QWERTY vs. Dvorak

Readers of this blog fiercely debated the validity of the QWERTY keyboard story a few months back. As the legend goes, Christopher Sholes engineered the QWERTY layout that is still in use today in order to slow typists down and prevent key jams. One commenter (ludvig) pointed to this 1996 article from Reason magazine by Stan Liebowitz and Stephen E. Margolis that “put the torch to the QWERTY myth.” In the article, Liebowitz and Margolis argue that the persistence of the QWERTY myth undermines the theory of efficient markets, since it is usually used as an example of luck winning out over innovation:

The typewriter keyboard is central to this literature [against market efficiency] because it appears to be the single best example where luck caused an inferior product to defeat a demonstrably superior product. It is an often repeated story that is generally believed to be true. Interestingly, the typewriter story, though charming, is also false.

But then another commenter (saharvetes) linked to this furious letter to the editor that appeared in the next issue of Reason. Randy Cassingham, who wrote the letter, also wrote a book called The Dvorak Keyboard in 1986. Dvorak is a layout invented after QWERTY by Dr. August Dvorak that minimizes finger movement by prioritizing letters that are used more often (like vowels), allowing for greater speed and less chance of carpal tunnel. Cassingham claims that Liebowitz and Margolis reported biased research that “proved” QWERTY’s superior efficiency in order to disprove the myth, mainly relying on the research of a man who hated Dr. Dvorak and destroyed his own data so that his findings couldn’t be verified.

If you feel like going down the rabbit hole of this old feud, check it out. And if you decide that Dvorak *is* the board to beat, here’s a wikiHow that will show you how to switch your keyboard and operating system. According to the Wiki entry, you’ll need a month of training with the new Dvorak layout to get up to your old typing speed and the transition period may be initially painful, because you’ll be using new muscles. And don’t even think about using a QWERTY keyboard during your Dvorak training. You’ll only undo all the progress you’ve made.

My question: Is there anyone who cares enough about a few extra words-per-minute who would be willing to give up the ability to type for a whole month? Learning the Rubik’s cube or baking a kitty litter cake might be time better spent.


Dvorak is counterproductive in quite a few cases. As you note, the layout is optimized for the letter frequencies. But those frequencies change depending on the language, so you need a separate - largely different - layout for every language. A layout optimized for one language would be less optimal for another language than Qwerty.

And that means that if you use more than one language (I use three regularly), you need to learn one separate layout for each one. And, of course, remember to switch layout every time you change the language you happen to type in at the moment.

And it means that would Dvorak actually become widespread, you'd come to public internet access points or computers in hotels, airports and wherever, and be faced with a layout that has nothing in common with the one your're used to, rather than the same basic layout with a few frills altered you have now.


I just don't see this having widespread adoption unless schools start teaching the Dvorak keyboard instead of the QWERTY (assuming most people first learn proper typing in school). I didn't find it on first glance of the links, but does it show anywhere the difference in typing speed for someone who once used QWERTY and has now mastered Dvorak?


The reported benefits of Dvorak over QWERTY are not limited to a few words per minute. Dvorak proponents claim their layout reduces repetitive stress disorder type injuries including sore hands and the like. For many that my make the transition worth it.

warcraft maps guy

I'm proficient with both Qwerty and Dvorak.

Two main non-obvious advantages of Qwerty are that 1) the rest of the world uses Qwerty so using different computer doesn't cause any confusion 2) many keyboard shortcuts and typical commands were designed with Qwerty in mind. Typing 'ls' or typical emacs keyboard shortcuts can be a pain in Dvorak.

However, Dvorak is indeed very comfortable and fast for typing large amounts of English text.


A few comments actually --
First, I've heard that the productivity benefits of changing to Dvorak would save companies "millions" although that of course is although theoretical money.

Another story I've heard in regards to luck creating an "inferior product" is with gasoline engines on cars. I don't claim to be an expert by any stretch, but someone once told me that cars originally ran on steam engines. In a race in Europe, every car in the pack was a steam engine car, except for the race's winner -- a gasoline powered car -- which apparently won in a fluke. As a result, every automotive engineer pursued gasoline engines, which has of course led to our oil dependence and poor environmentalism.


