sign in sign up
<<
Home , Dvorak keyboard layout

373

The Dvorak and Possibilities of its Regional Adaptation

Tomislav NakiC-AlfireviC, Marijan Durek Faculty of Electrical Engineering and Computing, University of Zagreb tomisluv.nakic-alfirevic@,kr. hr: marl:ian.diurek@,fer. hr

Abstract. During the last several decades, the 2. The Dvorak layout keyboard has proved to be the most important The Dvorak layout was constructed to computer input device. It was inherited @om the [1][2][3][4] be the optimal layout where "optimal" simply means mechanical from whom computer "fastest". There are a number of factors that have keyboards inherited a suboptimal key layout. Better some impact on speed and comfort: letter solutions have been kept out use by economic of frequencies, key reachability, human hand anatomy, pressure. An arguably optimal layout for the certain regularities found in any text in any is the Dvorak layout. This article language, and one or two others. discusses the possibility of applying the same guidelines and ideas that shaped the Dvorak layout on keyboards used in other languages. Some of the questions answered are how good the Dvorak layout is in EngIish. how good is it in other languages and how a language specific version would look like. Keywords. Dvorak, layout, keyboard, left handed, right handed, frequency, regional, national, language Obviously, not all letters occur with the same frequency: are a good example of very 1. Introduction frequent letters. Knowing that it takes a certain amount of time to reach any key on the keyboard A machine with a complicated system of levers and and knowing that some keys can be reached faster weights called the typewriter was invented around than others an optimization problem arises: how 1870. During the next several decades typists used a should the letters be placed on the keyboard in order typing technique called the Columbus technique to minimize the total time necessary to type a (typing using two index fingers). At the time, typing certain amount of text? Presuming that the reach competitions were organized in order to motivate time for each key is known, placing the letters on typists to improve their typing skills. A Frank . the keyboard could be done by minimizing a McGurrin won one such competition in 1888. using function like this one: a technique he developed himself called (typing using ten fingers). His victory was (layout,lung)= the beginning of modem dactylography and as such (fieq(c,lang)*(c,layout)) ('1 plays a vital part in keyboard layout design. CEILIIT.% The main goal of any typist is to type as fast with as where @eqO is the occurrence probability for few errors as possible. Breakthroughs like character c in a given language and to is the time to McGurrin' ten finger typing made the typists a lot reach c for the given layout. A layout analysis faster and had to be improved in order cannot, however, be based solely on this kind of to facilitate such fast typing. Finally, typists got so time analysis. Even if non statistical factors are fast that typewriters often jammed and got broken. ignored, occurrence frequencies of two- and three Unable to design better machines, engineers letter structures as well as top row - bottom row decided to rearrange the keys so that it would jump frequencies should be taken into account if actually be harder to type and their idea was a more precise results are desired. complete success. The problem is that the same basic layout used a hundred years ago to slow down Human hand anatomy includes a number of factors typists (known as "" by the letters on the left that must be taken into account in order to design a side of the upper key row) is used today on several good layout. It is easier to drum the fingers from the hundred million computers around the world. The little finger to the thumb than the other way around. fundamental computer input device -the keyboard - Also, not all fingers are equally strong or equally seams to leave a lot of room for improvement. fast: their speed and agility decreases from the Student paper 26th Int. Conf. Information Technology Interfaces /TI 2004, June 7-10, 2004, Cavtat, Croatia 37'A thumb to the little finger. Furthermore, it is a lot Most tangible results will come from statistical text easier to type altemating hands than typing series of analysis and because of that the question of sample characters with the same hand because there is a quality is raised. What is "normal" text? Language much higher degree of independence between two evolves relatively fast so the time of writing of a hands than between two fingers. certain text is very important: modem English can hardly be considered equal to the language that A key property of most languages is the alternation Hamlet was written in. Also, a book about e.. between vowels and consonants: a property that prenatal development is likely to