The QWERTY Keyboard: You're Still Using The First Draft

It's Tuesday, June twenty third. On this day in 1868, Christopher Latham Sholes (SHOLZ) and his colleagues were granted a patent for the first practical typewriter. You've probably never heard of him, but you use his most lasting invention every single day. Look at your keyboard right now, on your laptop, your phone, whatever. The top row of letters reads Q-W-E-R-T-Y. Have you ever wondered why? The letters aren't alphabetical. They're not arranged by frequency. They don't follow any pattern that makes obvious sense. I've been typing on this layout my entire life and I never once questioned it until I started looking into it for today's episode. The reason the letters are where they are has nothing to do with helping you type faster. It might actually be the opposite. And the reason we're all still using it 158 years later, even though the original problem it solved hasn't existed since the 1980s, might be one of the most important concepts in economics that nobody ever teaches you. Sholes was a newspaper editor from Wisconsin who spent the late 1860s developing a writing machine with his associates Carlos Glidden and Samuel Soule (SOLE). Their first prototype had keys arranged alphabetically, which seems like the logical choice. It wasn't. The mechanical typewriter worked by swinging little metal arms called type bars upward to strike an inked ribbon against paper. When you hit a key, its type bar would swing up, print a letter, and fall back. The problem was that if you hit two adjacent keys in rapid succession, the type bars would collide on their way up and jam together. You'd have to stop, manually unstick the bars, and start again. On an alphabetical keyboard, commonly paired letters like S and T or D and E sat right next to each other, which meant jams were constant.

Sholes spent five years rearranging the keys through trial and error, sometimes working with a mathematician named James Densmore (DENZ-more) who studied which letter pairs appeared most frequently in English. The goal was to separate commonly paired letters so their type bars wouldn't collide. By 1873, he had settled on something close to the QWERTY layout we use today, and he sold the design to the Remington Arms Company, which was branching into manufacturing after the Civil War.

Now, the popular story says QWERTY was specifically designed to "slow typists down." That's an oversimplification, and I want to be honest about it. The layout wasn't meant to make you slower. It was meant to prevent jams by distributing common letter pairs across different areas of the keyboard. The net effect was that typists couldn't hammer out certain combinations as rapidly, but the intent was mechanical reliability, not speed reduction. Some researchers have also suggested that telegraph operators, who were among the first heavy typewriter users, influenced the layout because they found alphabetical arrangements confusing for transcribing Morse code.

What's undeniable is that the layout was designed around the physical limitations of 1870s machinery. Those limitations disappeared decades ago. There are no type bars in your laptop. There are no type bars in your phone. The mechanical problem that QWERTY was built to solve has been gone for over forty years, and every person reading this is still using the solution. In the 1930s, an educational psychologist named August Dvorak (deh-VOR-ak) at the University of Washington looked at the QWERTY layout and thought it was insane. He studied the most common letters and letter combinations in English and designed a layout that placed them on the home row, the middle row where your fingers naturally rest. On a Dvorak keyboard, A, O, E, U, I, D, H, T, N, and S are all on the home row. About seventy percent of English typing happens on the home row of a Dvorak keyboard, compared to roughly thirty one percent on QWERTY. Your fingers travel less distance. The most used keys are in the most accessible positions. By every logical measure, it should be better.

The evidence that it IS dramatically better is actually mixed, and I think being honest about that matters. Dvorak himself ran a study for the US Navy during World War II that showed Dvorak-trained typists were significantly faster. But the study was designed and overseen by Dvorak, which is a conflict of interest that later researchers flagged. A 1956 General Services Administration study found no significant advantage. More recent analyses suggest Dvorak is probably five to ten percent faster once you're fully retrained, which is meaningful but not revolutionary.

So why didn't it replace QWERTY? Because by the 1930s, millions of people had already learned to type on QWERTY keyboards. Every typing class, every secretary school, every office in America had trained its workforce on one layout. Switching would have required retraining every typist in the country, replacing every typewriter, and accepting a months-long productivity drop while everyone relearned. The cost of switching was enormous, and the benefit, maybe five to ten percent faster typing, wasn't large enough to justify it. QWERTY wasn't the best layout. It was the first layout that got widely adopted, and that was enough to make it permanent.

