Qwerty Syndrome

Sub-optimum tools or techniques that become entrenched because no one person or organization can change it, because there is no benefit to changing that outweighs the costs. In other words, a majority would have to agree to change at the same time and swallow the temporary conversion costs that will have to be weathered before net benefits of change manifest themselves.

(This describes a NashEquilibrium)


On certain brands of typewriter often-used keys being closer together would cause the hammers in to hit each other causing damage to the machine. The rumor is that the QWERTY keyboard layout was originally designed to prevent keys from sticking in this way by putting common key combinations further apart, slowing one down for those key combos. However, in actuality to keep the mechanical linkage between hammers and keys relatively simple the keys had to be rearranged to make rearranging the hammers possible, with the side effect that it slowed down typing thus reducing the danger of hammers hitting each other even further.

I once saw a documentary about a speed-typing contest. This was when IBM typewriters had a ball with all the raised letters on it that would hit the ribbon and make the letter on the paper. All the top competitors had switched to a ball that effectively converted the ordinary QWERTY typewriter to Dvorak (touch typists never look at the glyphs on the keys anyway). When interviewed the competitors said that switching to Dvorak increased their typing speed. In my opinion this was better RaceTheDamnedCar evidence than any historical study of the motivations of early QWERTY and Dvorak makers. Perhaps there should be more discussion about the idiom? -- BrianFennell

Later models did not need such adjustments, but it had already become an entrenched standard. Of course as people learned to speed type at the new layout that slowdown became relative only. However, newer layouts optimized for typer speed have never caught on beyond a handful of fans. Almost all speed records are set on non-QWERTY keyboards.

Further, other brands took different routes that didn't need QWERTY-like arrangements. The one brand having QWERTY became the de-facto standard because it won in a speed contest. However, by some accounts that brand won simply because it was in use longer and there were more experienced typists available for the contest.


I don't think that definition is quite right. The layout actually makes the keys less likely to jam for those combinations even when typed at the same speed. It applies to almost all manual typewriters, rather than to one certain brand. There are key levers (I'm not sure what the correct term is, but I'm referring to those metal arms that swing forward and strike the page), and these have the potential to collide and jam. Two key levers that are far apart are much less likely to jam than two that are close together, since, although they are both aiming for the same spot on the page, the possible "collision area" is much smaller. If you can find a manual typewriter lying around, it's pretty easy to verify this by experimenting with how much easier it is to jam Q&W than Q&P (and so on), even when you type them at the same speed. [What about ED? Adjacent key levers for a common suffix! But you type them with the same finger, and on the old mechanical keyboards you couldn't do the finger-rolling trick, so you had to release one before you hit the other. QWERTY uses a combination of separation and slowing techniques.]

Perhaps this discussion should be moved to KeyboardLayouts? or something, since this is really about a concept named after keyboard issues. Anyhow, you might be right for keys typed by a different hand, but for keys typed on the same hand, it is generally faster to type letters that are closer together, at least for me.


Nope, this theory isn't true: "largely a myth perpetuated by none other than Prof. Dvorak, who they say held a patent on his keyboard and stood to gain if it triumphed over QWERTY" http://www.xent.com/FoRK-archive/feb98/0437.html

Also, more careful studies about the supposed advantage of the Dvorak keyboard usually show at most a 5% speedup, given equal proficiency, which makes the myth at best irrelevant to the point that most people have in mind when they cite it. There is some controversy surrounding this story. See TheFableOfTheKeys

Odd how it is hard to get objective studies of keyboard speeds (the jury is still fighting with one another). Anyhow, this topic is about the concept, not necessarily the reality of QWERTY's drawbacks.


This differs from ZombieTechnologies in that there is usually no dispute that something is inferior, but that it requires too much up-front investment and coordination to change.


Personally, I think the performance increase from switching from QWERTY to DVORAK is too small to make learning a new keyboard scheme worth the time. Also, QWERTY is so popular now that you have to be 'bidexterous' enough to switch when using a public computer - not easy.

Wimp. I say that as someone who *often* switches keyboard layouts. I'd use Dvorak all the time, but hotkeys in various apps are designed for QWERTY and it's easier to get used to different keyboards for different apps than to remember to hit Ctrl+Shift all the time.

Mac OS Classic and MacOsx have a Dvorak-Qwerty layout. The keys are normally in Dvorak, but the command key switches them back to Qwerty. In effect, the keyboard shortcuts are the same. --BrianSchack?


It seems C-style syntax is subject to QwertySyndrome. People are so familiar with it that anything that replaces it has to make them noticably more productive in order to switch. I don't think anybody would keep most of it if starting from scratch, but the alternatives so far don't knock anybody's socks off either. (Related: ItsTimeToDumpCeeSyntax)


I don't agree it is necessarily a myth.

The myth that Qwerty is superior to Dvorak hinges on some individual studies that were themselves debunked. Basic denialism.

