Slashdot Mirror
Why is Everyone Still Stuck in QWERTY?
theWrkncacnter asks: "I was recently giving some instructions over IRC to a long time QWERTY keyboard user who wanted to switch to the Dvorak layout, mostly because a good majority of the people in channel had made the switch and were all talking it up, myself included, about how our speeds had increased and how its much more comfortable. This made me think, why don't more people use the Dvorak layout? Searching around I found an older topic on the subject, but that didn't answer too many questions, as most people in the comment section seemed to think that Dvorak vs. QWERTY was a hardware issue, when it is really a matter simply changing the layout on your particular OS. I took the time to pry off and remap my powerbook keyboard's keys but I have no problem typing in Dvorak on a physically QWERTY mapped keyboard, and I know many others who don't have a problem doing so either. So given all of this, why don't more people switch? Is it that most people just can't be bothered to make the change, even when its more efficient and more comfortable?" Is it mostly due to the fact that most people learn to type first on QWERTY due to its popularity, and hence don't bother to learn anything else?
First, everyone learns on QWERTY. Why? See reason two.
You are more likely to find a QWERTY attached to any particular PC or terminal than anything else. Switching back and forth is a pain.
Thirdly, unlike you, it seems, not everyone is a touch typist.
I tried to switch... but the letters written in Sharpie ink rubbed off to quickly =(
~foooo
I hunt and peck faster than I need. I can hit the keys with my index finger without looking and I'm not going to learn to touch type ever, although I type for a living.
Changing keyboard layouts would reduce my typing speed with no benefit. The fact is that most people can't think faster than they can type, and only a fractiion of the population need to type very fast and would benefit from a change to Dvorak.
Reliable, Great Value Hosting: $7.95/mo 2.4G/120G
1. See the post above me (Everyone learns QWERTY)
2. Users don't like having to learn new input methods (partly the reason why soft (ie software) keyboards on PDAs are in the QWERTY layout, despite the fact that the skills related to tapping the keyboard with a stylus are completely different to those found in touch typing.)
"I won't mod you down - I feel the need to call you a twit explicitly, rather than by implication."
Slashdot had an article a long time ago about going beyond the efficiency of Dvorak and determining what is better through genetic algorithms. You can read the Slashdot article here.
I use QWERTY because it's the standard. I know it's not as efficient as DVORAK, but it's more than fast enough for my needs (and I spend all day writing code and emails), so why go through the hassle of relearning typing skills and using DVORAK? Especially in an office environment where I have to keep constantly swapping over to my co-workers keyboards - I really don't want to have to deal with swapping contexts all day long.
Then there's the fact that most apps come with keyboard layouts configured for QWERTY keyboards.
Dull answers to your question, but were you expecting anything else? People aren't going to inconvenience themselves unless the benefits FAR outweigh the problems. I'm sure it's the same reason why many people don't use Linux.
QWERTY keyboard from Taiwan is so cheap it's nearly free. I wouldn't know where to start looking if I wanted an alternative keyboard layout.
Also, how many readers are concerned with WPM ? The quality of my code tends to take a sharp nosedive when I type quickly. Lots of thinking, slow typing, a good editor with syntax highlighting that notices when I don't have enough close braces, etc. Why don't more people use smart editors ?
Now wash your hands.
Oh god. This is like a bad Twilight Zone episode.
...to the Twilight Zone.
<voice style="serling">
Limekiller. Reader of Slashdot and sometimes typist. He thinks he's seen every rant devised by man. He also believes that he has come to grips with the Slashdot editor's penchant for beating the proverbial dead horse. It is with this jaded approach that he will begin his evening routine with a bookmark. A bookmark that leads
</voice>
Dear sweet Christ would you let the qwerty debate die! Hath you no shame!?
My
Limekiller
"So given all of this, why don't more people switch? Is it that most people just can't be bothered to make the change, even when its more efficient and more comfortable?" Is it mostly due to the fact that most people learn to type first on QWERTY due to its popularity, and hence don't bother to learn anything else?"
