That's a question that nearly every computational scientist has heard until his ears bleed. It invariably happens at family Christmas parties or other social gatherings. Someone who thinks he knows something about almost everything--Uncle Ezekiel, say--wanders over, stares at you whimsically and says, "I hear that you spend your time running computer simulations."

Encouraged that you might have someone here interested in your work, you enthusiastically start to answer when the other shoe drops: What good is that?

And so begins another session of Defend Your Value. With that thought in mind, it's time that we start to take a look at what we can glean from the hypothetical planets simulation project.

"Alright," you say, "what can we learn by running a simulation of a solar system." Well, in analyzing the output of a classical simulation, such as a study of heavenly bodies interacting with each other, there are two basic types of quantities that can be calculated: static properties and dynamic properties. It may be helpful to outline the basic characteristics of each of these types of properties so that we might better understand what the specific properties to be calculated might mean to us.

Static properties might best be thought of as values that can be expressed as averages. If we take people as an example, a static property might be height. A roomful of people will yield height as the average for everyone in that room. No matter what the people do, whether it be running around, eating, jumping up and down, nothing is going to change that value as long as the basic system remains the same. That is, no catastrophe strikes wherein half the people lose their shins and a corresponding 18 inches of height, people don't leave, and so on. Therefore, the average height of the people in the room is static.

Dynamic properties are those that change over time. In our roomful of people, we might look at the distance between people's eyelids as an example of a dynamic property. For example, if we assume the room is a classroom, then the students will probably come in with some level of interest and be fairly alert. Their eyes are probably wide open, and the distance between their upper and lower lids is probably in a steady state. As the class progresses, some lively topic of discussion may cause excitement, with a concomitant widening of the eyes and the lid spacing. If the class drags on, we would undoubtedly witness drooping lids and narrowing spaces. Because it changes with time, the opening in the eyelids of the students in our mock classroom is a dynamic variable.

As with a roomful of people, a solar system full of particles can be described in terms of both static and dynamic variables.