Assistive Technology Research Institute
College Misericordia - Dallas, PA 18612
 
Founded and Sponsored by the Sisters of Mercy of Dallas

 

Efficacy of Alternate Keyboard Configurations

 



Efficacy of Alternate Keyboard Configurations: Dvorak vs. Reverse-QWERTY

Denis Anson, MS, OTR, Christa L. Eck, OTS, James King, OTS, Regina Mooney, OTS, Christopher Sansom, OTS, Bryan Wilkerson, OTS, and Daniel Wychulis, OTS

Abstract

Clinicians working to improve performance for clients in work or school settings need to consider the possible benefits of providing an ergonomic and easily learned keyboard for text generation. The standard QWERTY keyboard was designed to avoid mechanical problems in early typewriters while ignoring ergonomic and cognitive issues for the typist. In spite of this design, previous studies of alternative keyboard arrangements have either shown QWERTY to be superior, or showed no significant improvement when the keyboard was rearranged to a more ergonomic pattern. One likely reason for this finding is the subjects in keyboard studies have many years of experience on the QWERTY keyboard and only a few hours of practice with the alternatives. This study sought to provide a comparison of the Dvorak keyboard with the QWERTY while controlling for prior experience by inverting the familiar QWERTY keyboard. Results of the study support assertions that the Dvorak keyboard allows faster typing, may allow faster learning, but do not support claims for more accuracy when using this alternative keyboard. This study provides support for clinicians seeking a more effective text generation technique for individuals with physical disabilities.

Background and Significance of the Issue

In 1874, the first commercial typewriter was marketed by E. Remington and Sons (Rehr, n.d. b). On the "Sholes & Glidden Type Writer", the keyboard was modified from earlier unsuccessful devices, which used alphabetical order, by arranging the high frequency letters as far apart, mechanically, as possible. This modification minimized key jamming, and allowed, for the first time, typing to be performed at a functional rate. Over the next 130 years, this keyboard, commonly known as QWERTY for the first six letters of the top row, became the standard for all keyboard products, in spite of its generally recognized poor ergonomic design. Because only a relative few people used keyboards for any significant part of the day for the first 100 years of the typewriter's existence, the QWERTY design was generally accepted, and little impetus was felt to change the keyboard layout.

Almost 100 years after the invention of the typewriter, the development of the microcomputer brought a larger portion of the population into contact with keyboards on a daily basis. Nearly half of all American households today use the Internet. More than half of all US classrooms are connected to the Internet (White House, 2000). It has been estimated that over 80% of all jobs require some computer access. As the keyboard became a pervasive factor in daily life, the issue of computer ergonomics became more pressing. In 1994, repetitive stress injuries, commonly associated with keyboard usage, accounted for 60% of all job-related illnesses (Wolkomir, 1994).

Concern about keyboard layouts is not a new phenomenon. "As technology developed from the inception of the typewriter in the 1870's to the 1930's, the problem of key jams decreased accordingly"(Rehr, n.d. a.). "At the same time, the science of studying body motion began to emerge and the QWERTY keyboard was [re-evaluated]"(Maxfield & Montrose, 1998). The Dvorak keyboard, developed and patented by August Dvorak and W.L. Dealy in 1932, offered a more practical "arrangement of the keys that let 70 percent of the typing take place in the home row" (New York Times, 1999). The Dvorak layout "placed the most commonly used letters, both consonants and vowels, on the home row where finger travel was a minimum; thus increasing the speed and efficiency with which one could type"( Rehr, n.d. b ). A 1944 study demonstrated that typists could match their QWERTY typing speeds with an average of 52 hours of training on the Dvorak layout. "At the conclusion of the 83- hour study, the typists had reached speeds that were 74 percent faster than their previous QWERTY speeds while accuracy increased by 68 percent" (Tenenbaum, 1996, pp. 23). Why wasn't this efficient method embraced and adopted by anyone working on a typewriter or keyboard? "Even when daisy wheels and computer printers replaced type bars, forever eliminating the jamming problem that had originally motivated QWERTY, manufacturers of the efficient new technologies carried on the inefficient old keyboard" (Diamond, 1997). Several factors contributed to keep the Dvorak a technological footnote of history; first and foremost, the concept of path-dependence, the idea that small, random events at critical moments can determine choices in technology that are extremely difficult and expensive to change.. [In the end] the weight of numbers makes the leading product more valuable than one based on competing technologies. (Passell, 1996, para. 4)

Secondly, "change is not always readily accepted by those who have already been trained to use one method"(Brooks, 2000). Third, ".a world war impeded the acceptance of the Dvorak"(Rehr, n.d. a ), and by their close, the QWERTY was 'locked-in' to society as the standard keyboard.

