Besides size and energy (temperature), another important yardstick exists for measuring the universe, complexity. Complexity is accompanied by organization between components, by structure, orderliness, and low entropy. As we construct more and more intricate machines, we find that certain capabilities are a function of the level of complexity of the machine, more than of the particular components from which it is constructed. One can build a clock, or an elementary calculating machine, out of wood, metal, or plastic. The capabilities of the calculator are limited not so much by the materials themselves as by their organization. Such a simple machine cannot alter its preset operations, or make choices based on the state of the machine at any given moment. From more elaborate hardware one can build a digital computer. This machine now can do everything that the primitive calculating machine can do, and much more. It can accept and emit data, can recall, and can calculate in ways not only not preprogrammed in the hardware, but not even anticipated by the builders. It can make choices or decisions for future actions based on the current state of its information, and can "learn" to make better decisions from the outcome of previous trials.

The very same computer in a functional sense can be built from quite different raw materials. It can use vacuum tubes or transistors. Its physical memory storage may involve solenoid switches, mercury delay lines, cathode-ray tubes, or magnetic tapes, drums, or core.