have different played the key role in the design of the Dvorak letter frequencies than the daily news. Furthermore, layout. Why is this so important? Because it goes how much text is enough? hand in hand with the fact that typing altemating hands is naturally faster and easier. This means that The problems mentioned so far could be considered vowels can be grouped under one hand and the most preparation problems, problems that arise before frequent consonants under the other to make the any actual measurement. A question that should be basis for a very fast typing mechanism. answered as a result of data analysis is how good the Dvorak keyboard really is in English. The Another detail of the Dvorak layout is the special question will be answered mostly on the basis of attention to index fingers. Because other fingers statistical test results which will also show how normally cover only three keys (taking into account efficient the Dvorak layout is in other languages. only letter keys) and index fingers cover 6 keys What would a language specific layout based on the each, their keys must be a little less frequent to principles guiding the Dvorak layout look like? compensate such a heavy load. That is the main reason that not very frequent letters U and are The analysis results will obviously show a certain placed under the index fingers. level of similarity in Dvorak layout usage in different languages: how strong should the It is worth noting that the Dvorak layout is more a similarities be to justify use of a only slightly set of principles to guide layout design than a modified English Dvorak layout in another certain keyboard. August Dvorak also designed language? Should an important consideration in keyboards for one handed people, both left and right layout design be standardization? How much of an handed, for instance. It is exactly those principles efficiency loss would a certain level of that are going to be used in this article to shape a standardizations justify? language specific keyboard. But before the methodology is explained and results presented and Finally, one of the more problematic questions interpreted, due attention must be paid to problems addressed in this article is the question of language encountered during research preparation and specific letter placement: every language has it's analysis. special characters and on a keyboard of fixed proportions compromises must be made. 3. Problems and questions Statistical. text analysis will shed some light on A number of problems and questions appear when these problems. analysing keyboard layouts and their characteristics. These problems will first be explained and solutions 4. Analysis -methodology and results to most of them will be proposed later on. One of the key factors that make a certain layout To be able to compare layouts, a method of efficient is letter occurrence frequency. Occurrence comparison should be selected. Layout quality is frequencies show how much a certain letter is used closely related to three sets of factors: human hand and in order to calculate those frequencies a small anatomy, distance travelled by fingers during collection of programs have been written. The input normal typing and the nature of language. Of these is a collection of files, the content of each file is factors, the only one suitable for analytical read and absolute or relative letter occurrences in processing is the nature of language: it is possible to each file are printed out. Before results are measure letter frequencies, statistically analyse discussed, a few words must be said about the text those frequencies and make deductions based on samples. A number of sample texts[6][7] have been those results. It is, however, difficult to measure chosen from 19" Century Croatian literature, 20" factors related to hand anatomy and distance Century Croatian literature, some contemporary travelled by fingers can only he roughly estimated. translations to Croatian, 20" Century English It is, for instance, hard to quantify the fact that it is literature, a translation to English, 19' Century easier to drum fingers from the little finger to the German literature and translations to German. thumb than the other way around so it's impact to Although newer novels might have been a better typing speed is hard to estimate. choice, they are mostly under copyright and so they were not chosen. The novel as a literary form has been chosen because of it's size (most samples range from 200000 to 800000 characters) and expression style which is a much more appropriate Language Layout Distance than that of e.g. lyrical expression forms. index The number and size of individual files is greatest Croatian Dvorak 41,18 for Croatian samples and smaller for English and Croatian 57,63 German samples. The purpose of English sample analysis is to illustrate the basic ideas behind the English Dvorak 3 1,54

