the persistence of QWERTY
Why is QWERTY so persistent? Common wisdom is that QWERTY was deliberately designed to slow typists down. According to MIT, it wasn’t.
Rather, the layout was intended to place common two-letter combinations on opposite sides of the keyboard. On manual typewriters, each key is mechanically connected to a lever that has the reversed image of a letter on it. If a typist were to hit two keys on the same side of the keyboard in rapid succession, the second lever on its way up would hit the first on its way down, the keys would become stuck together, and the typist would have to stop typing and unstick the keys. The qwerty layout was a clever design that minimized this problem.
However solving the original mechanical problem seems to have left us with an inefficient design for modern keyboards. The sheer persistence of QWERTY has been held up by many as a classic example of market failure. But is this true? There has certainly been passionate debate among academics on the economics of QWERTY. Apparently it’s quite a controversial subject, and is often cited along with standard-gauge railways in discussions on path dependence. There is a firm view among some academics that QWERTY has persisted simply because it has successfully beaten off its rivals for the last 130-odd years.
Alternatives have been around for decades. Take Dvorak for example. It was invented in 1936 and was intended to alleviate the apparent deficiencies of QWERTY. Barbara Blackburn, the world’s fastest typist (according to the Guinness Book of Records), clocked 212 wpm on a Dvorak keyboard.
But is Dvorak inherently superior? The claims were based on a 1944 US Navy study, which seemed to show significant improvements in typing speed with Dvorak. According to (among others) the Economist and MIT, unfortunately, the original studies showing Dvorak to be better were flawed. Not to mention being conducted by Dvorak himself. In fact, a carefully controlled study in the 1950s “unreservedly concluded that retraining typists on Dvorak was inferior to retraining on QWERTY.”
Sadly, there seems to be little advantage in switching to Dvorak, especially taking into account the switching effort. Estimates of the time to go from QWERTY to comparable skill with Dvorak range from 40 hours to 100 hours to six weeks. Compare this to the flawed Navy study cited above, which claimed that the entire retraining cost is recaptured 10 days after the start of retraining. Again according to MIT, most people don’t want to switch because they’re afraid of how long it will take them to learn the new keyboard, and of having to continually work with two different layouts because of the sheer pervasiveness of QWERTY.
But hey, it’s available as an option in Mountain Lion, and a jailbreak adds it to iOS6. And I have three weeks of leave over Christmas. Maybe I will give it a try. Or maybe I will learn Elvish instead. I have been meaning to…