It's because nobody cares. It creates more problems than it solves. Do you really want to retrain your fingers just so you can type a little faster? Is your keyboard really your bottleneck? (Linux masochists excluded from that question.) Do you really want to move your keyboard shortcuts around? Do you really want to use a non-standard keyboard? What do you tell friends that come over and use your computer?
There may be benefits to it, but we're not excatly talking about a live issue here. I mean if we're going to discuss this, why don't we discuss why people should use Procomm instead of Telemate for visiting BBS's.
"Derp de derp."
A) Actual research does not support the efficiency gains of the Dvorak layout. The most-commonly-cited study in favor of the Dvorak layout was published by ... guess who ... Mr. Dvorak himself, and the science behind that study is deeply questionable.
The data entry industry did their own studies, which do not support the claimed efficiency boost of the Dvorak keyboard. Since they make more money if their data entry personnel type faster, they had every reason to conduct a fair and honest study of the two formats. They stuck with QWERTY.
B) QWERTY is actually pretty damned good. The common urban legend about QWERTY being designed to slow typists down is just that, an urban legend. It is true that QWERTY was designed to reduce jamming on mechanical typewriters, but it did not do this by intentionally slowing typists down, as the legend claims.
Instead, it does this by ensuring that commonly-pressed pairs of keys are not next to one another (and in the days of mechanical hammers, this would also mean that the hammers were not next to one another). Conveniently, this means that successive keystrokes are likely to be pressed by alternate hands, which actually makes typing faster instead of slower.
C) Your own anecdotal stories are, I'm sorry to say, worthless.
This is for two reasons: first, you probably didn't do a formal study of your typing speed before and after the test, and you also didn't have a control group of people who remained with the QWERTY layout but put an equal amount of effort into attempting to improve their speed.
Second, even if it is true that you really do type faster with Dvorak, that's not conclusive. Some people can do math faster with an abacus than they can with a calculator, but that doesn't conclusively prove that the abacus is a better tool. It just proves that there are some people for whom the abacus is a better tool. Unless you do a large-scale test and find both the positive cases (you) as well as the negatives (people who tried the Dvorak layout and don't like it), you really have no clue which is better.
Again, these sorts of studies have been done. Every one I am familiar with concluded that the benefits of the Dvorak layout were minimal at best.
ZFS: because love is never having to say fsck
I can't believe this submission beat my "Why did some people prefer Gobots to Transformers?" story.
"Derp de derp."
Game, set, match... QWERTY.
Comment forecast: Bits of genius surrounded by a sea of mediocrity.
I used Dvorak for a while.. but there were a couple things that made me switch back to Qwerty:
/rant
1. public computers: internet cafes, computer labs, libraries, or even helping my girlfriend out with her computer required me to un-wire and switch back to qwerty for a while.
2. Vi: Vi was made for the qwerty layout, with the home row movement keys (hjkl). Remapping the keyboard is possible, but not without breaking all of the memnomics (sp?) that I had previously had. i.e., that row becomes "dhtn", 3 of which have other (non-movement) meanings (d = delete, t = to, n = next). What now becomes my delete/to/next keys? And what are the memnomics?
3. I was never taught to type correctly. My hands are not on the home row, my fingers are extended, and my form is a mess.. I basicly use like 3 fingers on each hand to type, moving my hands a lot. I get decent speed doing this (~60 wpm, I would guess), but it isn't accurate and doesn't translate to dvorak. When I learned dvorak, I realized it was designed for touch typists with the standard home row configuration. To this day, whenever I use dvorak, I change to the home-row stance. I am not as comfortable or as confident in this position and it makes my typing slow.
So, I found myself constantly switching back and forth between qwerty and dvorak.. my bad typing habits were created for qwerty.. and after months on dvorak, I still found qwety to be faster. That is why i reverted to qwerty. I wish I was better at dvorak, i really do, but damnit, i want Vim to work the way it should.
-molo
Using your sig line to advertise for friends is lame.
