Keyboards without Keyboards:

A Survey of Virtual Keyboards

Mathias Kölsch Matthew Turk
Dept. of Computer Science Dept. of Computer Science
University of California at University of California at
Santa Barbara, CA Santa Barbara, CA
+1 (805) 893-4321 +1 (805) 893-4321
matz@cs.ucsb.edu mturk@cs.ucsb.edu
ABSTRACT 2. RELATED WORK
Input to small devices is becoming an increasingly crucial factor
A large body of related work exists for physical keyboards and
in development for the ever-more powerful embedded market.
typewriters, their characteristics, usability, efficiency, history, and
Speech input promises to become a feasible alternative to tiny
anthropomorphic backgrounds [18][6][1]. As the keyboard
keypads, yet its limited reliability, robustness, and flexibility
becomes one of two largest components of computing devices
render it unsuitable for certain tasks and/or environments. Various
(next to the display), research on smaller and more mobile text
attempts have been made to provide the common keyboard
entry methods and devices has made great strides (see [15] for in-
metaphor without the physical keyboard, to build “virtual
depth coverage). We are not aware of a survey that focuses on
keyboards”. This promises to leverage our familiarity with the
input interfaces that retain the keyboard metaphor yet remedy the
device without incurring the constraints of the bulky physics.
restrictive device. More related work is referenced from
This paper surveys technologies for alphanumeric input devices throughout the paper.
and methods with a strong focus on touch-typing. We analyze the
characteristics of the keyboard modality and show how they 3. ALPHANUMERIC INPUT
contribute to making it a necessary complement to speech Since language and its manifestation in sentences, words, and
recognition rather than a competitor. letters is the human’s primary means of communication, it has
been researched extensively for the human-to-computer
Keywords interaction as well. In this section, we take a look at three main
Small device input, virtual keyboard, user interface, computer categories: Speech recognition, handwriting recognition and sign
vision, gesture recognition. language. Touch-typing is covered in section 4. Discussed are
characteristics and shortcomings.
1. INTRODUCTION 3.1 Speech Recognition
Touch-typing or machine writing was invented for mechanical Probably the most hailed UI, speech recognition (SR) is now at a
typewriters which had the current QWERTY key layout since stage where it can be successfully deployed in limited domains,
1874. While this interface is come to age, it survived because of such as in call centers for customer service or for voice dialing on
its many positive aspects. Yet it is not feasible for the ever-smaller mobile phones. Under these conditions it provides a highly user-
computing devices that house ever-more advanced functionalities. friendly, unobtrusive, flexible and efficient interface method. But
New alphanumeric interfaces include numeric keypads augmented careful estimates suggest that 100% recognition rates for large
with letters as on the cell phone and the Graffiti handwriting vocabularies will not be possible in the near future. Noisy
characters. environments, speaker particularities (accents, speech
In this paper, we survey the state of the art in alphanumeric input impediments) worsen the situation. One of the strongest
interfaces. After an overview of the related work in the area in arguments however for why we should not solely rely on SR lies
section 2, we lay out the general space of interfaces for text input in how our brain processes speech generation. As Shneiderman
in section 3. In the following section we discuss characteristics summarizes in [26], producing sounds that make up words and
and human factors of touch-typing in particular. Section 5 sentences occupies parts of the brain that are also used for general
explains the criteria we examined with. Section 6 introduces and problem solving. This is contrary to how for example body
compares various touch-typing input methods and devices. We movements are processed: Physical coordination does not conflict
conclude with stating that keyboards – whether virtual or real – with problem solving. In other words, it is harder to think when
are very well suited to the task of alphanumeric input. the thought also has to be spoken aloud. Overall, even the
versatile SR interface has its limitations.
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are 3.2 Handwriting Recognition
not made or distributed for profit or commercial advantage and that Handwriting recognition is another interface that offers itself to
copies bear this notice and the full citation on the first page. To copy HCI due to its widespread mastery. It is equally hard as SR,
