entirely chaotic, unpredictable. Similarly, complexity, behavior that is orderly but can only be estimated within limits as a probability occurs with relatively simple mathematics. And the emergence of completely orderly behavior that can't be predicted from the rules can be seen nicely in the computer program "Life". Stephen Wolframs' A New Science is a nice synthesis of this, but has the major fault of removing it from its historical background, making it seem as if Wolfram conceived of these ideas independently.

Copyright 2003 by James J. Vornov