@2 - I switched over a couple years back. I typed around 70 words per minute using QWERTY, and I've been able to hit around 90 using Dvorak. The majority of the speed increase is in longer words, because having to bounce between hands for vowels and consonants gets more efficient the more you do it, I think.

It took me about a month to get going, and the first week or so actually WAS painful. You end up using your hands a lot more than your wrists, and you go side to side more than up and down.

I actually got into it because I was curious how muscle memory was related to language, and whether or not typing was a certain type of language, too. I figured a lot of basic speech is training your mouth to make certain sounds in succession to form words. What I noticed when learning Dvorak was your fingers go through the same type of training. The first day or two, I was hunting around the keyboard for the right letters, but once I started getting routine groups of letters ("ing," ".com," "the," "wh," etc.), I really sped up. The remaining three weeks or so was just getting the intricacies down.

I've switched back and forth in the past few years. It takes me a couple days to get my "fluency" back; longer with QWERTY than Dvorak, funnily enough.



I took the Dvorak plunge about a year ago. As others have said, the first month was painful, but now I never want to go back. My primary motive for switching was not speed (though that is a nice benefit) but reduced risk of repetitive stress injury (i.e. my wrists were bothering me). So I simultaneously switched to a contoured keyboard and the Dvorak layout. (more details on my blog)


I switched many years ago because it reduces the amount of finger travel. I can still type in Qwerty. Switching back and forth is not seamless, but if I need to use a Qwerty keyboard I rapidly become proficient. One trick: I use Qwerty shortcuts because that keeps the Qwerty locations fresh in my head.

Why? As noted above, to reduce pain from too much typing. Also to see how long it would take, what it would feel like to learn a new version of an old skill and how well it would work versus Qwerty. Answers were: took a fairly long time to get very fast, felt very weird like my mind was being bent, and it works very, very well for both speed and reduced strain.

The mind bending feeling was worth the effort. It's rare one gets an opportunity to subject one's self to an experiment like this. Comparables are probably more hard-wired: changing hands for throwing or writing, changing legs for kicking a soccer ball.

I was probably drawn to trying Dvorak because I was born left-handed and was forced by an accident to switch. I can still write leftie and do many things leftie and I know that my brain and my body respond differently when things are done by one side or the other. Some "things" are physical: for example, my eyes are set up as a left-hander with a dominant right eye, so I was a natural left-handed batter. Some "things" are hard to quantify, as in the way my writing content changes when I use my left hand.



One thing - the benefit of Dvorak over Qwerty is when _touch_typing_. Something most people never learn. And if you hunt-and-peck, or type at a decent speed with your own, non-standard non-formal finger stance you just won't see much of an improvement in either speed or comfort, and you'll still have all the drawbacks of a nonstandard keyboard.

If you want to increase productivity, don't change to Dvorak. Just learn to touch type at all; that's where the big savings could come from.


Very interesting discussion.
JanneM: I think most people who spend a lot of time on keyboards are touch typists.

As a country we never did fully adopt the metric system.
The educational system would fully adopt the Dvorak keyboard.


Luck winning out over innovation? I know another keyboard example, the various ergonomic keyboards. A standard keyboard is not designed with the human in mind but countless millions of identical devices were introduced. As soon as a contoured keyboard came out, I was sure people would flock to it, but no. Sure, it's a bit odd at first, but it's so much easier on your wrist and hand. Why not change to the superior form? Why not indeed!

Josh Millard

Futurist speculation: I don't think we'll see any mainstream adoption of a non-Qwerty layout until keyboard devices start coming with some sort of dynamic keycap displays standard.

I think the ubiquitousness of Qwerty is insurmountable under the current circumstances, essentially: the life cycle of the keyboard loop is a long one, and conditions have to be right to make it worth it to the status quo to endure a whole life cycle to make a mainstream change. That's a lot to ask of the general computing public, and there's no way to accomplish this with fixed-layout devices without thrusting an unfamiliar layout in the face of the computing public.

(I know that keyboard configurations can be changed with relative ease, but the physical keyboards cannot be: for most folks, the folks who don't know DVORAK from Adam and aren't going to learn a new layout recreationally, the keycaps are the layout. It's not even Qwerty, in the layout-discussion sense; it's just How Keyboard Are. So, for the moment, a Qwerty-capped keyboard is a Qwerty keyboard, period.)