otherwise, or republish, to post on servers or to redistribute to lists, especially connected-cursive handwriting. This is the reason for
requires prior specific permission by the authors and/or a fee. slightly modified alphabets like Graffiti [2] or entirely new pen-
UCSB Technical Report 2002-21, July 12, 2002. based scripting languages like Cirrin [16] or Quikwriting [19] that
© 2002 Mathias Kölsch & Matthew Turk. are better suited for recognition by a computer. They achieve
good recognition rates and can be implemented on devices with keyboard is an input device with one degree of freedom (1 DOF)1,
limited capabilities. Just as in writing on plain paper though, the a pen or a mouse has 2 DOF (a 2-dimensional surface), and SL
throughput is limited because a single actuator (the pen led by the operates in 4 DOF (3-dimensional space plus time).
fingers) is performing the communication task. One unit of
An exception which we will say to have 1.5 DOF (time being
information, a character, consists of a combination of straight
the extra half dimension) are buttons that can generate more than
and/or curved lines and/or dots. The amount of information a
one character by pressing them repeatedly, for example those of
human can output through the medium of a pen is limited by
mobile phone keypads. These keyboard types severely limit the
motor skills rather than by cognitive skills. Furthermore, the
achievable throughput, although a study [20] has shown that both
improvement that can be achieved by training of motor
motor and cognitive skills adapt to the common use with only the
capabilities peaks out much quicker than it does for cognitive
two thumbs. Shannon’s statistical methods and prediction [25],
capabilities. Stenography remedies this shortcoming of a single-
frequently called word disambiguation [4] and used for example
pen interface by coding of language into more, and more complex,
in the T9 algorithm [28], can be employed to achieve bandwidths
symbols. Yet stenography requires a significant amount of
of up to 46 wpm [27]. (A more recent theoretical approach [14]
training, too much for the common user (see [3] for an
suggests that raw typing bandwidths up to 60 wpm are possible.)
explanation of the difficulties).
Another way to increase the amount of information each key can
3.3 Sign Language produce is to use complex moded or combinatorial operation,
Sign languages (SL) also code language into more symbols than called chording keyboards. Individual key strokes do not generate
there are letters in the English alphabet, thus a higher bandwidth a character, but only combinations of keys pressed
can be achieved with fewer atomic elements. “Speaking” SL simultaneously. Since n keys map to n+m symbols, chording
naturally does not involve the vocal system, although mouth keyboards also have a higher DOF than one. The maximum
movements frequently accompany hand and arm gestures. But – addressable space is 2n-1 symbols (less one because at least one
without proof – we guess that, contrary to speech processing, key has to be pressed to recognize an action) with n keys.
gesturing SL originates in parts of the brain that do not conflict Similar techniques as word disambiguation can be employed to
with sentence forming. Therefore, it is conceivable that a SL-like even relax the requirement on the strict sequential ordering of key
modality can be developed that remedies the problems of SR and presses by performing context-sensitive reordering. This is
handwriting interfaces. Yet the drawbacks of using SL as input frequently implemented in word processors that for example
modality that must be overcome are big: The skill penetration in correct “hte” to “the”. This of course comes at the expense of
the population is small, extensive movements are required to greater context sensitivity of the input method.
communicate, signals are highly context dependent, and last but
not least there are many difficulties associated with computer Experimental observations suggest that the speed of classic touch-
recognition of sign language. typing with ten fingers is limited by cognitive skills rather than by
the motor skills required to independently and rapidly move the
ten fingers. In other words, the bottleneck for touch-typing
4. TYPING AS INPUT MODALITY bandwidth is the brain. For pen-based interfaces it is the
As derived in the previous section, no alternative input modality
dexterousness of the fingers rather than cognitive skills. This is
is the silver bullet to the UI problem. In this section, we show that
good news: The brain can learn much quicker and better how to
touch-typing, while certainly not “perfect” either, does not have
control body parts than the human physique takes to adapt to new
many of the drawbacks of other methods.
requirements.