I learned Dvorak a few years ago. At the time I learned it on a physically QWERTY keyboard, which helped enforce proper finger positioning. I ended up being about 15WPM faster in Dvorak (85WPM overall), which certainly wasn't bad - however, I was also typing "correctly", while my QWERTY was an ad-hoc mess that I'd learned as I went along. Spending less time just forcing myself to learn QWERTY properly would probably have resulted in much the same speed increase.
Nowadays my desktop machines have IBM keyboards with removable keycaps, so they're all physically Dvorak - on the other hand, my laptop is both physically and logically QWERTY because other people want to use it occasionally. Switching takes a few seconds, but isn't a major problem.
How hard is it to jump between QWERTY and Dvorak layouts? I've thought many times about switching to Dvorak, and I'm pretty certain that I could be back up to speed in only a few months, but there's no way that I can be restricted to using *only* the Dvorak layout, so the ability to remain moderately productive on a QWERTY keyboard is a prerequisite.
So, how hard is it to jump back and forth? Is it like having two separate modes, each equally capable, or do the two sets of muscle memories stomp on each other? I've known people who spend enough time on telephone and adding machine keypads to develop excellent "touch-typing" skills on both, and they could bounce between them flawlessly, never missing a single stroke even at high speed, in spite of the different layouts. OTOH, there are fewer keys and more "environmental" clues to distinguish between them.
I notice that (spoken) languages often seem to create the same sort of "modality", whereby a person fluent in two languages can trivially jump between the two with little risk of accidentally mixing them. OTOH, I find that I have a strong tendency to mix keywords and syntax across multiple programming languages, particularly if I'm not using different development environments (my theory is that the different IDEs provide some context that helps).
So, how does it work?
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
...but I have to use public computers a lot at school (commute, can't afford a laptop). Switching back and forth was horrible. I went from being a decently fast QWERTY typist to a crappy QWERTY typist on public computers and an almost decent Dvorak typist at home. It wasn't worth the hastle of switching back and forth on a regular basis, I'd never be able to make myself more efficient without being able to focus on one or the other. Since QWERTY is more common, I decided to just use it at this point.
I will admit that I miss having the semicolon where the Z is on a QWERTY keyboard.
I'm starting to think this isn't the best place to promote my Anti-Sig Campaign.
The best reference I have found on the subject is The Fable of the Keys.
This paper basically attempts to prove that QWERTY vs. Dvorak was not an example of market failure -- in other words, that the best keyboard really did win and it wasn't because QWERTY was an entrenched standard that nobody was brave enough to challenge (which is the typical argument that the losers in any such fight give).
ZFS: because love is never having to say fsck
I almost believed you, until I looked for myself for the studies you refer to. While your comments on the subjectivity of his experiences are correct, your statements regarding the results from several major studies is biased. Do a google search for Dvorak vs qwerty and you can read a good subset of those results again, for yourself.
My primary reference on this subject is The Fable of the Keys, which seems to be a pretty comprehensive look at the entire debate. I have read other papers regarding the subject, and generally found the same facts.
f you agree that moving your hands a smaller distance is more efficient, than it follows that the Dvorak layout is more efficient.
The only statistic worth debating is typing speed. Not hand movement or anything else. You can debate numbers all you want, but unless you've done a study showing that Dvorak is faster, you're just engaging in mental masturbation.
This happens in damned near everything -- film vs. digital, MP3 vs. CD, CD vs. vinyl -- people make assertions about what is better without actually bothering to do a fair comparison. I'm tired of it. Point me to studies which show that Dvorak is better than QWERTY, or be quiet.
ZFS: because love is never having to say fsck
The last time I wanted to switch to Dvorak, many years ago in college when one does things like that, I went all out. Switched keyboard layouts, actually physically swapped the keys on the keyboard, etc.
The problem is that some programs used command-keys that were based on keyboard position, and some were based on actual letter (so command-o on the dvorak layout might be either command-o, because they were using the letter, or command-s, because that is the key in the same space on the qwerty layout). So I never knew from program to program which keyboard shortcut I'd be using.
