Two discrete levels represented one after another on a graph are normally considered to be connected by a vertical line which has no time value. But the realities of engineering teach us that any vertical line has a time value, no matter how short. If that graph is charting the sound from a single source, we may observe that even extremely dissimilar sounds are connected in a continuum; that they are not items that are arranged, but rather a continuous process of change. Changes may be periodic, transient or random, and they exist at different time scales from “audiotime” (changes in milleseconds) to “conscious time” (changes in seconds). These changes attribute to a sound's “meaning.” The problem for a composer of computer music is how to generate these changes meaningfully.