The keyboard that refused to die

QWERTY is 150 years old, designed for a machine that jammed. You are typing on a relic from 1873. Why?

On June 23, 1868, a Wisconsin newspaper editor named C. Latham Sholes walked into a patent office in Washington D.C. and filed a document that would, eventually, shape how nearly every human being on the planet interacts with machines. The machine in question was a typewriter. Sholes and his collaborators had been tinkering with it for years, rearranging metal bars and watching them smash into each other, and by 1868 they had something that kind of worked. Not well. They knew it.

What they couldn't have known is that the layout they landed on, after years of trial and error, would still be sitting under your fingers in 2026. On a phone, a tablet, a laptop, a desktop. On glass, on plastic, on mechanical switches that click and clack exactly the way typebars never did.

They couldn't have known it would outlive the Remington company that manufactured it, outlive mechanical typewriters themselves, outlive the electric typewriter, the CRT terminal, the first personal computer, and probably the last one too. That it would become an object of study in economics departments. That people would write PhD theses about why it won't go away.

The jamming problem

Sholes' first typewriters had the letters arranged in alphabetical order, row by row. This made sense if you thought of a typewriter as a printing press for one person to use. But the typebars, those metal arms that swung up to hit the ribbon and the paper, didn't like alphabetical order. When you typed fast, adjacent typebars would collide mid-swing and lock together. You'd have to stop, reach in, and pry them apart with your fingers.

So Sholes started rearranging. Not by any grand theory, just by trying things. He'd put common letter pairs on opposite sides of the machine, so their typebars wouldn't swing from the same direction. He worked with his brother-in-law and with a man named Amos Densmore, who apparently studied letter pair frequencies by hand, by candlelight, to figure out which combinations were most common. Not bad for a guy in the 1860s with a stack of newspapers and a lot of patience.

By 1873, Sholes had sold the design to Remington, the gun and sewing machine company (which is a whole other kind of strange). Remington put the Sholes & Glidden Type Writer on the market in 1874. It typed only in capital letters. The letters were hidden under the carriage where you couldn't see what you were typing. It cost $125, which is about $3,200 today. And it didn't sell well.

The TYPEWRITER myth

There's a story you've probably heard, that the top row of QWERTY was arranged so salesmen could type "TYPEWRITER" quickly using only that row. It's a beautiful story. It's also unsupported by any known evidence. A pair of Japanese researchers named Yasuoka and Yasuoka published a paper in 2011 that traced the myth back through decades of repetition and found no primary source for it anywhere. The top row of a Sholes & Glidden didn't even spell TYPEWRITER — the original layout had the period where R is now, and some letter positions were different.

The story persists because it's neat. It's the kind of story you want to be true. But the truth is more mundane: Sholes and Densmore just kept shuffling letters until the machine stopped jamming, and this is where they landed. The "TYPEWRITER" thing is an urban legend that somehow became the received history.

The better keyboard that wasn't

In 1893, a man named George Blickensderfer produced a typewriter with what he called the "Ideal" keyboard. Its home row was DHIATENSOR — ten of the most common English letters, right under your resting fingers. It was, by any metric, a better layout. It didn't matter. Remington had already won. The first-mover advantage was so complete that even a clearly superior alternative had no chance. Blickensderfer's company died. QWERTY marched on.

Forty-three years later, in 1936, August Dvorak, a psychology professor at the University of Washington, patented the Dvorak Simplified Keyboard. His layout put five vowels on the left home row and the most common consonants on the right home row, like a mirror image of the Ideal. It was supposed to revolutionize typing. Years of tests followed, including a study by the U.S. Navy during World War II that claimed dramatic efficiency gains.

In the 1990s, economists Liebowitz and Margolis published a famous paper arguing that the whole "Dvorak is superior" narrative was an urban legend propped up by shoddy science and wishful thinking. The debate still simmers in certain corners of the internet, mostly among mechanical keyboard enthusiasts who have strong opinions about everything.

Path dependence

So why are we still using QWERTY? The economic answer is path dependence, a concept that the economist Paul David wrote about in 1985 using QWERTY as his primary example. Once enough people learned QWERTY, the cost of switching became astronomical. Not just the retraining, but the coordination problem: for Dvorak to make sense, everyone would have to switch at once, or documents would need to be keyboard-agnostic, or every keyboard would need to be reprogrammable. None of those things were true in the 20th century.

The network effects worked against the better product. Every QWERTY typist made Dvorak slightly less viable. Every new typist who chose QWERTY (which was all of them, because that's what was taught in schools and available in offices) reinforced the standard. This is the same dynamic that keeps the QWERTY layout on every phone screen in 2026, where there are no typebars to jam and no mechanical reason whatsoever to maintain the arrangement. It's purely legacy. The ghost of a machine from 1873.

The boring truth

Here's the thing. QWERTY isn't that bad. It's suboptimal, but typing speed is mostly determined by the typist, not the layout. A good QWERTY typist can hit 80-100 words per minute without breaking a sweat. The bottleneck is cognitive, not mechanical. Moving to Dvorak might buy you a marginal gain, but not the kind of gain that justifies months of frustration and productivity loss.

There's a lesson here that goes beyond keyboards. The QWERTY story is a parable about how technologies become standards. It's not about merit. It's about timing, market power, and the weight of everyone who came before you. The best technology doesn't always win. Sometimes the adequate one just arrived first, sat down at the table, and never left.

I'm writing this on a laptop keyboard that has a "QWERTYUIOP" row, a "ASDFGHJKL" row, and a "ZXCVBNM" row. On a laptop built in 2023. The metal arms that caused all the trouble are gone. The ribbon is gone. The paper is gone. The carriage is gone. The bell that rang at the end of each line is gone.

What's left is the layout. The ghost arrangement that made sense once, in a specific mechanical context, with a specific set of constraints. And it's still here because a hundred million people learned it before anyone could convince them to learn something else.

That's not a design failure. It's a social fact. And social facts are harder to change than any typewriter.