The existing keyboard was designed experimentally by Christopher Sholes, the inventor of the typewriter, to slow the typist down, because the keys in his 1873 machine fell back into place by gravity.
It is often frustrating to see all of the supposed emphasis that the U.S. and Canadian economies are putting on productivity. Over the past year, Business Week magazine has had many articles and even one entire issue devoted to the problems of productivity. I have written them several letters, pointing out where the Dvorak Simplified Keyboard could be of real benefit. None were published. Instead, I received a letter stating that they had gotten my letters, but could not publish them because ``we receive so many letters from readers, etc.'' I found this hard to understand. The Dvorak Simplified Keyboard (DSK) offers an increase in productivity of 30% to 50%, while the articles in Business Week were making a big fuss over any businesses that were able to achieve an increase of 10%!
The internal combustion engine is not the only sacred cow that's giving us trouble these days. In 1962, when I was in Seattle for the World's Fair, I called up Dr. August Dvorak, the inventor of the DSK. This keyboard is a scientific rearrangement of the letters on the typewriter keyboard which allows efficient and speedy typing. When I called Dr. Dvorak, I was somewhat surprised to find out that I was talking with a bitter man:
``I'm tired of trying to do something worthwhile for the human race,'' he said. ``They simply don't want to change!''
He told me that he'd been trying to introduce his keyboard for 30 years, and had been blocked at every turn. At the time, he was already 68 years old, had retired from the University of Washington in Seattle where he was a professor of education for many years, and had evidently given up all hope of his invention ever gaining acceptance.
This conversation with Dr. Dvorak sparked my interest. I eventually converted my own typewriter to his keyboard, and learned his system myself.
Before going on with the history of Dr. Dvorak's struggle to have his invention accepted, we should look at how the typewriter keyboard most in use today came about. It turns out that this keyboard was designed experimentally by Christopher Sholes, the inventor of the typewriter, to slow the typist down. The keys on the early machines hung down in sort of a basket arrangement, and pivoted up to strike the platen (roller) from underneath. To see what you had typed, you had to lift up the roller so you could look at the paper. Since the keys had no springs on them, they fell back into place by gravity. This meant their action was very sluggish and if two keys that were close together in one quadrant of this ``basket'' were struck rapidly, one after another, they would jam. To overcome this problem, Sholes moved the keys around experimentally until the machine seemed to operate with a minimum of jamming. What he actually did was to make many commonly-used letter sequences awkward and slow to execute. Thus, by ``anti-engineering'' his typewriter from a human factors point of view, he was able to slow it down so it would function to his satisfaction. Now, when we have typewriters that are mechanically quite responsive, we are still bound by the old keyboard found on those first (1873) machines. What an irony.
Upon analysis, Dr. Dvorak found that the ``standard'' keyboard had several defects. These can be summarized as follows:
After several years of intensive research, during which hundreds of keyboard arrangements were studied and rejected, Dr. Dvorak received a patent for his Dvorak Simplified Keyboard in 1932. The DSK solves the basic problems inherent in the standard keyboard.
Better hand alternation: The hand overload problem is solved by maximizing alternate hand stroking. This is particularly important in maintaining rhythm. As much as possible, successive strokes should fall on alternate hands. This allows what is called ``play for position.'' That is, while a finger on one hand is in the process of stroking a key, another finger on the opposite hand can be getting into position to stroke the next key, and so on. The longer such alternation keeps up, the more even the typing rhythm. Dr. Dvorak solved this problem by putting the vowels (which comprise 40% of all typing) on the left hand side of the keyboard, and the major consonants that go along with those vowels on the right hand side. This guarantees good hand alternation since most syllables are made up of alternating vowel/consonant letter sequences.
The figure above shows a short selection from a story called Fraser Street. This text is made up almost entirely of one-hand words on the standard keyboard. Made into a typing test, Fraser Street is extremely difficult for a normal typist. Only exceptionally good typists (in the 80-word-per-minute-and-up range) can even be expected to finish the test at all in the standard time.
