11/16/2009 19:3

Power

Power is a measure of the work done in a given amount of time. In physics, that's the change in the energy of the system, without regard for what is being accomplished, but in human life we may prefer to consider only useful work. In other words, power in human society is referenced against human goals; if you are going to play a power game in human society you need to know what work you want accomplished.

It occurred to me the other day that in actual reality we spent too little time thinking in terms of power. Power, I suspect, may have gotten a bad connotation, perhaps because the idea has been overplayed by some people. In particular, people with whom many of us would prefer not to have play have played power games disproportionately. Often, those who overemphasize a particular play do not make that kind of play particularly well. So our negative reaction to power may credit them with more than their actions warrant. Even if they make power plays successfully, unless they do so in a way which constitutes a successful play in the actual reality game it is of little import to us.

Why do we want to consider power in making our plays? This follows from the fact that we want to accomplish our goals, to do useful work, which is the very definition of power.

The matter arose in my mind while I was considering the attitude of several computer programmers to a suggestion for changes that should be made to certain computer code. Two objections to making the change were raised. The first was that it would be a lot of work and take up a lot of time (which presumably could be used for other purposes). The second was that we could get by without making the change, that the cost of doing nothing would be small.

The first objection is clearly a question of power. How much work can be done in a given amount of time is a question of how much power you have. I was frustrated by the objection because, as computer people, we have at our fingertips the power of computerization. We could -- and I did -- make the complete set of changes in a very small amount of time by applying the power of automation. My frustration lies in the fact that these computer programmers were not thinking in terms of the power available.

The second objection is also a question of power. To see this, you need to recognize that there is a cost to doing nothing: You save the cost of the change, but this is offset by the cost of the lost opportunity. If we do not make this change, future changes will cost more. We'll still do the same work, eventually, but more of the work will be spread over a longer period of time: less power.

The application of more power is not inevitably a better decision. There can be reasons to use less power, that is to spread the work (and therefore the cost) over a longer time. For example, if the resources are limited, you can choose to apply them to multiple efforts (at lower power) or to a single effort (at higher power). If the system were completely efficient, the total cost would be the same whichever choice we make. But actual reality is not completely efficient and the cost is always higher for work delayed.

In actual reality, if you want to play the game well, you have to be willing to balance present and future costs. And that means you need to think about power.