And here's the part that really gets me: you can switch right now. For free. Every major operating system, Windows, Mac, iPhone, Android, lets you change your keyboard layout to Dvorak in the settings. It takes thirty seconds. You don't need new hardware. The physical keys on your laptop still say QWERTY but they'll type Dvorak. The switching cost is literally zero dollars. And almost nobody does it. Because the real barrier was never money or technology. It's the fact that you'd have to relearn how to type, and that relearning period, a few weeks of feeling slow and frustrated, is uncomfortable enough that billions of people choose to stay on a suboptimal layout forever rather than push through it. The cost is entirely psychological. And it's still too high. If that sounds familiar from the skills episode we did last week, it should. Same mechanism. Your brain avoids the discomfort of early failure even when the long-term payoff is better. Economists have a name for what happened to the keyboard: path dependency. It means that decisions made early in a system's history constrain all future decisions, even when the original reasons for those decisions no longer apply. The system gets locked in, not because the current arrangement is optimal, but because the cost of changing it is higher than the cost of living with something suboptimal. You're not using QWERTY because it's the best way to type. You're using it because Remington sold a lot of typewriters in the 1880s and everyone learned the layout and nobody could coordinate a switch.

The most absurd proof of this is in your pocket. QWERTY on a smartphone makes even less sense than on a laptop. You're thumb-typing on a flat piece of glass. There are no type bars. There's no home row. There's no mechanical anything. The entire ergonomic logic that justified QWERTY doesn't apply to a touchscreen. And yet every smartphone on Earth uses the same 1870s layout because when Apple launched the iPhone in 2007, they could have used any arrangement they wanted and they chose QWERTY because that's what people already knew. Path dependency followed us from typewriters to desktops to laptops to the phone you're holding right now, and at no point along that chain did anyone stop and ask whether the layout still made sense for the new device.

Once you see path dependency, you start noticing it everywhere. The United States uses Fahrenheit for temperature. Every other country on Earth except Liberia and Myanmar uses Celsius. Fahrenheit isn't more intuitive, it's not more scientific, and it makes international communication harder. But switching would require changing every thermostat, every weather report, every recipe, every medical device, and retraining three hundred million people's intuitive sense of what "72 degrees" feels like. So we don't switch. We just convert when we have to and complain about it never.

Same with imperial measurements. The US is one of three countries that doesn't use the metric system. NASA lost a $125 million Mars orbiter in 1999 because one engineering team used metric units and another used imperial and nobody caught the mismatch. The spacecraft entered the atmosphere at the wrong angle and disintegrated. A hundred and twenty five million dollars and years of work destroyed because America uses a measurement system that was old-fashioned in 1799. But switching would require changing every road sign, every speedometer, every building code, and every American's sense of how far a mile is. Path dependency wins again.

The streaming episode we did yesterday is the same pattern. Cable was a bundle nobody loved. Streaming was supposed to replace it with something better. Now we've rebuilt cable as fifteen separate apps at a higher total cost, and switching back to a single bundle would require consolidating companies that are legally and financially independent. We're locked into a worse version of the thing we replaced, and the cost of fixing it exceeds the cost of complaining about it. Here's the thing that's been sitting with me since I started researching this episode. The keyboard has survived for 158 years because nobody could agree on a better layout. Every alternative required retraining millions of people, and the improvement was never large enough to justify the pain. QWERTY beat Dvorak. QWERTY beat Colemak. QWERTY beat every challenger, not because it was better, but because it was already there.

The thing that might actually kill it isn't a better arrangement of keys. It's not typing at all. Voice-to-text has gotten good enough in the last few years that a growing number of people are starting to dictate instead of type. The accuracy has improved dramatically, the speed is faster than most people can type, and it doesn't require you to learn any layout at all. The keyboard that survived for 158 years because no alternative could overcome the switching cost might get replaced by something that skips the keyboard entirely.

Sholes, for what it's worth, never stopped trying to improve his own design. He filed a patent for a completely different layout called XPMCH in 1889, a year before he died. He knew QWERTY wasn't the best arrangement. He just couldn't convince anyone to switch, because by then, his first version had already won. The inventor couldn't escape his own path dependency. And 158 years later, neither can we. So if this comes up in conversation, here's how to think about it. The QWERTY keyboard was designed in the 1870s to prevent mechanical typewriter keys from jamming by separating commonly paired letters. The mechanical typewriter hasn't existed in decades, but the layout persists because millions of people learned it and the cost of switching to something better has never been worth the disruption. The Dvorak layout, designed in the 1930s, is probably five to ten percent faster, but it never gained traction because QWERTY was already everywhere. Economists call this path dependency: early decisions lock in systems that persist long after the original reasons disappear. The same pattern explains why America uses Fahrenheit, why we measure in miles, and why streaming rebuilt cable in a worse configuration. The inventor of QWERTY himself tried to replace it with a better layout a year before he died and couldn't. 158 years later, you're still using his first draft. The thing that might finally kill it isn't a better keyboard. It's skipping the keyboard entirely.

Stay informed, stay curious, and we'll see you tomorrow.