Logo IMG


Qwerks of History

Brian Hayes


It's no secret that the success of a technology is not always explained by merit alone; there are also qwerks of history and qwerks of fate—named for the most famous example, the qwerty keyboard on which I am typing these words. The qwerty layout is not the best possible arrangement of the letters—it may even be the worst—but once everyone has learned it, the cost of change is too great. A slightly different kind of qwerk is sometimes cited to explain the ascendancy of the gasoline-fueled automobile engine. Whether or not this powerplant was the best choice at the outset, the argument goes, so much engineering effort was invested in its development that rivals couldn't keep up. By now, the global infrastructure supporting gasoline engines gives them an almost insuperable advantage. The story of color television offers yet another variation on the theme: The broadcast format adopted in the 1950s was chosen not because it gave the highest picture quality but because it was compatible with existing black-and-white TV sets. Even 50 years later, replacing that suboptimal format with the digital HDTV standard has been slow going.

These stories imply a certain mental model of technological evolution. When a new product category appears, there's an initial free-for-all period, in which diverse ideas compete on a more-or-less equal basis. But multiple alternatives cannot co-exist indefinitely. Eventually one technology becomes dominant—for whatever reason—and all others are driven out. A latecomer to the market has little hope of breaking in.

The trouble with this neat formulation is that counterexamples are as easy to find as examples. If it were true that an established technology could never be dislodged from its niche, we would still be listening to vinyl LP records. If competing technologies could never co-exist, we would not have three incompatible standards for cellular telephones in the U.S., with a fourth standard prevalent elsewhere. It appears that some technologies remain fluid over long periods, whereas others freeze solid early in their history. How do we tell them apart? What is it about audio recording that has allowed such a profusion of media (wax cylinders, 78s, 45s, 33s, eight-track cartridges, cassette tapes, CDs, MP3s), while keyboards have remained steadfastly faithful to qwerty? Perhaps close examination of specific cases and circumstances would explain these discrepancies, but I don't see much hope for a simple, predictive theory of technological evolution.

The realm of computing offers a great many further illustrations of qwerkiness. Here I shall discuss two examples of strangely long-lived technologies, both of them distinguished members of the class of 1969: the Intel family of microprocessors and the Unix operating system.

comments powered by Disqus


Of Possible Interest

Computing Science: Belles lettres Meets Big Data

Technologue: Quantum Randomness

Technologue: The Quest for Randomness

Subscribe to American Scientist