Several studies have been conducted in the past that compared the QWERTY to the Dvorak, (see Liebowitz and Margolis, 1990) but biases and confounding variables existed during those research trials that clouded the results - almost everyone has been exposed to the QWERTY format at some point during their life, which gives it a certain familiarity. This new study will attempt to eliminate that factor by flipping the QWERTY format both vertically and horizontally thus creating a ReverseQWERTY. This allows the study subject to begin the study with two new and unlearned keyboards while keeping the biomechanical properties of the QWERTY key layout intact.

Focused Literature Review

As previously mentioned, change is not always readily accepted. In his book, The Design of Everyday Things, Donald Norman (1998) describes the history of the QWERTY and other keyboard layouts. He states, "There is a better way - the Dvorak keyboard - . It is easier to learn and allows for about 10% faster typing, but that is simply not enough of an improvement to merit a revolution in the keyboard.The severe constraints of existing practice prevents change, even where the change would be an improvement"(p. 147-150). Debates continue from various professional & educational arenas about which layout is superior. Diamond (1997) argues against the QWERTY layout stating,

Only 32% of strokes are on the home row; most strokes (52%) are on the upper row; and a full 16% on the bottom row.Not more than 100 English words can be typed without leaving the home row.Qwerty perversely puts the most commonly used English letters on other rows. The home row of nine letters includes two of the least used (J and K) but none of the three most frequently used (E, T, and O, which are relegated to the upper row) and only one of the five vowels (A), even though 40% of all letters in the typical English text are vowels (para. 8).

Is the QWERTY layout as inefficient as Diamond suggests and is the Dvorak far superior as claimed by its proponents? In their article, "Fable of the Keys", S.J. Liebowitz and Stephen Margolis (1990) argue that the Dvorak has no proven advantage over QWERTY. They refer to a study conducted in 1956 by the General Services Administration of the Federal Government. At the end of this experiment, it was "concluded that retraining typists on Dvorak had no advantages over retraining on QWERTY" (p. 7).

Advocates of Dvorak look to a study done in 1944 by the U.S. Navy. This study compared the efficacy of the Dvorak versus the QWERTY. This study claims that subjects learned the Dvorak layout nine times faster than the QWERTY. In addition, the test subjects unanimously noted their approval in the use of the Dvorak keyboard in relation to the ease of learning, increased accuracy, and decreased fatigue (Tennebaum, 1996). Liebowitz and Margolis (1996) imply that the data collected from the Navy study was contaminated due to conflict of interest. August Dvorak himself conducted the study as "the Navy's top expert in the analysis of time and motion studies during world war II" (p. 7). They go on to say "he (Dvorak) owned the patent on the keyboard."

Proponents of Dvorak argue it has advantages other than faster learning, stating Dvorak was generally the first to recognize and focus on the human factors (Shieh & Lin, 1999). He studied digraphs or two-letter sequences commonly found in printed text, such as rt, tr, ed, de, fr, sw, ty, gy, and lo. While using the standard QWERTY keyboard, the two letters of these diagraphs are typed by the same finger, resulting in a decrease in typing speed. In the Dvorak layout, the number of times the same digit types consecutive letters is greatly decreased allowing faster parallel processing of typing. Further, "the Dvorak keyboard.reduces travel of the hands lessening fatigue" (Struck, 1999, p. 55). Due to the growing concern about repetitive stress injuries, or cumulative trauma disorders (CTD) over the past twenty years in relation to typing, researchers were prompted to investigate the use of the Dvorak keyboard in reducing the incidence of CTD.

