Why didn't the 'better' keyboard ever replace QWERTY?

Mainly because QWERTY got there first, and the rival's advantages turned out to be debated and relatively small. QWERTY traces back to the Sholes and Glidden typewriter sold from 1874. Sixty years later, in 1936, August Dvorak and his brother-in-law William Dealey patented a rival layout designed to be faster and more ergonomic, with proponents claiming it needs less finger motion, causes fewer errors, and is more comfortable to type on.

The numbers behind that claim are striking. On a QWERTY keyboard, only about 32 percent of typing happens on the home row where your fingers naturally rest, while 52 percent lands on the top row and 16 percent on the bottom. Dvorak flipped that: about 70 percent of letter keystrokes on a Dvorak layout happen right on the home row. On paper, it should be a landslide win.

It wasn't, and the reason is as much about evidence as ergonomics. During World War II, Dvorak himself, while serving in the Navy, ran a study claiming typists could switch to his layout in just 10 days, though he discarded at least two earlier studies whose results are unknown. When the US General Services Administration later commissioned Earle Strong to settle the question, Strong's carefully controlled study found no benefit to Dvorak at all, and he accused Dvorak of running biased experiments. The catch: Strong had a personal grudge against Dvorak and had publicly opposed new keyboard designs before his study began.

So neither side was a clean referee. Later experiments found that many layouts, alphabetical ones included, let trained typists reach very similar speeds, suggesting Dvorak's careful design may have had little effect. Today Dvorak is a standard example in economics textbooks of network effects, where a standard everyone already uses is hard to dislodge, though that use of the story has been criticized. Most keyboards remain QWERTY, even though nearly every major operating system lets you switch to Dvorak.