~ Dvorak layout. German samples analysis results are English 9wem 57,37 used as a reference point for the much more precise results for Croatian: any common property is likely German Dvorak 37,3 1 to show up during such an analysis. Capturing those German 9wem 63,64 similarities allows construction of hypothesis about the underlying principles of language specific adaptation of the Dvorak keyboard in more than one language. Two quantitative criteria are introduced to measure to what degree a certain layout is adequate for use in a given language (or vice versa). The first criterion is a total finger travel distance measure. Total finger movement during typing can be approximated using these simple rules: the distance from a finger to the key under it is considered zero, the distance to a key in a different row or a different column is and

* the distance to all other keys is 2d. Formula 2 calculates a total finger travel distance index for a given vector of letter occurrence frequencies: (layout,language) = (2) c (fie4(e 1 * &t (e)) c€lnre,.~ where freq0 is the measured relative occurrence frequency for letter c, and dis@ is the distance to the character as defined above. To estimate a distance instead of a distance index, d would have to be expressed in real unit (e.g. 1.5 cm), hut since it is only used to compare layouts for a given language, the absolute total distance is of no practical importance. Application of the formula on the described samples gives the results presented in Table 1.

where t, is the time needed to press a key, LochOis the time required to reach a certain key, is the total number of different letters and reach() is a factor that takes individual finger abilities into account. The reach0 function assigns reach difficulty to letters ranging from 8 to 10 from the middle to the little finger, respectively. The index finger is assigned the value 8 rather than 7 because occurrence frequencies (in %) of letters in a given it reaches for 6 instead of 3 keys which means that a language. For example, in an average Croatian text, loss of parallelism occurs (because the same finger the letter A is the most frequent letter. It occurs types the whole word, e.g. "hum" on the qwerty about 92 times in a thousand-letter text block. The keyboard). The assignment of these values is purely most frequent English letter is E and it occurs 97 an educated guess, but should be good enough to times in a thousand-letter text block and so on. illustrate the layout differences and demonstrate the One of the obvious conclusions that can be derived methodology. More precise factors could possibly from the data in the character frequency table is that be calculated using, for instance, Fitt's law [SI. although differences are obvious, most of the first According to that law, movement time is a IO characters (the ones that matter most if typing logarithmic function of distance when target size is speed is the goal) are the same for all three kept constant, and movement time is also a languages which points to the possibility of Dvorak logarithmic function of target size when distance is usage in Croatian and German without too much kept constant. However, Fitt's law predicts key modification. To this end it is also very important reach time in one dimension only. More that language specific letters for Croatian or importantly, the precision of parameters based on German don't show up in the most frequent letters Fitt's law would still be hard to estimate and would table. Such low language specific character still be suboptimal compared to real measurements. occurrence frequencies (under 4% for 5 characters in Croatian altogether) could be a solid enough Results for all three languages and both layouts are reason to merely adapt the existing Dvorak layout presented in table 2. T* denotes the total typing instead of totally rearranging it. This solution time without the time t1 to press a key. If tr is would, however, be suboptimal in several ways: the estimated at at 3 time units (compared to 8-10 to typing speed would be somewhat better than using reach the keys), the differences are a bit smaller the qwerty layout, but not as good as it could be, the than those found using the total distance formula. right hand little finger could be seriously strained An estimate seems to be the best that can be done having to control 6 or 7 keys and possibly others. here because key press time depends highly on key Aside from the just described Dvorak-qwerty hybrid resistance, key press depth (laptop keyboards are, solution, it would be possible to suggest a for instance, a lot more shallow than their standard completely new layout, custom made to fit a desktop equivalents) and so on. The final results specific language. Further analysis of regional estimate an increase in total typing time from 30% language differences and it's influence on layout (Croatian) to over 60% (English) with German still usability and other factors could result in a somewhere in between (50%). definition of a language specific layout based on the Table 2. Total typing time index per a given letter occurrence frequencies and hand alternation. language-layout pair Figure suggests one possible solution for Croatian. Language Layout T* T Croatian Dvorak 355 457 Croatian qwerty 484 586 English Dvorak 2631 341 - English qwerty 478 556 Figure 2. A half-way solution for a ~ Croatian Dvorak layout German Dvorak 318 408 The problem of suggesting a national layout is German qwerty 530 620 partly simplified by using letter keys of the Dvorak layout as a starting point. The basic idea is to sort all the letters in English and in Croatian by their occurrence frequencies and place a Croatian letter in the place of an English letter with the same occurrence frequency index. At the same time, keys should still be limited to the same five keys in order to enable hand alternation during typing. To follow both guidelines as much as possible, vowels are separated from the consonants and then assigned to keys according to their respective occurrence frequency indexes. 377