4.1 Touch-Typing: Definition Table 1. Bandwidth comparison of different UI methods for

We define touch-typing as any input method that employs discrete communicating alphanumeric data. All values are commonly
sensors, or sensed areas, or buttons, for one or a set of atomic used quantities, aside from the one for sign language which we
symbols (letters, digits, or characters) of a language. Examples are guessed equal to reading prose, based on the ability to simulcast
the common keyboard, the keypad of a mobile phone, and on- American Sign Language for live television.
screen keyboards on PDAs. This definition explicitly includes Bandwidth in words-
“virtual” buttons that only differ from the surrounding physique in UI method
per-minute
that their extent is sensed by some technique for touch by a finger Conversation 280
or pointer. We use the term keyboard, or keyboard metaphor, Reading (prose) 250
interchangeably for touch-typing interfaces. Handwriting (paper) 30
Graffiti 20
4.2 Characteristics and Human Performance Sign language 250
Touch-typing was born with the invention of the mechanical Keyboard touch-typing (average) 50
typewriter, and the common QWERTY layout followed us since Keyboard t-t. (professional) 150
1874. Its greatest benefit is that all ten fingers can be used to
operate it in very rapid sequential order. The interface is not
restricted to communicating through a single pen. A seemingly
related yet orthogonal benefit derives from the fact that each 1
A keyboard discretizes the 1 DOF input gesture into a set of
button generates only one bit of information. So in a sense, a zero-dimensional or binary bits of information. This is an
important capability of keyboards that distinguishes them from
one-dimensional devices like volume sliders for example.
Another advantage of the keyboard metaphor over alternative operating devices. A third alternative are devices that could
methods of text entry is that it supports novice and expert users potentially function as a virtual keyboard, but it has not been used
equally well. Keyboards afford both hunt and peck typing, which yet as such.
is advantageous if the key locations are not memorized, as well as Gear: What devices are employed to realize the VK? Are there
touch-typing with ten fingers for advanced users. alternative methods that could shrink equipment size but can still
produce the same VK method to the user? We also distinguish
4.3 The Influence of the Key(board) Layout two main types of incarnations:
The QWERTY key layout (i.e. which key maps to which language a) Sensing areas are on a flat surface, e.g. a table: A regular key
symbol) is not optimal in a number of ways. Various fixed-layout layout results in high familiarity with the interface method. Finger
keyboards optimize the key arrangement for frequent alternating impacts with the surface produce tactile feedback. These VKs also
use of the left and right hand, short travel distances of fingers, allow for simple ways to visualize the keyboard.
decreasing load from index finger to pinky to compensate the
decreasing strengths of these fingers and so forth. Most frequently b) Sensing areas are on the user’s hand or on a worn glove: No
cited is the Dvorak keyboard, whose key layout supposedly is far surface is needed for this type, and typing can happen in “stealth
superior to QWERTY. Yet after a closer look [18][13][17], mode”, e.g. while the hand is in the pant’s pocket. Tactile
Dvorak layout is only marginally better in terms of maximum feedback is provided implicitly by finger contact. But at least with
typing speed of experienced users. It does, however, fare superior today’s technology a glove is essential to implementation.
in that it is easier and quicker to learn. Other approaches are more Method of key press detection: What does the device actually
promising overall, for example Zhai et al [31] analytically predict register – the touch of a surface, an interruption of a light beam
the performance of various layouts and devise one with 40% etc. This has a big impact on how robust the method is, i.e.
predicted asymptotic performance improvement over QWERTY. whether actions that were not intended to be key presses might be
We will not delve into further discussion here. recognized as such.
To mention are “soft” keyboards which (can) dynamically change Number of discrete keys: How many keys does the method
the key arrangement for different users, situations, or even from and/or device have? Is this number a hard limit or can it be
keystroke to keystroke. Often times they also provide keys that increased easily? This is important if other or expanded alphabets
produce a sequence of characters per key press. This benefit is of like Chinese are to accommodated. We denote a device that has
particular interest to virtual keyboards as they might combine a similar characteristics as a regular keyboard as having 52 keys and
projection method with the input detection technique, thereby a plus “+” symbol stands for expandability. (The common