For our purposes, Fraser Street becomes in illustration of how the DSK guarantees good hand alternation by putting all of the vowels on the left hand side. In the figure above, all of the left hand strokes are shown by white letters on a black background, while the right hand strokes are printed in the normal ``black letter on white background'' manner. It can be seen how the DSK automatically breaks words up into left-right vowel-consonant sequences, thereby insuring superior rhythm.
Better finger loads: The ``dactylographs'' shown here illustrate, by the lengths of the fingers, the relative work done by each finger on the Dvorak and on the standard keyboards. The bold numbers 1, 2, ..., 8 are placed on the fingers to indicate the relative abilities (combination of strength and dexterity) of each. Thus, the right index finger (number 1) is the ablest, while the left little finger is the weakest. Note how the DSK arrangement precisely divides the finger loads according to relative finger capabilities. On the standard keyboard, the finger loads are nowhere near being properly distributed.
More work (70%) done on home row: To minimize excess finger movement, the most frequent letters and letter sequences were placed on the home row of the Dvorak arrangement where 70% of the typing is then concentrated. One can compare where the same 70% is done on both keyboards by looking at this figure.
The effect of placing the most frequently used letters and letter sequences on the home row is illustrated in the figure below. In this example, the ``Gettysburg Address'' was typed with all off-home-row strokes shown by white letters on a black background, while the on-home-row letters are printed with black letters on a white background. The preponderance of off-home-row strokes on the standard keyboard is dramatically evident in the left hand diagram of the figure below, while on the DSK there is almost the exact opposite situation with 70% of the work being done on the home row.
Awkward strokes minimized: The rest of the characters, comprising the typing that has to be done off the home row, are placed on the DSK in positions on the remaining rows according to how hard it is to strike the keys in those rows. This is done such that the total number of awkward strokes is minimized (the ``awkwardness'' of various types of strokes was determined using high-speed time-and-motion movies). The effect of this can be seen in the diagrams below. Since awkward strokes are sometimes slower by a factor of three to one, and since the DSK reduces the number of these strokes by a factor of ten to one, one can see how it is possible to achieve a faster typing rate on this keyboard (and why Dr. Dvorak's students hold 12 out of 15 unbeaten world typing records).
With all these design improvements, it thus becomes easy to believe (a belief that has been proven experimentally) that the DSK is:
At this point, a quote from Dr. Dvorak himself might prove enlightening:
``The reason the DSK, patented in 1932, is not generally used today is the same reason that it took 35 to 70 years for the railroads, steamships, radio, telephone, telegraph, airplanes, and yes, even the automobile to gain general acceptance. Each required implementation: investment of time, money, and effort. Each was opposed by those who had investments in the status quo, e.g., the canal and barge owners, the Pony Express, the stagecoach operators, etc. The general public shies away from ``new-fangled'' things and ideas, especially when they are given a strong negative sell by dealers in what was good for their grandfathers. Incidentally, standard keyboard typists, remembering their laborious and frustrating travail in learning to type, worry about ``unlearning'' the old keyboards and ``won't go through that again.'' I'll venture that if I could give you a Rolls-Royce car with a seven forward, two reverse gear shift, you'd learn to use it, with no concern about unlearning your present car's shift. In Africa,'' the Doctor says, ``I learned to drive on the left side of the road and streets with a fouled-up gear shift with little strain.''
Let's look at some of the problems that Dr. Dvorak encountered when he brought out his new keyboard. First, in 1932, the year his keyboard was patented, the country was in the depths of the Great Depression. The typewriter companies were almost broke; so, naturally, they didn't take too kindly to an inventor coming to them and saying, ``If you put my keyboard on your typewriters, you will be able to do twice as much work with the same machine.'' The manufacturers took this to mean, ``Oh, you mean we will sell half as many typewriters? Well, thank you very much. Don't call us, we'll call you.''
This feeling was seemingly passed over to the typewriter dealers. If you went into a shop then (and in many cases even today) and asked for a typewriter with the DSK, they would almost always try to talk you out of it. The reasons were always similar:
Certainly it is easy to understand that the typewriter companies were concerned with what they would do with their current stock of machines if a change were made to the new keyboard. They probably just felt they were protecting their investment.
In a letter from a typewriter head office to one of their branches, they pointed out that ``The Dvorak keyboard is not new ... has not been commercially accepted by the public ... for the reason that the present so-called standard keyboard has considerable merit and that typists for years have been taught the touch system on that keyboard. To introduce a keyboard with the alphabet keys rearranged in as radical a manner as the Dvorak keyboard would cause considerable confusion. ... If a school trained its typists on the Dvorak keyboard, they would have difficulty in locating a position where the machines were in use equipped with that keyboard. ... There is no definite evidence that the Dvorak keyboard will increase the speed of a typist regardless of statements to the contrary. Our most expert typists are able to write over 150 five-stroke words in a single minute, which means that they are able to hit over 750 keys in 60 seconds. A keyboard that is capable of being operated at that high rate of speed cannot be so badly arranged after all.''
Before continuing, some of the above arguments ought to be answered:
Another company commented openly in their advertising: ``No one has ever studied typewriting without worrying about the madly inconvenient arrangement of the keys ... (produced) to avoid jamming keys and similar problems. ... From very early in typewriter history, the idea of changing Sholes' nonsensical keyboard has been hopeless. Typist opinion was against change, and all of the companies that tried to prove that a more sensible key order was desirable, have long since departed! Typewriter buyers of the country know how to typewrite by `touch' and don't want to learn a different system. And before you invest time or money in a keyboard-reform scheme, consider the facts. If people would buy it, [name of company] would be selling it!''
It is interesting that here, the company has actually recognized that the standard keyboard arrangement is bad, but then proceeds to tell us why we should still not try to change it!
One could agree that typist opinion would likely be against the DSK. But, most people don't type. Many of them would like to. Why should they be forced to learn an old (1873) and awkward keyboard, when a more modern and scientifically designed one is available? That's as if the world's typists all belonged to a huge union that says: ``To join our most-esteemed group, you must learn to practice our trade exactly the way we learned to do it. No matter that you think you have discovered a better (faster, easier-to-learn) way. We are against your way.'' One can hardly argue that that makes any sense.
From 1906 to 1932, 26 years, typewriter manufacturers used annual World Professional and Amateur Typewriting Contests to prove the merits of their machines and for advertising. For the professional contests, manufacturers maintained ``speed stables'' of outstanding typists whose duties were to practice speed typing (while under full pay from their sponsoring company) and periodically demonstrate the superiority of their employers' machines. Some professional typists practiced up to 25 years to improve their speed and accuracy, which were widely advertised as ``World Records.''
Outstanding student typists from high schools and colleges, usually winners in state typewriting contests, competed in the ``World Novice and Amateur Contests.'' A student with one school year of typing instruction who could type at 60 net words per minute (n.w.p.m.) for 15 minutes frequently was the ``world novice champion.'' A student with two years' instruction with 70 n.w.p.m. frequently became the ``world amateur champion.''
Eventually, these contests were combined and included in the International Commercial Schools Contest (I.C.S.C.) to be held each year. The I.C.S.C. also include categories for shorthand, machine calculation, and dictating machine transcription, in addition to the typing events. The extent to which the equipment manufacturers and publishers of shorthand materials subsidized I.C.S.C. was not widely publicized.
Beginning in 1933, Dr. Dvorak started entering his DSK-trained typists in the I.C.S.C. His students began ``sweeping the field.'' Ten times in 1934-41 DSK typists not only placed first in their class event, but also placed first in events for contestants with much more training. In the 1935 contests, nine DSK typists won twenty awards. The contest officials became more and more upset. In 1937, after Dvorak spent $1,600 bringing nine contestants to Chicago, the I.S.C.S. Committee informed him that DSK typists were to be disbarred from competition because they were ``unfair competition.''
However, Dr. Dvorak was not one to be bullied. He hinted that the newspapers might find the disbarment of his students most interesting, especially since the Contests were supposed to be held to advance the skill of typing, not hold it back. The Committee reconsidered, and Dr. Dvorak's students were allowed to remain in competition.
After these events, however, Dvorak and his students were not received in a friendly sportsmanlike manner in the interest of commercial education. Standard keyboard typists objected to being placed near DSK typists because the noise of their high rates was disconcerting.
One year, Dvorak's machines were even sabotaged before the contests began. Someone reset the margin stops on all of their machines by just enough to cause line length and paragraphing errors. Many typists were disqualified because of this. In the following years, Dr. Dvorak had to hire security guards to watch his typewriters prior to the start of the competition.
Another interesting thing was the way they reported the winners. The score would be given, along with the brand name of the typewriter they used (e.g., IBM Electromatic, etc.). However, when Dvorak's students began winning with quite superior scores, there was no mention that they had used the DSK, only that they had used a machine produced by such-and-such manufacturer (after all, the real purpose of the contests was to prove the superiority of each manufacturer's machines, not the keyboard used on those machines).
During World War II, the I.C.S.C. were cancelled. In 1946, when they started up again, Dr. Dvorak had no students ready to compete because he had been serving in the armed forces. With no DSK typists in the contests, the performances on the standard keyboard were so dismal (at least one contestant won a third place with a zero net score) that they did not bother to announce the winning scores of each winner at the awards ceremony, as had been the case in previous years. After that, they decided to cancel the competition altogether because they ``proved nothing'' (except perhaps the superiority of the DSK?).
Many people have suggested proving the worth of the DSK by running experimental classes. The question is: What will be done if the experiments show that the DSK is indeed all it is claimed to be? The plain facts are that the DSK has been proven experimentally, but those in power in each case chose to disregard the results of the experiments and everyone just went on using the standard keyboard arrangement.
Some examples would be helpful here:
Tacoma Schools Experiment. During the Depression of the 1930s, an experimental program in personal typing was instituted by the school district in Tacoma, Washington. Great care was taken to choose students who wanted to use the typewriter for personal use, rather than in a business environment. Parents understood that they would have to purchase DSK typewriters for their children to use after finishing these experimental classes.
Two thousand seven hundred students were put through the various courses in DSK typing. These classes showed that senior high school kids could learn the DSK in one-third the time it took to learn the standard keyboard. The program was an outstanding success, and was reported in various educational publications.
But, then came a school board election. And typing in the schools became a political issue: whether or not they should allow the Simplified Keyboard classes to continue, etc. The man who was against the new keyboard won the election. So, what did he do? He ran a survey. He asked businesses in the area how many DSK machines they had in their offices. Answer: None. Then, he asked how many standard keyboard typewriters the had. Answer: Why, all of them, of course. On these grounds, he closed down the personal typing classes (regardless of the fact that these students were not planning to go into office typing, but wanted the typewriter for their own personal use). It's amazing what one man can do to help shoot down a good idea.
U.S. Navy Department Report: One of the most interesting experiments was conducted by a group of management engineers in the U.S. Navy Department in 1944. In this test, they retrained a group of standard keyboard typists on the DSK in a period of about two and a half months. The retrainees' progress was also compared with that of a group of standard keyboard typists who were given some additional training on the regular keyboard. The results, together with the data supporting them, were most conclusive. The DSK retrainees increased their productivity by an average of 74%! Not only that, the total cost of retraining was completely amortized in only ten days after the tests were finished.
The improvement in the comparison group was much less dramatic, amounting to only about 26% increase, and the comparison group took twice as long to acquire their slight increased performance.
On the basis of this test, the Navy Department issued a request for bids for 2,000 DSK-equipped typewriters. They figured that the amount of money that would be saved during the war effort would be tremendous. But the request was turned down by the Procurement Division of the U.S. Treasury Department (which was responsible for all government purchases of typewriters at the time). No satisfactory reason was given (at least from the viewpoint of the Dvorak proponents). The request was simply denied!
Later on, Dr. Dvorak heard over the ``grapevine'' that the reasoning went something like this:
Although it would probably be very hard to prove whether or not this rumor was in fact true, one can nevertheless ask: ``If this is not the case, then why was the order turned down by the Treasury Department Procurement Division?'' Surely they must had some reason for the rejection. If it was not political, then what was it?
The report is in two parts:
Another interesting thing about this report was that after it was completed, it was given a security classification by the Navy, which again meant that an ordinary citizen could not gain access to it. What a report on typewriters could have had to do with ``endangering the national defense'' was never quite explained. This classification was lifted later on; but, nevertheless, it still happened.
``Unbiased test'' by the General Services Administration: What really almost killed the DSK was a test conducted by the U.S. General Services Administration (GSA) in 1956, under the direction of Dr. Earl P. Strong. If you ask anyone in the U.S. Government about the DSK, they will invariably say that the GSA test ``proved'' that the Simplified Keyboard wasn't any good. The conclusions of this study state, in part: ``... the Standard Keyboard results are better ... recommendation for the adoption of the Simplified Keyboard for use by the Federal Government cannot be justified based upon the findings of this experiment.''
However, some old correspondence that has recently come to light seems to indicate that Dr. Strong might not have been the ``impartial researcher'' he claimed. In a letter dated September 13, 1949, Dr. Strong states: ``... I have developed a great deal of material on how to get this increased production on the part of typists on the standard keyboard. Consequently, I am not in favor of purchasing new (i.e., Dvorak) keyboards and retraining typists on the new keyboard, and I am out to exploit it to its very utmost in opposition to the change to new keyboards.'' This seems to indicate that the U.S. Government ought to conduct some new, more objective tests, and that the data and results of these tests ought to be made freely available to the public for open discussion.
If new tests are to be run, they ought to be run on a strictly scientific basis, and the testing should be carried out over a period of at least a year in order to allow proper evaluation. At least four groups of students should be provided for:
Most importantly, provisions should be set up to insure that all of the experimental data (including test papers, progress records, etc.) are saved and made available to the public (or to qualified researchers). It seems this was not done with the GSA tests, much of the class material evidently having been either lost or destroyed at the end of the trial.
The importance of typist efficiency can be understood when you stop and think that there are, at present, several million people who earn their living wholly, or in part, with typing. What would be the value of a 50% increase in stenographic production (which the DSK could provide) to business and government? For two million workers earning $5,000 a year, it would be $5 billion annually. And the increased cost to typewriter manufacturers of assembling type for the DSK arrangement? Relatively, very small.
In case you want to write the GSA itself, you can ask for A comparative experiment in simplified keyboard retraining and standard keyboard training, sponsored by General Services Administration, Washington, D.C., 1956.
The evidence all seems to indicate that the Simplified Keyboard is really better. Many people may ask: ``Surely we are more progressive today. Don't you think the DSK will catch on now?
Unfortunately, that is not necessarily the case. It seems that there are still many forces around that have an interest in the status quo, in preserving the sacred cows.
Take the American National Standards Institute, for example (abbreviated ANSI). This organization is responsible for determining various standards needed in business and industry today in the U.S., and includes a committee that recommends keyboard standards. In a recently published article in Datamation, a well-known computer journal, the ANSI Keyboard Committee proposed a keyboard standard based exclusively on the old layout that is on most of our typewriters today.
Let me quote one of the reasons given for this proposed standard:
Research consistently revealed that the overriding criterion for continued use (of the Standard Keyboard arrangement) was the millions of people already familiar (and those trained annually in schools) with this arrangement.
No thought is given to those people who don't now type, but might like to learn if only it weren't so hard and time-consuming to do so. It looks like preservers of the status quo are at it again! And who are the members of the Keyboard Committee? As you might guess, there are many representatives of manufacturers of keyboard devices, most of whom, it might be said, have heavy investments in the status quo. The membership list seems to indicate that there might be a slight bias for this committee to choose the normal keyboard layout over any other that might come up. The percentage of ``users'' on the committee does not constitute a majority of the voters. Thus, the manufacturers can always have their way by voting together. (I have heard there are other ANSI committees with a similar balance of power in the hands of the manufacturers and suppliers.)
The ANSI Keyboard Committee certainly can't feign ignorance of the DSK. One of their members has for some years been trying to get them simply to include a mention of the DSK arrangement in their standards: to define a ``family of standards,'' so to speak. Then let people choose which member of the family best suits their needs.
As a matter of fact, the DSK has been before the standards organization for many years.
In the October, 1943, issue of Industrial Standardization, published by the American Standards Association (predecessor of ANSI), there was a discussion of the Simplified Keyboard. One member of the current committee has frustratedly written:
There are good reasons for asking for delay in the case of the X4.7 typewriter keyboard standard, in that the proposal was arrived at without laboratory research and without consulation with qualified research engineers, psychologists, or specialists in education and training.
I, for one, hope this is not the way our standards organizations are being run. Were it true, I would question whether we could really trust a vested-interest-laden committee to make very important decisions for everyone else in the country. Such decisions should not be based largely on tradition.
One thing that is holding back wide-scale training efforts on the DSK is a way of converting a typewriter easily back and forth between the standard keyboard and the DSK. IBM is currently evaluating the potential uses of the DSK on their Selectric Typewriters (Selectric is a trademark of the IBM Corp.), which utilize removable typing elements. Unfortunately, IBM wants to charge $20,000 for the initial tooling of the first DSK element. Few organizations, outside of the governmnent, have operations large enough to justify an expenditure of this size.
Perhaps the typewriter companies should follow the example of Xerox Corp. Xerox designed their machines so that all you had to do was to press a button to get a copy. This greatly simplified the entire process, and the copiers caught on like mad. In a similar way, you can make typing simpler. Then more people will use the typewriter. Besides the benefits of increased productivity in business and government, the DSK conversion would mean increased sales of typewriters to the public. Because typing could suddenly become easy to learn, many more people would decide they wanted to learn to type, who would never have considered doing so before. After all, most people do recognize that typing is a valuable skill; they just figure it's too hard, or they don't have the time. Undoubtedly, many of the new machines sold would go to people who don't type now (a market the manufacturers have been after for many years, without much success). The end result would be beneficial to us all.
Henry Ford is said to have remarked, ``Hire the best engineers you can find to build a car that morons can drive.'' I think it's safe to say, he did, and they do!
Imagine how many people would be driving today if it took three full years (the time it takes to gain an ``acceptable'' skill, on the average, with a standard keyboard typewriter) to learn how to operate a car.
Also, Dr. Frank Gilbreth (under whose direction Dr. Dvorak began the research that led to developing the DSK), the father of time and motion study, said:
It is cheaper and more productive to design machines to fit men rather than try and force men to fit machines.
The U.S. space program strives to obey this principle, as its importance becomes very obvious under stressful conditions. Why can't we give the millions of typists in the country the same considerations?
We are attempting to face the problems created by the internal combustion engine, and have recognized the need to convert to the Metric System of weights and measures. Since more people use the typewriter to make a living than any other single business machine in the world today (as was pointed out by Dr. H. Forkner, the inventor of Forkner Shorthand), isn't it only fair to examine needed improvements in this area too? I think the answer is yes!
More than thirty years have passed since the appearance of the article above. A few points from this perspective:
Parkinson's article above was used without permission. In the process of typing it in, I modernized the punctuation slightly, but tried as much as possible to adhere to the original wording. All emphasis in the original with underlining; my choices of when to use italics and when boldface were somewhat arbitrary. All artwork is scanned from a photocopy of the original article.