Knowledge, Truth, and Values in Computer Science

Metaphor And Law

We were led to these questions, interestingly, when, in our study of abstraction in computer science, we found ourselves considering the role of metaphor in computer science (Colburn & Shute, 2008). Computer science abounds in physical metaphors, particularly those centering around flow and motion. Talk of flow and motion in computer science is largely metaphorical, since when you look inside of a running computer the only things moving are the cooling fan and disk drives (which are probably on the verge of becoming quaint anachronisms). Still, although bits of information do not ``flow'' in the way that continuous fluids do, it helps immeasurably to ``pretend'' as though they do, because it allows network scientists to formulate precise mathematical conditions on information throughput and to design programs and devices that exploit them. The flow metaphor is pervasive and finds its way into systems programming, as programmers find and plug ``memory leaks'' and fastidiously ``flush'' data buffers. But the flow metaphor is itself a special case of a more general metaphor of ``motion'' that is even more pervasive in computer science. Descriptions of the abstract worlds of computer scientists are replete with references to motion, from program jumps and exits, to exception throws and catches, to memory stores and retrievals, to control loops and branches. This is to be expected, of course, since the subject matter of computer science is interaction patterns.