alleviating the need to switch the focus of attention from display typewriter has 52 keys while modern keyboards usually have 101
device to a physically separate input device. or more keys.)
Keyboard layout modifications (as opposed to key layout which Key-to-symbol mapping DOF: Does each key correspond to
refers to changing the key-to-character mapping) such as the split exactly one symbol/character (1 degree of freedom, DOF), or is it
keyboard on the other hand have been shown to have a very a one-to-many characters mapping (1.5 DOF, see section 4.2),
consistent impact. First proposed by Klockenberg et al in 1926 disambiguated by either temporal methods (multiple successive
[12], split keyboards reduce the physical stress on hands and keystrokes), by statistical prediction or by chording methods
fingers for all users, novice and experienced. (multiple keys pressed simultaneously produce one character).
Temporal significance interval for key press: For the device to
4.4 Virtual Keyboards register a key press, how long does the key have to be pressed?
We define a virtual keyboard as a touch-typing device that does On a physical keyboard, a key does not have to be depressed for a
not have a physical manifestation of the sensing areas. That is, the noticeable amount of time, but this might be different for virtual
sensing area which acts as a button is not per se a button but keyboards. Obviously, this has an immediate impact on the
instead is programmed to act as one. So a sensing area could for potential typing speed with the VK.
example be realized with photo-electric sensors, active finger Number of discrete operators: Are all ten fingers used to operate
tracking methods, or a touch pad. The latter is different from a key the keys of the virtual keyboard or is it only the index finger,
pad as it does not have a-priori designated areas for buttons. thumb etc. that can press a key? The number of discrete operators
Virtual keyboards that employ discrete sensing areas for each influences the parallelism that the human can utilize and therefore
symbol (rather than chording methods) inherently allow for the potential bandwidth achievable with this device or method.
realization of a soft keyboard.
Operator-to-key mapping: Can any operator press any key or
only a subset thereof? There are three different cases: A one-to-
5. METRICS AND CHARACTERISTICS one mapping means there are as many keys as operators, and each
We examined each of the methods and devices, called “virtual operator works with the one key only. A one-to-many mapping is
keyboards”, VK, for a number of characteristics which we exemplified by how most people touch-type on a keyboard: One
consider relevant to a user interface. To simplify the comparison finger is responsible for a set of keys, and the key set for a finger
and future discussion of UIs we introduce a taxonomy specific to does not intersect with any other finger’s set, i.e. one key is
user input interfaces (and VKs in particular). always pressed by the same finger. Finally, a many-to-many
Method or device: Since we did not limit the survey to actual mapping allows any operator to press any key. It must be noted
devices, we have to distinguish between mere suggestions of that many devices theoretically have a many-to-many mapping,
character-producing methods without implementations and actual but individual users tend to operate a specific key with always the
same operator. This is contrasted by the musical piano keyboard, every key as having a unique mapping to a character – a one-to-
where the keys are pressed by whatever operator is nearest by at many key mapping has an additional effect on the final, perceived
any moment in time. (The authors are not aware of any interaction accuracy.
methods that use a many-to-one mapping.)
Operator-key switch time: This is related to the number of 6. VIRTUAL KEYBOARDS: METHODS
discrete operators that can be employed with the VK, but it is an AND DEVICES
independent quantity. It gives an idea about the human factors The following subsections explain the main characteristics of each
aspect of the time between pressing two different keys. A physical VK2. Some criteria from section 5 are addressed exclusively in the
keyboard has a very small switch time, especially when two tabular comparison in subsection 6.13.
different fingers are used: We can hit two keys almost
simultaneously. On a thumb-operated cell phone pad on the other
hand, it might take a more significant amount of time to switch
6.1 Visual Panel
The Visual Panel [32] consists of a camera and a sheet of paper.
from one key to the next. Fitts Law [9] provides a widely accepted
The location of the extended index finger in reference to the paper
method for quantitative analysis. Due to the complexity of this
is located with computer vision means. The primary application is
quantity (it varies for different operators, keys, and operator-key