Dvorak puts the most-used keys on the home row: AOEUIDHTNS, so it is trivially superior. Querty prevents old-fashioned keys from jamming.
I'm Sympathetic to this point of view, but I do wonder whether the 'u' is one of the most-used keys? Maybe I can come up with a more often used key, in this one case. I'm not trying to debunk Dvorak. I might mention, in this connection, that the left hand home row in Dvorak is the vowels a e i o u. I suppose that might have a certain neatness to it--easy to remember. I certainly don't give any supposed 'studies' purporting to show that Dvorak gives only a minimal speedup any bandwidth. Not just, because I think Dvorak might indeed be significantly faster, but also because I think that may not be the right approach at all to settling this debate. I might put more weight on which is easier to learn? Seems quite likely to me, that Dvorak is easier to learn--seems quite likely to me, that this could be a very significant difference.

I took typing in school, and it was hard to learn. The logic of the Dvorak's layout, has an immediate appeal to me. I write as somebody who has decided to learn Dvorak. The strange fetish for 'debunking the myth' of Dvorak, this scientism trumping the supposed scientism of Dvorak, it's an interesting debate to me, because there are many debates about pseudo-science and generally, I rather like to side with sceptics of stuff like Rorschach &etc., let alone homeopathy. But Dvorak has obvious logical appeal. It might be, that as long as people who learn Dvorak are self-selected, they get a placebo effect, an ego boost, they defend their decision to learn it, perhaps? But, only those people really have an opinion, based on knowing both Dvorak and qwerty. And I don't know of any cases of somebody learning Dvorak, and not becoming notably infatuated with it, which does count for something with me. Hard to know what to measure, here. I don't think it's an easy question. Perhaps not impossible, but not to be left to you people, as it were.

As to the key issue, sub-optimum tools or techniques that become entrenched, well, it's obviously an issue, and I think dubbing it 'qwerty syndrome' is pellucid. We all make many decisions every day, not to fix what isn't broken, not to switch phone companies, etc., it's the other end of the continuum from faddish early-adopter syndrome. Thinking that you've found the silver bullet. But we all know how almost immediately, when a codebase grows from its very start, day 1, there are suboptimal things left in place--sometimes you go 'dammt I'll go back and fix that, nah, I don't think I will just now', and then it never gets fixed. This is worse on a team. My experience tells me to suspect people's motives. Those who are just there for the money, and are truly sociopaths, once you get to know them, simply con artists, are on the same continuum as me. Do I make an issue out of this? No, I won't stick my neck out, don't want to get fired today..we'll get our hair mussed, but not many dead, and they're not my little ones. This is human nature.

{It's an established myth, probably perpetuated by Dvorak himself, that QWERTY was intended to prevent old-fashioned keys from jamming. Not hitting two keys at once prevents old-fashioned keys from jamming, regardless of key arrangement. Sholes (inventor of the modern typewriter) went to considerable effort to speed up typing, not slow it down. The modern arrangement of keys -- largely established by Remington -- was intended to aide non-typist salesmen in typing the trademarked name of the device, TYPE WRITER, without giving away the fact that the top row of keys was configured for marketing purposes. At the time the typewriter was commercially introduced, alphabetical order was generally considered less important than it is now. It certainly wasn't considered particularly helpful in finding keys, as it was recognised then (as now) that finding keys would quickly become more a matter of "muscular memory" (like playing piano by rote, etc.) than visual searching.}

Even if true, it doesn't significantly change the key issue (no pun intended) of this topic: the cost and benefit curves of changing to something better. Whether the non-optimal layout was driven by mechanical forces or marketing forces is less important than the issue of how or if to move forward to remedy it. It's more about why something is not getting fixed rather than why it's bad in the first place.

By the way, if I start a keyboard or typing training company, I'll name it ASDF. Association of System Data Fundamentals?

The British Railway Gage Standard was set by Julius Caesar. I don't think Julius Caesar really knew what the best standard would be for rail cars. This is another example.

There is some luck-based selection. There were competing standards floating around and Ceasar's just happened to be the best compromise at that time.


Our 10-based numbering system is perhaps a case. Base-12 would be much friendlier if you ignore the "finger" issue. It divides nicely by 3rds and 4ths, common slicing needs. And matches months. Damned early tetrapods!


perhaps related to BandWagonSyndrome

Wagons are easy to get off of; standards are not.


http://www.scribd.com/doc/61087053/24/Chapter24-Simulating-Typos-with-Perl Sean M. Burke's article modeling typing errors; he determined that a Sholes keyboard results in less chance of mistyping and producing something else that happens to be a word that a Dvorak keyboard.

This. All the common letters in Dvorak are right next to each other, and it's very common to have adjacent letters on adjacent fingers. Sholes has more of a right-left-right-left flow, like walking.

EditText of this page (last edited August 9, 2013) or FindPage with title or text search