Hacker News new | past | comments | ask | show | jobs | submit login
The QWERTY myth (reason.com)
25 points by numeromancer on Jan 19, 2010 | hide | past | web | favorite | 31 comments

For the record, here's the response to that article: http://www.dvorak-keyboard.com/dvorak2.html

So are there any reliable studies that show 1) the superiority of one over the other or 2) the superiority of one over the other for programming?

I don't know of any recent studies, haven't really looked, but I've been typing in Dvorak for about two years and this is my take on it.

1. I don't think the superiority of one or the other is really the issue here. The issue is making the switch. Those studies are addressing whether it's worth it to switch.

The bottom line is that Dvorak is more efficient in terms of movement. It's optimized for typing in English (other romance languages are quite similar) and QWERTY just isn't.

2. From personal experience, I think that the mappings of punctuation are superior in Dvorak. I especially like having the dash and underscore easily accessible (next to Return) for programming. I also find punctuation more natural overall in Dvorak.

I've been typing Dvorak over 10 years now and more than anything I appreciate the comfort. Even though I type all day long I've never had a typing-related hand problem.

Last time I checked, there weren't any studies showing the superiority of the Dvorak layout that Dvorak himself wasn't involved with.

What's the story on the "KEYTIME" study alluded to in the response above?

tl;dr: it's a wordy but alright article about how the idea of an inferior product winning because it got a head start is a myth that has no actual evidence. QWERTY vs DVORAK is one example of this myth, but it's not central to the piece, actually.

Moreover, it seems to assume that being lucky is the same as being first. Providing examples of second-comers who were successful (VHS, DOS) is not a sufficient argument that luck doesn't matter in markets.

Interesting read, though.

Most claims of "luck" don't hold up when examined closely.

There are rather few cases of some technology being superior in every conceivable way to its active competitor; usually, each has relative pros and cons. Beta wasn't superior in every way to VHS - at every point while they were still earnestly competing, VHS tapes had longer run times. Early Beta tapes only lasted one hour, not even letting you record an entire movie. (ETA: It was very popular to record on LP and SLP modes in VHS, further reducing quality to get extra recording time. Recording fidelity wasn't a winning factor.)

Almost as bad, Sony waited a long time to license the format, turning away interested manufacturers who then produced VHS machines.

End result? A (more) open format that was both more available and distinctly superior in an key aspect won out. Not exactly a roll of the dice.

At first, looking at the scrollbar I thought you may have been harsh in your judgement "It's not that long" I said. But then I realized how closely packed the paragraphs were, how many sentences each had, and just how small the font was, I thought, "He has a point."

Then when I scrolled to the end for the heck of it and I saw the "1 2 3 4" links, I swore I'd come back and upvote you :)

Incidentally, I'm now glad that I'm not weird for reading the comments first, and the story second.

Or, at least, that I'm not alone.

I always do... and if there are many comments then I'll open the link in another tab and flip between them. Seriously, the comments are why I check HN first and Techmeme/other news second.

I think in another post it turned out that majority of HNers here do the same.

I'm surprised the article doesn't mention (or refute if need be) one of the "tales" i have heard for why QWERTYUIOP is the top row: you can spell TYPEWRITER using these letters, which was a neat demonstration for potential buyers, back when the name for this new technology was up for grabs...

The article seems to assume that social factors like product prevalence should have no bearing on the actual ultimate economic benefits of the dominant player, and that's silly, in my opinion.

We've only seen it more and more often; something may be technically better, but there's more than technical consideration in choosing a product, and if the newer product can't overcome the social inertia associated with the old product, then it doesn't offer a net benefit.

People choose PHP because there is an abundance of PHP programmers and businessmen want to stiff people on pay more than they want a good interpreter running their software.

People choose Windows because the system is familiar and all of the software they know, often niche, rare, or unmaintained, works correctly on it.

And so on.

And then there are instances where the consumers don't really have much say, like Blu-ray v. HD-DVD. They were well about even until the one supergiant corporation (Sony) swooped in with extra incentives for the other supergiant corporations (movie sudios) to get them to choose Blu-ray over HD-DVD, even though HD-DVD is ultimately superior.

So I sympathize with their intention to preserve faith in free markets, but their postulation that markets always make the right choice technically is silly. There are many other factors, including retraining costs, meddling by big players, etc., that can effect the ultimate dominance of a product.

I disagree that this article assumes that product prevalence has "no bearing at all". That would be silly.

Rather, I think it merely demonstrates that product prevalence is not the only factor, and often not even the most important one. It dismisses the popular delusion that early success is all that is required to become an entrenched, inferior monopoly.

When high profile figures with political influence fall prey to these delusions -- even in the face of contrary evidence -- it bears a detailed criticism.

I don't necessarily agree with the idea, but there is an economic theory that essentially states that the market is always right given that it has perfect information. You can agree or disagree with that theory, but I don't think what they're saying is necessarily absurd.

Even if you hold that theory, there's an implicit limit as t->infinity. Certainly if a new keyboard that's objectively better was introduced today, by tomorrow it wouldn't have >50% marketshare.

QWERTY was not designed to slow typists down. It was designed to maximize the typing speed without the types(?) getting stuck. This is not the fastest layout now, but was the fastest for mechanical typewriters.

I'm a bit irked by the conclusion here: "Apparently the typewriter example is of such importance to many writers because it can so easily persuade people that an interventionist technology policy is necessary. How else to explain its continued use in this literature? Since an interventionist technology policy is no more likely to benefit consumers than are the myriad other government interventions in the market, we should not be surprised that good examples are largely fictional."