a mouse pointer, clicking is achieved by resting the fingertip in its
sequences) we give only qualitative guidelines.
current position for three seconds. The authors demonstrated text
Feedback: Does the VK sport a feedback mechanism other than entry by interpreting pointer locations as the keys of a keyboard,
characters appearing on a screen? If so, what human sensors are which were printed on the sheet of paper. An audible notification
used? There might be a time delay between the typing action and signals the recognition of a character after the 3 second
the character and/or feedback being visible (audible…) by the significance interval.
user – how big is this delay? The usability and “feel” of the VK
will strongly depend on good feedback characteristics.
Visual incarnation of a keyboard: Is a projection or display
mechanism available or feasible? With what technology?
Visualization of the “keys” is an important consideration for
novice users.
Familiarity: Is the VK an entirely new method to input text, does
it have a remote relation to conventional keyboards (same layout
but does not support all its affordances), or is it very much in the
style of a physical keyboard? This will have a big impact on how
easily users can transition to this kind of virtual keyboard, and in
our opinion also on the broad acceptance of the VK.
Estimated bandwidth: This can of course be only a rough 6.2 Finger-Joint Gesture Wearable Keypad
guideline to how many characters a human will likely be able to The FJG [11] suggests viewing the phalanges of the fingers
input per time when using this kind of device or method. We state (besides the thumb) of one hand as the keys on phone keypad. The
the bandwidth in characters per minute (cpm). To compare to thumb is used to press the virtual buttons. This is similar to
words per minute (wpm), a conversion factor of 5 can be used in
accordance to the mean word length of 5 characters per word for
the English language. We report raw key stroke numbers, not the
number of characters after word completion or related methods.
Invisibility: How apparent is the device and its use to other
people? Can the VK be used in “stealth” mode, without disturbing
others or even without being noticed?
The next three criteria are very important UI characteristics, yet
limited information about almost all devices and methods
rendered a comparison impossible.
Cost: What are mass production costs of the device? We put this
in relation to the cost of a physical keyboard, which ranges in the
tens of dollars.
Reliability and robustness: We are used to perfect accuracy of Drawing reprinted with permission,
physical keyboards – no key press is registered unless we actually copyright by the authors of [11].
press a key, and no key press goes unnoticed. VKs might have
non-zero false positive and false negative rates, for example
caused by non-human disturbances to the recognition process or 2
Note: While we are confident about the information reported
by borderline key press actions by the human.
below, some of the products’ available specifications are not
Accuracy: Strongly related to this is the accuracy with which the very comprehensive. We inferred data wherever it was fairly
intended key press is recognized. For this measure, we regard obvious and explicitly stated otherwise.
Thumbcode, but it solely relies on word disambiguation to 6.5 FingeRing
produce more than 12 characters. Yet the drawback of this 1.5 FingeRing [10] uses accelerometers on each finger to detect
DOF key-to-symbol mapping might be mitigated by the familiar surface impacts. In the wireless version depicted in the figure
layout. Also, less complex hand configurations might be less below these rings communicate with a wrist-mounted data
tiring for the user. Just as Thumbcode, FJG has no user feedback processing unit. The interaction method is designed for one-
method beyond skin contact sensations. handed use, but could be extended to two hands with obvious
implications. In the current version, the finger movements to
produce one character are extensive: two chording patterns have
6.3 Thumbcode to be typed within a time interval, each consisting of a
The “Thumbcode” method described in [21] defines the touch of combination of fingers hitting the surface. Due to this piano-style
the thumb onto the other fingers’ phalanges of the same hand as typing method, users with prior piano experience fare much better
key strokes. Consequently there are 12 discrete keys (three for with this device; in fact, the full 2-stroke chord mapping is
each index, middle, ring finger and pinky). To produce up to 96 rendered too difficult for novice users.
different symbols, the role between keys and operators is broken
up: The four fingers can touch each other in eight different ways,
each basically representing a mode, or modifier key that affects
the mapping for the thumb touch. Tactile user feedback is implicit
when touching another finger with the thumb. A glove
implementation was tested by the author.