Qwerty gets taught because Qwerty is how the keyboards in schools are shipped, how home computers come set up, how what electric typewriters still get sold are configured. It's what the computer at work, and at the library, and at the internet cafe uses. Using DVORAK, or any other layout, in the wild? Hard. Hard things like that aren't going to just happen without some tremendous incentive. People won't want to learn it, manufacturers won't want to sell it, workplaces won't want to install it.

So how do you get that new layout out there?



Learning to touch-type is easily the most effective thing I've ever done to boost productivity. Easily.

I've never understood why employers don't just insist on staff learning it. It's simple - if a little like hard work - and the training is straightforward. Looking through the CEO's goldfish bowl, it often looks as though we could be saving £20k per year, there and then with a little basic computer literacy training.

However, why don't more people use those machines that stenographers use? They seem like the fastest option possible.

Or, why did the Microwriter die out - 80's gizmo as endorsed by Douglas Adams (http://en.wikipedia.org/wiki/Microwriter)?


I have used the Dvorak before, and I am certain that my brain has plenty of room for both Dvorak and Qwerty.

I never did switch over for good because I couldn't find a Dvorak kb for a reasonable price. I tried moving all of the keys on a spare qwerty, but the f and the j were keyed so they had to go in their appointed spots. After using my hacked keyboard for a few days, I was able to type on it at a reasonable rate, but not at my qwerty speed. I could switch back without much effort.

It is really fun to sneak into some unsuspecting person's computer and switch the keyboard map. They get very confused, and the helpdesk techs think they are crazy.

The benefits are too marginal for the average person to take the effort. That might be different if I could try out 20 dollar dvorak keyboard.

Keyboards are pretty inefficient anyway. I suspect we will find something better in the next few years.



It does not take a month. It took me 2 or 3 days to be able to type; maybe a week to type at a normal speed. Maybe it helped that I used a program to teach me.


It seems to me that once you've mastered touch-typing a la Dvorak, it shouldn't matter how the keys themselves are arranged.

It's easy to tell the computer it has a Dvorak keyboard even when it doesn't -- just never look down and you're golden.

Interesting that some WWII coding was done using a similar technique: touch typing on scrambled keyboards. Once you've internalized the pattern, muscle memory takes over and the keys could be blank for all you care.

And just so it's never forgotten -- blowhard columnist John Dvorak had nothing to do with this invention. (-:


isn't this gonna be obsolete when speech recognition software becomes efficient?


If you ever try a French keyboard then just a few letters out of place, not to say umpteen punctuation keys, make typing a pain.

Even US/UK differences are irritating.

Even a random distribution standardised would be better than deviation. For that matter, significant deviation is probably better than slight, because you then HAVE to look at the keys. Thus, I'd rather use US Qwerty and Dvorak, than 55 subtly different Qwerty keyboards.


In the book "Mastering the Dynamics of Innovation" by James Utterbach of MIT, the Dvorak/QWERTY issue was addressed in what I believe to be a proper perspective: in the early days every typewriter manufacturer had its own keyboard layout for the same reason that all plumbing manufacturers had their unique pipe diameters and thread pitches: to make it hard for customers to change suppliers. The QWERTY keyboard happened to be on the first typewriter that displayed the print immediately after typing... an innovation that became an overwhelming standard and dragged the QWERTY keyboard with it. The QWERTY is like 60 cycles 115volts AC...neither tecnically good nor bad (compared to alternatives) but the overwhelming standard upon which other standards are hung. About 20 years ago I wrote an essay on this "Standards, Innovation and Survival" which can be found on my web page at www.elew.com


I've switched from QWERTY to Dvorak several times. I found Dvorak wasn't necessarily faster, but was definitely more comfortable since most words could be formed with significantly less finger movement. The only reason I ever switched back was during times when I was working on many different computers that were not mine and switching back to QWERTY on the fly was too irritating.

Eventually I resigned myself to having to know both, so I used Dvorak at work (on a TypeMatrix keyboard) and QWERTY at home (on an IBM M15). Now I naturally fall into the appropriate layout depending on the feel of the keyboard. If I type on another TypeMatrix, Dvorak comes naturally but QWERTY is difficult, on the M15, the opposite is true. Using keyboards with significantly different tactile responses made it much easier to keep the wires from crossing in my head.