I don't see how you can infer that because the basis for an argument of government intervention by some writers is flawed, that government intervention is flawed. Regardless of whether or not government intervention is the right idea, it seems strange for a magazine based on reason to make the claim : "A → ~B" is true, But A is false, therefore B is true.

I don't think Reason is saying that this example proves that gov't intervention is flawed. I think they take it as a premise that gov't intervention is generally flawed, and the conclusion here is that we have an example of champions of gov't intervention disingenuously promoting this phenomenon as an example of why gov't intervention is needed.

The problem is that I don’t personally know of any “big-government” proponents who use QWERTY as an example of how regulation is necessary (only that they say it proves that markets can have path-dependency, and inefficiencies, &c., which is true whether this example holds or not).

"Since an interventionist technology policy is no more likely to benefit consumers than are the myriad other government interventions in the market[...]"

This seems to imply that they're taking "government intervention is flawed" as a given for the argument, rather than a conclusion. This isn't surprising, since _Reason_ is pretty libertarian.

If it's on that site, it's always (as far as I can tell) an article arguing for free markets or less government or something like that. Indeed:

"Founded in 1968, Reason advances a free society by developing, applying, and promoting libertarian principles, including individual liberty, free markets, and the rule of law."

So it's not really a serious economics journal (with varying points of view) or anything like that, it exists to promote one particular political view. I think that if we see articles from there, we can reasonably expect to see articles posted here from the other side, too, like, say, Paul Krugman or marxists.org or whatever. My desire is that neither one should appear here and that we stick to startups and hacking and things of that ilk, as those don't tend to produce the "SOMEONE IS WRONG ON THE INTERNET!" reaction quite so much.

For a more academic-economics-oriented article on the same theme by the same authors, see http://www.utdallas.edu/~liebowit/keys1.html , from _Journal of Law and Economics_.

Incidentally, I would be very interested in a pointer to a comparably serious/academic rebuttal of Fable of the Keys. It has been a very long time since Fable of the Keys was published, and it has been a long time since I last tuned into the controversy, and at that time my understanding was that a rebuttal had long been promised but never produced.

It's not an economics journal at all: it's a pop-politics and pop-culture magazine, similar to Scientific American or Popular Mechanics, but on different subjects. You wouldn't say that "Popular Science is not a real astrophysics journal" is a mark against the magazine, I assume.

The authors point out that the QWERTY example is currently the most favored by those arguing for government intervention. By attacking the QWERTY example, the authors are attacking the case for government intervention.

> Stanford University economist Paul Krugman offered the central claim of this literature boldly and with admirable simplicity: "In QWERTY worlds, markets can't be trusted." The reason that he uses "QWERTY worlds," and not DOS worlds, or VHS worlds, is that the DOS and VHS examples are not very compelling. Almost no one uses DOS anymore, and many video recorder purchasers thought VHS was better than Beta (as it was, in terms of recording time, as we have discussed at length elsewhere).

In an argument about government intervention, the burden of proof should always be on those arguing for government intervention because they are arguing that freedom should be restricted. To take a simple example, if you publish something on the Internet and I want the government to force you to take it down, the burden should be on me to show why the government should interfere with you. If I don't make a case that is more convincing than yours, you should be left alone. I should not be allowed to require that you prove why you should be allowed to publish.

This is an important distinction because freedom often depends on this. Many activities that free people engage in would not happen if people had to ask permission first-- either because there is no plausible reason why the particular activity should be allowed, or just because it would not be worth the trouble.

In a free society, government must have a good reason to prohibit. In a non-free society, people must have a good reason to get permission. The authors are assuming a free society, so they content themselves with attacking government's reasons to prohibit.

Is this the case of the metric system vs English system? IMHO, the metric system is superior, but it is too expensive to change in the US. Or driving on the left or driving on the right. In this case both are similar but it is desirable to have a worldwide standard.

It wasn't too expensive to change in the rest of the world, though.

[June 1996]

Edit: this comment score has ranged up to three or four points and back down to zero. Is it not relevant to point out that an article was published 14 years ago?

It depends on what point you think you're making. (And perhaps you should make that point explicitly, so that I don't need to guess it before criticizing it, but if I'm guessing wrong, I apologize in advance.) If you are arguing out that an article from 1996 not timely, your argument is "relevant" but not necessarily correct. Note that the thread http://news.ycombinator.com/item?id=1062031 was on the front page when I first noticed this article. It contained a comment linking to the 1996 article. I don't know why the the 1996 article became a standalone thread instead of just a comment on the 1062031 thread, but it might be that someone who read the 1062031 comment found the 1996 article surprisingly worthwhile.

IMO, the 1996 article (and even more the 1991ish "Fable of the Keys" academic journal article which preceded it) remain timely because both QWERTY-is-superior and QWERTY-demonstrates-market-failure stories are remarkably hardy perennials, presented as established fact without acknowledging the Liebowitz/Margolis counterargument. (E.g., _Knowledge and the Wealth of Nations_ by David Walsh (2006), carries approving blurbs from _The Financial Times_ and _The Economist_ on its cover, and should be easy to find in a library, maybe even a chain bookstore. "Paul David" and QWERTY appear in the index many times, but Liebowitz and Margolis do not.)

My point is simply that the publication date provides some important context for the contents of the essay.

qwerty and dvorak are open standards. beta and vhs are licensable. windows and mac os are retail products. although all three pairs are "competitors" in a sense, i don't think it's reasonable at all to compare their adoption by the market.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact