Memory Addresing

[mjg59]

QWERTY, for those who don't already know, was a layout designed specifically to solve the problem on mechanical typewriters, of people typing so fast that they would jam the keys. Dvorak, for example, is a much better layout and allows faster typing, but you are not likely to come across one of these. Why? Because QWERTY is everywhere. Why? because that's what people know. Why? Because that's what they are taught. Why? because QWERTY is everywhere.

QWERTY was designed to allow typists to type faster without jamming the keyboard. The problem with mechanical typewriters is that the arms swing out and intersect each other. The closer the arms are to each other, the more likely it is that they'll cross each other's paths (and so jam). QWERTY helps with this by making you use arms that are further apart. As a result, you can type faster on QWERTY without jamming it. People who claim that QWERTY reduces jamming by slowing you down are obviously either bad typists or have never used a mechanical typewriter - the first time I used one, I jammed it repeatedly.

Coincidentally, the first time I used a mechanical typewriter was while I was learning Dvorak. Comparing the two, I'm about 10 words per minute faster with Dvorak (85 compared to 75). Almost all of this is because I learned Dvorak properly. My QWERTY style is ad-hoc - I don't use the correct finger positioning. The original study that showed that Dvorak was faster was conducted by Dvorak himself. When people attempted to repeat his results, they found that QWERTY and Dvorak were about the same speed. http://www.independent.org/tii/news/990403Liebowitz.html has more on this.

There's certainly evidence that Dvorak is slightly better, but it's certainly not much better.

[mjg59]

FluffyMcDeath wrote:

In that slowing the typist down so that the keys don't jam is faster than unjamming the keys all the time.

No. If that was true, then it would be impossible to type fast enough on a QWERTY keyboard to jam it. But it's entirely possible to do so.

By your argument, there is a maximum speed that can be typed on a mechanical typewriter without jamming. With an ideal keyboard layout, typists would hit that limit and jam the keyboard, and as a result end up slower overall. QWERTY slows typists down, and so they don't hit that limit.

Now, this is obviously not true. It's possible to type fast enough on QWERTY typewriters to jam them. The only reason QWERTY is useful is that it raises the maximum speed that can be typed without jamming. Seriously. Find an old mechanical typewriter (and we're talking about one from circa 1920 here, more modern ones are less prone to jamming and have a different arrangement between the keys and the hammers) and try typing in QWERTY. You'll notice that normally one hammer will be followed by one a large distance away. Now pretend that the keyboard is layed out in alphabetical order. Hammers will then often be followed by nearby hammers. Dvorak has similar qualities to QWERTY in this respect.

And I jammed the keys on a QWERTY repeatedly. Novice typists will do that. But mechanical devices need time for the hammers to swing in and swing out again. Anything that delays the next keystroke helps avoid those collisions.

And anything that causes you to use hammers that are further apart (like, say, QWERTY) helps avoid those collisions.

As to the Liebowitz article, the conclusions are more based on a desire to believe that the markets are rational and efficient than on whether Dvorak is better than QWERTY. Since they believe that the market knows best, they believe that QWERTY must be the same or better, and they bend the evidence to fit.

No, that's entirely not what the article is about. The article is debunking the claim that the prevelance of QWERTY is a market failure. It's not. Dvorak is slightly better than QWERTY, but not massively so. It appeared many years after QWERTY became popular. It offers only a small advantage. The cost of retraining is moderately high. Things that are expensive but offer only a small advantage don't become popular. If you want to take over the market, you need to be either cheaper or much better.

One rebuttal here

Which again misses the point (the Liebowitz article isn't about proving that Dvorak is worse or the same as QWERTY. It's about proving that it's not much better), and claims that the Liebowitz article is accusing there of being a coverup on the Navy report. It doesn't. Liebowitz says that it was difficult to get hold of the Navy report. This suggests that most people who talked about it had never got a copy themselves. The entire "rebuttal" is based on misinterpretations, and in the end it admits that there is little or no decent statistical evidence that Dvorak offers large improvements over QWERTY.

That's more than 10% faster, and thats not a small amount.

I type Dvorak correctly. I use the correct finger positionings. I keep my hands over the correct parts of the home row. I type QWERTY badly. I use whichever fingers happen to be in about the right place. Comparing the two directly without taking that into account is not a fair comparison. I'm glad I learned Dvorak, because it's a lot easier to learn something properly from scratch than it is to unlearn bad habits.