Rick Robertson, in a study conducted in the Department of Physical Education and Rehabilitation at the University of Pittsburgh noted that using the Dvorak keyboard over the standard QWERTY reduced tendon movement by 6 to 10 percent thereby reducing the potential for CTDs. Robertson also stated, "6 to 10 percent doesn't sound like much, but it can make a significant difference to people who do a lot of typing" (Robertson as cited in Tennebaum, 1996).

Based on the review of the literature and the previously conducted studies, the debate continues as to which layout is superior. Further experimentation is needed to substantiate which layout is better in terms of speed and ease of learning.

Figure 1. The ReverseQWERTY Keyboard Layout (Click image for larger view)
The ReverseQWERTY Keyboard Layout has the keys reversed horizontally and vertically.

Some of the difficulty in comparing the two keyboard layouts for human performance may stem from the pervasiveness of the QWERTY keyboard. While few typists are familiar with the Dvorak layout, most potential research subjects have many hundreds of hours of exposure to the QWERTY keyboard layout. Thus, the QWERTY layout has a marked advantage in direct comparisons of typing speed over any reasonable duration. In order to control for this prior learning, this study uses a variant of the standard QWERTY layout, developed to maintain the biomechanical relationships of the keys while providing a novel arrangement of the keys. This is accomplished by inverting the QWERTY layout vertically and horizontally. A dual axis inversion is considered necessary because Matias, Mackenzie and Buxton (1994, pp.88-94) have demonstrated that right to left inversions are easily learned. The ReverseQWERTY (See Figure 1) pattern, used in this study, allows comparison of two novel keyboard patterns, while maintaining the inter-key relationships of the QWERTY keyboard.

Purpose

The claims of the advocates of the Dvorak layout suggest the following research hypotheses:

  1. The Dvorak keyboard pattern allows faster learning of typing than the QWERTY;
  2. The Dvorak keyboard layout provides faster typing than the QWERTY;
  3. The Dvorak keyboard layout provides more accurate typing than QWERTY.

Method

Research Design

A single-subject repeated measures design was utilized in this quantitative study to test twelve subjects on typing fluency as measured by speed and accuracy. The time required to learn the keyboard was measured by the number of trials it took to reach fluency. During the study, the subject's typing speed and accuracy were monitored to determine the efficacy of the QWERTY, via the ReverseQWERTY design, and the Dvorak keyboards, respectively.

Procedure

Sample Subjects

The subjects included seven females and five males subjects. Each researcher located two subjects within their respective communities who voluntarily participated in the experiment. Subjects had to be able to sit unsupported for up to an hour at a time, to read printed English text, and to operate a standard keyboard in order to be included in the study.. The subjects ranged in age from 20 to 40 years. Ten of the subjects had minimal typing experience prior to taking part in the study and 2 had frequent or daily typing experience prior to the study. Three subjects were left-handed and the remaining nine were right handed. The majority of testing was completed in the morning or evening hours. None of the subjects had identified upper extremity limitations, identified cognitive limitations; uncorrected visual limitations, or prior experience with the Dvorak or ReverseQWERTY keyboards.

Although several computer stations were used in the course of this study, each subject completed all trials at a single station. In order to control for possible order effects in learning keyboards, half of the subjects started with the ReverseQWERTY keyboard, and half started with the Dvorak keyboard.

Equipment

Each researcher utilized a digital countdown kitchen timer to time the 20-minute typing tests. The printed test text was placed on a copyholder on the subjects preferred side of the computer monitor in a vertical position during each test.

Computer System.

Each researcher used a Windows-based standard personal computer with a 104 key keyboard, a 15-inch computer monitor and Windows 95 or 98 operating systems for this study. All computers used during the study had processors operating faster than 200 MHz, and had memory greater than 64 MB. (This was necessary to assure that the computer did not slow down the typist.) The key labels for each layout were affixed to keyboard safe-skins , which were placed over the keyboards for all typing trials. The Dvorak layout was provided using Keyboard control panel of the Windows operating system. The ReverseQWERTY set-up was provided using "Keyboard Layout Manager," a software based keyboard redefinition program.

All text was typed into Microsoft Word 97. No other programs were operating during the testing session.

Software

The Microsoft Word tool 'Compare Documents' was used to determine the accuracy of each test by marking differences in the test document as compared to the original. For purposes of this study, each difference counted as a single error, although there might have been more than one character difference within the detected block of text. A single error might include a missing letter, an incorrect letter, a letter reversal, or even an entire missing line.

The number of words typed in a twenty minute block was determined using the "Word Count" feature of Microsoft Word. The counted number of words typed was divided by 20 minutes to determine the average typing rate in words per minute.

Text

The subjects were provided with novel text for each typing session from Aesop's Fables found on the Guttenberg Project web site of online books. Aesop's Fables was formatted to 14 point, New Times Roman font. The text was broken up into 500 to 600 word increments for individual text blocks, and labeled from 1-40. The text was printed for use in the test trials and a disk copy was maintained for use in the compare documents tool. Each subject had the same text in equivalent trials (e.g. Trial 5 of the Dvorak), but an individual subject used unique text in each consecutive trial. The degree of difficulty of the typed text, as calculated by Microsoft Word Flesch Reading Ease Score, remained similar throughout the experiment ranging from grade 6.3 to 9.7 for the Dvorak and 6.6-8.9 for the ReverseQWERTY trials.

Data Collection Procedures

At the beginning of each trial, the researcher converted the keyboard to the appropriate layout for the subject and trial. The subject was seated with feet on the floor and the keyboard placed so that his or her elbows were flexed at approximately 90 degrees during typing. The monitor was placed at a distance and height that the subject reported to be comfortable. At the beginning of each trial the subject was provided with the text for that session. The text to be typed was supported in a vertical plane on the subject's preferred side of the monitor. The subject was provided with new text for each trial period of 20 minutes of typing. The subjects were instructed that the test was on speed in words per minute and accuracy. The subject was told, "When I say 'Go', type this text as quickly and as accurately as you can. After 20 minutes, I will say 'Stop.' At that time, please stop immediately." After each data collection section, the typed text was saved with a unique code so that the date and trial number were preserved. This typed text was then counted for the typing rate, and compared to the source text to determine the accuracy of the typing in that session. Trials were continued until the subject achieved fluency on the keyboard layout being tested, although no more than three sessions were performed during each meeting. Fluency was achieved when the subjects typing speeds were within 7 percent on three consecutive trials. Once fluency was attained, the subject was switched to the other keyboard layout (ReverseQWERTY or Dvorak). Again the subject continued typing until fluency was achieved. The ReverseQWERTY and Dvorak fluency scores were compared based on speed in words per minute and accuracy as they were determined through the number of variances detected by "Compare Documents". The time required learning the keyboard was measured by the number of trials required to reach fluency.

Data Analysis

The data for each trial was entered by the researcher who collected it into an Excel spreadsheet. This spreadsheet calculated the Words per Minute and error rates for each trial. Additionally, the spreadsheet generated graphs of the data for visual inspection.

Results

Of the subjects in the study, all twelve of the participants achieved fluency on both the ReverseQWERTY and Dvorak keyboard layouts. At the point of reaching "fluency," none of the subjects was touch-typing; rather all used some form of two digit typing like "hunt-and-peck." The number of trials, and typing speed at fluency, are shown in Table 1.

Table 1. Typing rate at fluency and number of trails required to achieve fluency.

 

Subject

Fluency in WPM

Trials to Fluency

Dvorak

Reverse QWERTY

Dvorak

Reverse QWERTY

A

12.3

9.48

11

9

B

17.2

9.36

14

10

C

14.2

14.0

6

6

D

12.1

10.7

7

8

E

16.2

9.4

9

6

F

14.6

12.0

9

7

G

13

9.8

8

8

H

12.05

10.3

8

9

I

14.73

12.65

12

9

J

16.56

14.5

12

9

K

14.35

15.2

7

10

L

18.1

13.4

8

10

Figure 2. Typical performance when moving from Dvorak to ReverseQWERTY. (Click image for larger view)
This subject plateaued at more than 12 words per minute on the Dvorak keyboard, then, when changed to the ReverseQWERTY, showed a marked drop, with an eventual plateau at about 10 words per minute.
Figure 3. Typical performance when moving from ReverseQWERTY to Dvorak. (Click image for a larger view)
This subject plateaued at about 10 words per minute with the ReverseQWERTY layout, and when switched to the Dvorak layout, showed continued improvement to about 20 words per minute.

While the number of errors varied among the individual participants, there were no systematic patterns in the error rates. This does not support hypothesis 3, that the Dvorak keyboard provides more accurate typing than the QWERTY. The number of trials required to achieve fluency on the Dvorak was typically higher than that required to achieve fluency on the ReverseQWERTY keyboard, with a range of 6-14 trials. The number of trials to reach fluency on ReverseQWERTY ranged from 6-10 indicating that more time was required, on average, to obtain fluency and master the use of the Dvorak as compared to the ReverseQWERTY. Six out of twelve participants required more time to obtain fluency using the Dvorak keyboard layout versus four out of twelve requiring more time on the ReverseQWERTY while two participants demonstrated fluency acquisition on both layouts in the same amount of time. When examined on a trial by trial basis, however, we find that, for those typists who took longer to achieve fluency on the Dvorak keyboard, they had already exceeded their fastest typing on ReverseQWERTY at the number of trials where they had achieved fluency with the ReverseQWERTY. From this point, they continued to make gains before reaching a plateau on the Dvorak keyboard. Thus, while these typists took longer to achieve their final typing speed, they achieved the limiting typing speed of the ReverseQWERTY faster on the Dvorak keyboard than on the ReverseQWERTY. The results of typical typists in this study are shown in Figures 2 and 3. It will be seen that Subject A began typing faster on the Dvorak than the ReverseQWERTY, while, after 7 trials, subject B was typing faster on the Dvorak than the ReverseQWERTY, but continued to make gains over a number of trials.

Discussion

The results of this study support the assertion that the Dvorak keyboard allows faster typing than the QWERTY. Each of the 12 subjects in the study achieved faster typing using the Dvorak layout than the Reverse QWERTY layout. Ten of the twelve participants showed higher initial typing speeds using the Dvorak than the Reverse-QWERTY.

The results do not support the assertion that the Dvorak keyboard allows more accurate typing than the QWERTY. There were no systematic differences in error rates between the two keyboards, or over the course of the trials. The errors per word did not appear to relate to the keyboard being used, the trial number, or the typing speed of the subject. It appears that some other factor, not examined in this study, determines accuracy of typing, at least for the range of speeds that these subjects achieved.

The assertion that the Dvorak keyboard may be learned faster and more easily is partially supported. While six of the 12 subjects required more time to achieve fluency, they had achieved faster typing rates at an equivalent time. Hence, for a criterion-based measure of learning (time to type 12 words per minute, for example), the Dvorak would be considered faster to learn. When considered from the point of view of ultimate fluency, this argument would not be supported, as half of the subjects took longer to achieve fluency with the Dvorak. The argument that although the overall time required to plateau with the Dvorak keyboard is greater than that of the ReverseQWERTY, the speed of typing is superior.

Additionally, many of the participants perceived the Dvorak keyboard layout to be easier to use compared with the ReverseQWERTY. One subject remarked, "After working with the ReverseQWERTY and the Dvorak layouts, the Dvorak seemed much easier to learn and use." Another subject remarked, "The ReverseQWERTY was very frustrating to use." There are two possible origins for this perceived difference. Since the ReverseQWERTY was presented as related to the QWERTY, but with the keys in different locations, the subjects may have experienced a cognitive dissonance that increased their feelings of frustration in trying to use the keyboard that was not felt when typing on the Dvorak keyboard. If this were true, one would expect the effect to be strongest for the most experienced typists. However, the two subjects with daily typing experience performed at about the same rate as other typists in the study, and did not express any stronger degree of frustration than the other typists. The difference may also derive from the fact that the QWERTY layout was originally based on the mechanical needs of early typewriters, while the Dvorak keyboard was developed based on letter frequency in normal typing. Hence, familiarity with English language and spelling would assist in learning the Dvorak keyboard, while the ReverseQWERTY keyboard provided no such supporting experience. The frustration subjects felt in working with the ReverseQWERTY may mirror the experience of learning the QWERTY keyboard originally, in which case, the Dvorak keyboard may well be easier to learn, as the subjects suggested.

There are some assumptions and limitations in this study that should be considered when applying the results to clinical populations. We assume that the ReverseQWERTY keyboard provides a valid homologue for learning the QWERTY keyboard. This assumption is weakened by the observation that, on the standard QWERTY keyboard, 52% of keystrokes are on the upper row of keys, and only 16% are on the lower row. This relationship is reversed for the ReverseQWERTY. If it is true that reaching to the lower row is more difficult than for the upper, then the effort of typing on the ReverseQWERTY does not mirror that of the standard QWERTY. There are two reasons to question this assertion, however. First, it has not been demonstrated that the "n" key is harder to reach than the "y" key, for example, so the comparative difficulty of top and bottom rows cannot be asserted. Second, none of the typists in this study achieved touch-typing with either keyboard. They all achieved their typing using a one or two digit "hunt-and-peck" approach that makes any differences in rows moot.

The results of a single subject design study may be considered less generalizable than those of a group design with random assignment to control and experimental groups. It must be noted that the subjects for this study were selected from different regions of the United States ( Pennsylvania , Texas , and Arizona ), and that all 12 achieved similar results. This provides strong support for the findings of the study.

Finally, it can be argued that, because the subjects in the study were able-bodied, the results of the study cannot be generalized to a population with disabilities. The main focus of this study was to determine if the use of the Dvorak keyboard led to a superior level of speed and accuracy in keyboarding. As with the wheelchair study performed by Deitz, Jaffe, Massagli, & Anson (1991) where ".the focus was on wheelchair performance," able-bodied subjects were used to separate the effects of the disability from the effects of the technology. Because this study was focused on determining the ease of learning, typing speed and accuracy of two "new" keyboards (ReverseQWERTY and Dvorak); the use of able-bodied subjects should not affect the overall results.

Conclusions

Prior researchers have failed to show any significant advantage of the Dvorak keyboard pattern over the QWERTY pattern. While it is possible that this is because the two keyboards are, in fact, equivalent for typing, it is more likely that this finding is biased by the pervasive experience of subjects with the QWERTY keyboard. The use of a novel keyboard pattern with the same physical relationships between the keys provides a means of comparing learning and speed on two novel keyboard patterns, with results that can be generalized back to the experienced QWERTY typist.

The results of this study support the assertions that the Dvorak keyboard allows for faster and easier learning of typing, and faster typing for the experienced typist. This is an important finding for clinicians who are considering alternative keyboard patterns for clients with fatigue or performance limits to productivity. Because initial typing rates on any new keyboard will likely be lower than on a familiar one, and because the subjects took longer to achieve fluency on the Dvorak keyboard, the study suggests that a clinician teaching this alternative pattern to clients must provide support over the time required to achieve fluid typing. Because the clinician now has evidence that, ultimately, typing will be faster using the Dvorak layout, he or she can provide this support with confidence.

As with any study, there are avenues for suggested advancement. In the current study, none of the subjects worked to achieve touch typing fluency. A future study should extend training on the Dvorak keyboard until touch-typing fluency, and compare ultimate typing speeds. The assertion that the ReverseQWERTY results represent typing on the QWERTY keyboard for a novice typist can be tested by replicating this study with young children who have not yet learned touch typing. (Because of the pervasiveness of the computer in the home, it would be very difficult to find subjects that are not exposed to the QWERTY pattern, but until they attempt touch typing, this may not matter.) Such a study of school-aged children could directly compare Dvorak and QWERTY typing.

References

Brooks, M. (2000, March). Dvorak comments and history . [Online] Available: http://www.mwbrooks.com/dvorak/comments.html [2001, August 17]

Costa, P.T., Jr., & McCrae, R.R. (1989). Personality continuity and the changes of adult life. In M. Storandt and G.R. VandenBos (Eds.). The Adult Years: Continuity and Change . Washington , DC : American Psychological Association.

Diamond, J. (1997, April). The curse of qwerty. Discover Magazine. [Online] Available http://208.245.156.153/archive/output.cfm?ID=1092 [2000, Sept 26]

Chronister, N. (2000) Friends don't let friends type QWERTY . [Online] Available: http://www.catskill.net/evolution/typing/ [2000, Sept. 26]

Deitz, J.C., Jaffe, K.M., Wolf, L.S., Massagli, T.L., & Anson , D.K. (1991) "Pediatric Power Wheelchairs: Evaluation of Function in the Home and School Environments." Assistive Technology , Vol. 3.1

Diamond, J. (1997, April). The curse of qwerty. Discover Magazine. [Online] Available: http://208.245.156.153/archive/output.cfm?ID=1092 [2000, Sept. 26]

Kincaid, C. (1999). Alternative keyboards. Exceptional Parent, 2, pp. 34-35.

Liebowitz, S., & Margolis, S. E. (1996, June). Typing errors. Reason On-Line . [Online] Available: http://www.reason.com/9606/Fe.QWERTY.html [2000, Sept 26]

Liebowitz, S. J. and Margolis, S.E. (1990, April) The fable of the keys. Journal of Law & Economics, 33 [Online] http://wwwpub.utdallas.edu/~liebowit/keys1.html [2001, July 19]

Matias, E., MacKenzie, I. S., & Buxton, W. (1994). Half-QWERTY: A one-handed keyboard facilitating skill transfer from QWERTY. Proceedings of the INTERCHI '93 Conference on Human Factors in Computing Systems (pp. 88-94). New York : ACM.

Maxfield and Montrose Interactive Inc. (1998). Circa 1936 ad: The dvorak keyboard . Madison , AL : [Online] Available: http://www.maxmon.com/1936ad.htm [2000, September 26]

Norman, D.A. (1988). The Design of Everyday Things . New York : Bantam Doubleday Dell.

Passell, P. (1996, May 5). Why the best doesn't always win. The New York Times . [Online] Available: http://www.nytimes.com/archives.html [2000, Sept. 26]

Patton, P. (1998, May 14). Public eye; windows shopping: What price freedom? The New York Times. [Online] Available: http://www.nytimes.com/archives.html [2000, Sept. 26]

Poole , C.J.M. (1995). Computers and the handicapped. British Medical Journal, 311, pp. 1149-1152.

Quinn, E. (1998, February 22). Keyboard ideology . [Online] Available: http://www.gontier.org/Quinn/WWW/ideology/Dvorak.html [2000, Sept. 26]

Rehr, D. (n.d.) Consider qwerty [Online] Available: http://home.earthlink.net/~dcrehr/whyqwert.html [2000, Sept. 6]

Rehr, D. (n.d.) The first typewriter. [Online] http://home.earthlink.net/~dcrehr/firsttw.html. [2001 , July 19]

Shieh, K.K. & Lin, C.C. (1999). A quantitative model for designing keyboard layout. Perceptual Motor Skills, 88, pp. 113-125

Struck, M. (1999). Focus: One-handed keyboard options. OT Practice, 4, pp. 55-56.

Tenenbaum, D. (1996, July). The typist's long-lost friend. Technology Review On-Line. [Online] Available: http://www.techreview.com/articles/july96/trends.html [2000, Sept. 26]

To speed up, typists slowed down . (1999, February, 18). New York , NY : The New York Times. [Online] Available: http://www.nytimes.com/archives.html [2000, Sept. 26]

The White House, Office of the Press Secretary. (January 21, 2000). Information technology research and development: Information technology for the 21 st century . [Online] Available: http://www.whitehouse.gov/WH/New/html/20000121_2.html [2000, Sept. 26]

Wolkomir Richard (1994, May). When the work you do ends up costing you an arm and a leg. The Smithsonian Magazine. pp. 90-102.


Please send all correspondence to Denis Anson at danson@misericordia.edu

Fentek Industries, Inc., P. O. Box 2278 , 470 South Main Street, Cottonwood , AZ 86326 , U.S.A. (800) 639-0710 . www.fentek-ind.com.

Keyboard Layout Manager: http://solair.eunet.yu/~minya/Programs/klm/index.html .

http://promo.net/cgi-promo/pg/t9.cgi?entry=45&full=yes&ftpsite=ftp://ibiblio.org/pub/docs/books/gutenberg/