Table 3. Letter occurrence frequencies in Croatian, English and German

Since Croatian has more letters than English, the The key properties of the Dvorak layout are letter F (the rarest letter in Croatian) wasn't assigned hand alternation, intensive middle row key a key. The same goes for English specific letters , usage and precedence of inner (index and , and .. Inclusion of these letters in the middle fingers) over 'outer ( and little) suggested layout necessarily implies rearrangement fingers. Organising keys so that these three of special keys like -, /, =, [ and others. A measurement of special character frequencies would properties are secured has two important be of use, but a more important problem would be advantages: because of reduced necessary finger choosing alternative letter locations. The qwerty movement typing is faster and less strainous. solution for Croatian suggests combinations Regionalisation of the layout would be a natural including the ANGr modifier: possibly with the U next step. There are, however, a few obstacles along and E keys to reach the square brackets and with C the way. First of all, a wider Dvorak layout and S to reach the curly brackets. Similar acceptance would make the regionalisation effort assignments could be done for several other more important: low interest for the Dvorak symbols and so some space would be cleared for keyboard (especially on the regional level) at this placement of the letters F, Q, W, Y and X. time doesn't really support a lot of research on As the previous solution, the solution in part possibilities of application of the same ideas to presented by figure 2 has some good and bad other languages. From the technical point of view, characteristics. Even thought the exact layout research on the importance of minimum layout proposed here might not be the fastest layout for the compatibility or standardisation and a more precise Croatian language, the basic guidelines could very measure of relative usability for different languages well present a basis for the best layout. While this needs to be done in order to define a relevant was the goal of the whole research, it is arguably layout. Suggested further areas of research should the only good side of the presented layout. The also include an analysis of two- and three letter major drawback of the proposed idea is maximum structures and their influence on typing speed as localisation implying that no compatibility with the well as possibilities of avoiding top row - bottom original Dvorak layout is maintained at all. row finger jumps. Compatibility with the existing standard is an The results and solutions presented in this article important issue but not one whose importance can are partly based on estimates of otherwise hard to be measured easily. measure factors. Measuring these factors would definitely give the results more credibility, but those 5. Conclusion measurements depend on a number of problematic There is a lot of controversy surrounding the factors: individual typist abilities, amount of training with a specific layout, text language, text Dvorak layout. Results shown in this article topic, keyboard physical parameters (key resistance suggest that it is an improvement over the and size, key height ...) and possibly one or two qwerty layout in most respects. It is a result of others. Precise input parameters would have to be research guided by the goal to achieve determined in dozens of categories ("expert maximum speed without taking into account typistlergo keyboarditechnical text", "normal century old problems that occurred at high typistlregular keyboard/technical text", "expert speed typing and so defined the qwerty layout. typistllaptop keyboardhegular text" and so on) 378 rather than one, general category. That is why these [4]Kozierok C.. The PC Guide - Keyboard measurements were not done. layouts Aside from the results themselves, some scientific http://www.pcsuide.com/ref/kb/layout/index.htm value is provided through the presented analysis 12/5/20041 methodology. Given a more precise set of input [5jRohmert W.Forschungsbericht zur parameters (for example, in a separate study with ergonomische Gestaltung von the goal of measuring those input parameters), Schreibmaschinentastaturen. Eggenstein-Leopol; results based on the formulas provided in this article 1982. would instantly provide numbers which could be [6]A collection of publications in Croatian analysed in a fashion similar to the one presented in htto://www.ffzp.hr/infoz/dzs/uopis.htm this article. [2/5/2004] A single, exact number showing just how much one [7]Proks 2.:A collection of publications in Croatian layout is more appropriate than the other is out of http://home.t-online.de/home/Zoran.Proks/ reach at this point, but however approximative the [2/5/2004] results might be, they strongly suggest that the [8]Fitts,P.M. and Peterson,.. Information capacity Dvorak keyboard in its current form is significantly of discrete motor responses. Journal of more efficient than the qwerty standard - the Experimental Psychology, 67, 103-1 12.; 1964. standard in dominant use today. (summary available at http://ei.cs.vt.edu/-cs5724/g1/index.html) [9]Rumelhart, D. E., &Norman, D. A. Simulating a skilled typist: A study of skilled cognitive-motor performance. Cognitive Science, 6, 1-36.; 1982. [IOINorman, D. A. The Dvorak revival: Is it really References worth the cost? Consumer Products Tech Group: the Human Factors Society. 8, No. 3, 5 - 7.; [IIABCD - A Basic Course in Dvorak 1983. http://www. gialiwood.com/ahcd/ 12/5/2004] [I IINorman, D.A. and Draper, S.W. User Centered [2]McNamee D. Dvorak Intemational's FAQ System Design: New Perspectives on Human- http://www.cse.osi.edu/-dylan/dvorak/Dvoraklnt Computer Interaction. Hillsdale, NJ: Lawrence -Lhtml [2/5/2004] Erlbaum Associates.; 1986. [3]Brooks, M. Introducing the Dvorak Keyboard [12]Dvorak, A., Merrick, Dealey and Ford http://www.mwbrooks.com/dvorak/[2/5/2004] Typewriting behaviour American Book CO, 1936.