It might not be as much of an issue now, with a more modern OS. On the other hand, now I really don't care.
=Brian
---
There is nothing so good that someone, somewhere, will not hate it.
Quay is typable with only the left hand on a Dvorak keyboard. So is pope, pupae, pike, and (probably) others. Note that the calculator on that page is not very good - it is unable to say 100% for "same hand". Try typing the word "i" for an example. I'm not saying whether QWERTY is better, just that your facts are wrong.
Trees can't go dancing
So do them a big favor
Pretend dancing stinks!
My passwords wouldn't be as easy as "asdf" anymore.
It's jsut like switching to the metric system...sure it's easier...but everyone is just too set in their ways and too lazy to switch.
One more time, based on my own research,
Dvorak - about 70% of all keys you will ever press are on the home row.
Qwerty - About 35% of all keys you will press are on the home row.
Conclusion,
You won't be moving as much. It is so much more comfortable for me to use Dvorak. The only way I can describe it, it feels like your fingers are flowing over the keys. It looks strange to watch someone with a Qwerty keyboard type because you see how much more movement and funny hand positions that they have to make.
It's not THAT much faster, but it is noticable.
Two disadvantage that I have to admit that I've found from my experience are:
1. On cold nights, your hands tend to stiffen more *because* of the lack of movement.
2. It's harder to type with one hand. Since the keys are placed so that the hands alternate for most keystrokes. Oh no!
Actually, that's not quite right. It looks like most of the arguments you're presenting were based on The Fable of the Keys by Liebowitz and Margolis.
... guess who ... Mr. Dvorak himself, and the science behind that study is deeply questionable.
As I pointed out when the topic came up last year, Leibowitz and Margolis are economists, and while their discussion of market externalities was correct, they don't quite represent the cognitive research on the Dvorak vs. Sholes (QWERTY) issue very accurately, or fairly.
A) Actual research does not support the efficiency gains of the Dvorak layout. The most-commonly-cited study in favor of the Dvorak layout was published by
The data entry industry did their own studies, which do not support the claimed efficiency boost of the Dvorak keyboard. Since they make more money if their data entry personnel type faster, they had every reason to conduct a fair and honest study of the two formats. They stuck with QWERTY.
Actually, the half-dozen or so well constructed lab tests comparing Dvorak to Sholes consistently show a 5-10% advantage for Dvorak (even Leibowitz and Margolis admit that Dvorak is somewhat faster). For a good overview of the research conducted on text entry, check out Jim Lewis's chapter "Keys and Keyboards" in the Handbook of Human-Computer Interaction.
The "data entry industry" study you're referring to is probably the Navy study mentioned in The Fable (and Lewis's chapter). Leibowitz and Margolis don't really describe the study correctly. This is in part due to the strange way it was conducted -- I'm away from my copy of it so I can't give a good description.
(On an unrelated note, it is pretty irritating to read Leibowitz and Margolis's character assassination of Dvorak. I once asked a well known economist about Stan Leibowitz and was told that his research seems to be too motivated by his political beliefs. I have no idea if that's true, and I would never use that as an argument to refute him in a peer reviewed article. Likewise, I think that the aspersions cast on Dvorak's reputation are a bit disingenuous and out of line for a scientific article.)
B) QWERTY is actually pretty damned good. The common urban legend about QWERTY being designed to slow typists down is just that, an urban legend. It is true that QWERTY was designed to reduce jamming on mechanical typewriters, but it did not do this by intentionally slowing typists down, as the legend claims.
Instead, it does this by ensuring that commonly-pressed pairs of keys are not next to one another (and in the days of mechanical hammers, this would also mean that the hammers were not next to one another). Conveniently, this means that successive keystrokes are likely to be pressed by alternate hands, which actually makes typing faster instead of slower.
Sort of right. Analyses of cross-hand keying do indicate that QWERTY is pretty good, but Dvorak is still better.
C) Your own anecdotal stories are, I'm sorry to say, worthless.
