Vox gets QWERTY wrong

This video produced by Vox.com on why we have QWERTY-standard keyboards was interesting but it didn’t actually answer the question as to why QWERTY was over-rated. It’s claim is that it was the result of collusion from typewriter manufacturers and how typing was taught. Sure, that explains how it started, but it doesn’t explain why it persists when we don’t have typewriters. After all, there is no cartel now. There are no typewriting schools. There aren’t even any typewriters!

They don’t cite it but the video tracks fairly closely the history of QWERTY as told in this 1985 paper in the American Economic Review by Paul David. (Actually, it is one of the most famous economic history papers of all time). It is a short paper but the video actually stops just three pages in. They didn’t read on to the section “basic QWERTY-Nomics.”

If they had they would have been introduced to network effects (or as David called it ‘system-wide economies of scale.’ Why do we use QWERTY now? Because every keyboard we encounter as QWERTY. A few years ago, my son realised that he could change the keyboard on his computer. He read about the Dvorak design and so decided to switch. He figured it would take him about a month to get better and he measured his performance to make sure. This, of course, was David’s point — it actually wasn’t too hard to change keyboard designs. There was learning but the rate of return to switching was positive.

Much to my amusement (as I had read David), my son eventually ran into a problem. He was going to computer camp (of course) and realised that he would be using a different computer every day. He thought about bringing a Dvorak keyboard with him but then realised that uploading drivers each time to Windows machines would be a pain. So he gave up.

This illustrates why we have QWERTY. It is not that once-off switching costs are that high. It is that on-going switching costs are. So we have coordinated on QWERTY.

My point is that the Vox video misses this completely! I cannot figure out why. Moreover, in the process, it actually answers nothing let alone saying something new.

For instance, we have QWERTY keyboards on our smart phones. What we don’t know is if that has the same rate of return issues as we have on full sized keyboards. One theory I’d like to put forward is that QWERTY is now back to being close to optimal. Your phone keyboard isn’t a true keyboard because the keys are too small. Instead, it is a predictive text instrument that anticipates the key you have tried to hit by taking into account what you did previously. So if you type ‘t’ it figures there is a high chance you want to type ‘h’ next even if you finger hits ‘g’ or ‘j.’ This was an innovation from Apple in the original iPhone.

However, the whole point of QWERTY was to ensure that it was very unlikely that you would type two keys close to each other. The Dvorak keyboard is designed to have you type faster and doesn’t have that characteristic. Indeed, the vowels are altogether in that design which would diminish the ability for predictive typing to work. Consequently, it may well be that QWERTY is now optimal again and the puzzle both David and Vox talk about was a short-lived decades long affair.

4 Replies to “Vox gets QWERTY wrong”

  1. See Liebowitz and Margolis, The Fable of the Keys, Journal of Law and Economics, 1990. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1069950

    This paper examines the history of the QWERTY typewriter keyboard, often put forward as the archetypical case of markets choosing the wrong standard. Contrary to the claims made by Paul David and Brian Arthur, we find virtually no evidence to support a view that QWERTY is inferior to DVORAK. Instead, using records of typing experiments, studies by ergonomicists, and examining the historical record of competition among different keyboard designs back when QWERTY first became dominant, we conclude that QWERTY is about as good a design as any alternative.



    1. thanks for writing this comment — you saved me the trouble. The “QWERTY is inefficient path dependent market failure . . . ” myth lives on — impervious to any serious examination of the historical record. A convenient “just so” story for the market failure crowd..

  2. Does that mean that the explanation I got years ago about QWERTY being formulated during the early mechanical days of typewriters to slow typists down so they didn’t overwhelm the capabilities of the machine is apocryphal? I’m crushed!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s