Actually, this is sort of true, sort of false, but these days probably irrelevant.
True: Only well designed scientific studies (or simulations) of human performance using various layouts can tell us which layouts are most efficient in which contexts.
False: Your anecdotal evidence is actually worth a lot -- to you. If you typed at 40 WPM using one layout and now type at 60 WPM using another layout, good for you. It doesn't mean anything for anyone else, but something about the switch (the new layout, the practice you had to engage in, your desire to prove that your layout is superior) helped you.
Irrelevant: Unless you are a transcriptionist (in which case, you probably should be using a specialized tra
Back in university, my roomates and I had a go at leaning dvorak. In our case, the most compelling reason was reduced risk of RSI, not typing speed. As an added bonus, it was also an opportunity for me to learn proper touch typing (which I never did with Qwerty... still haven't).
We never pulled it off.
We got all our X11 keyboards remapped. We changed the keycaps on my PS/2. We downloaded some tutor apps from the net. We even spent quite a bit of time actually practicing. Results were promising... But then reality kicked in.
At university labs, I was routinely using about 5 different keyboards a day, some X, some tty. Remapping all of them wasn't an option, so I was trying to learn Dvorak while still blasting out assignments in Qwerty. Then there's situations during the transition from Qwerty to Dvorak where there's no feedback... Trying to enter a password on a keyboard with Dvorak keycaps but a Qwerty layout is, uh, hard.
What killed the whole thing, however, is that I'm a vi user. vi at the best of times can be a disaster for bad typists. Just trying to navigate via ijkl in vi on Dvorak is futile, much less handling complex ingrained key patterns like df' or 'ay}. After years of vi use, I've got these patterns burned into my fingers. Learning a new keyboard without learning a new editor at the same time won't happen.
Yet another problem is that too much emphasis is placed on the letters. C/C++ programmers need a good symbol layout too and we make at least as much use of the symbols as the letters. Dvorak is, I found, a bit weaker in the symbol layout than Qwerty. {}[];()= are, I think, the most commonly used C symbols... This choice of symbols and the convenient placement on Qwerty is probably not accidental.
c.
Log in or piss off.
5% is quit a marginal increase, not minimal.
Consider as such:
Person A types 8 hours a day.
In these 8 hours, (s)he averages about 70wpm (while typing). 70wpm * 4 chars (4 chars/wd + 1 space) = 350 keystrokes/min.
In those 8 hours, assuming 1 hr break (30 min lunch, 15 min * 2 breaks, legal minimum, YMMV) this means 7hrs*60min/hr*350cpm = 147000 chars (/5cpw) = 29,400 words typed in a day.
This is, of course, theoretical.
In any case, a 5% increas in speed now puts you at 30,870 wpd. This is over 1000 words difference.
I can think a lot faster than 70wpm. the faster I can type, the faster I can get my code into the box before its forgetten.
If I get my code in faster, i have one of two choices:
a) more time for testing before production roll-out (assuming enough time)
b) actually making the deadline (if it is a really restrictive deadline)
Of course, you think 5% is minimal, so why bother?
Uhhh, error rates, finger fatigue, increased or decreased risk of RSI, retraining costs, application changes, documentation changes... these are ALL statistics worth debating.
!! You just found the next big revolution. They say that porn drives all technological improvements and pushes the boudnaries..
...
Keyboard layouts designed for one-handed use! Brilliant! You can write code with your [less 'favorite'] hand, while that pr0n video clip in the lower corner of the screen has your other hand
Well, you get the idea.
What an annoying assumption. John C. Dvorak didn't come up with this keyboard layout. It was devised by Drs. August Dvorak and William Dealey in the 1930's. You can read more here
C) Your own anecdotal stories are, I'm sorry to say, worthless.
Actually, no, they are not:
Before: I typed 80 WPM, I was about to quit because my hands hurt, and not anecdotally.
After: I still type 80 WPM, my hands no longer hurt.
I've been tyiping Dvorak for 6 years now. I'm not going back.