6.6 TouchStream
The TouchStream keyboard stretches our definition of a VK as it
has keys printed on the surface. Yet the underlying technology
permits software configuration of the sensed areas, equal to the
multi-point touchpad described in subsection 6.6. Despite
conventional touch-typing the TouchStream affords a number of
6.4 Chording Glove chording patterns as alternatives to modifier keys. These patterns
The Chording Glove [22] employs pressure sensors for each are pressed by one hand (anywhere on the pad) while the other
finger of the right hand in a glove to implement a chording input touches the key that is to be modified.
device. Almost all possible finger combinations are mapped to
symbols, making it potentially hard to type them. Additional
“mode switches”, located along the index finger, are used to
produce more than the 25 distinct characters. Yet user experiments
suggest otherwise: rates of up to 19 wpm are achieved after ten
training sessions “with no signs of leveling off”.

6.7 Multi-Point Touchpad

DSI Datotech Systems offers one of the few touchpads that
reports up to ten surface contacts and their pressure forces
independently and simultaneously [5]. While it has not been
implemented yet, one could use the 20x15cm large device to
Drawing reprinted with permission,
transfer the traditional keyboard modality in a one-to-one fashion
copyright by the authors of [22].
to an interactive, software-configurable surface. Inherent to this
device are the same user feedback methods as for any of the
devices employing tabletop units: finger surface impacts.
 6.10 VKB Projection
The virtual keyboard technology developed by VKB [30] is a
tabletop unit that projects a laser image of a keyboard on any flat
surface. Infrared cameras detect key strokes of all ten fingers.
Word disambiguation techniques are employed despite this 1
DOF mapping. Therefore, our guess is that engagement of all
distinct key locations is detected, yet with a fairly low accuracy.
These two characteristics in combination should result in fairly
good recognition rates. Surface impact of the fingers serves as
The multi-point touchpad, see subsection 6.7.
typing feedback.
6.8 VType
VType [7] detects the key stroke of each finger “in the air” with a
data glove (fiberoptical curvature detection). Different locations
of the key strokes are not distinguished, only which finger pressed
a key. Instead, disambiguation with standard statistical methods
on the word and sentence level solves the 1.5 DOF mapping
problem. There is currently no feedback mechanism incorporated
into the VType prototype.

6.11 Scurry
Tiny gyroscopes on each finger are the sensing technology in
Samsung’s Scurry [23]. The prototype suggests that these finger
rings communicate with a wrist-mounted unit where the data is
processed. Not much is known about this device, yet our guess is
that finger accelerations and relative positions are detected,
making it possible to distinguish multiple key targets per finger.
We further guess that a surface impact is required to register a key
stroke, also making for the primary sensory feedback to the user.
Little LEDs on the rings potentially provide additional feedback.
6.9 VKey
Virtual Devices Inc. recently announced a combined projection
and recognition VK [29]. Little is known about this device, but
their press release suggests that visual sensors (cameras) detect the
movement of all ten fingers. Just as the VKB device, the VKey
also consists of a tabletop unit and feedback is the tactile
sensation of hitting a surface.
6.12 Senseboard
The Senseboard [24] consists of two rubber pads that slip onto the
user’s hands. Muscle movements in the palm are sensed (with
unspecified, non-invasive means) and translated into key strokes
with pattern recognition methods. All further information
(obtained from the company’s web site) can be found in the
tabular comparison. The only feedback other than characters
appearing on a screen comes from the tactile sensation of hitting
the typing surface with the finger.

6.13 Tabular Comparison

Temporal significance interval for

Invisibility (+ high, o medium)

Visual incarnation of keyboard

Key-to-symbol mapping DOF
Number of discrete operators

Estimated bandwidth in cpm

Operator-key switch time
Number of discrete keys

Operator-key-mapping.
*: potentially n-n
Method or device

Familiarity

key press
Visual Panel d 52+ 1 1 1-n Medium Medium 3s 20 o fixed
Finger-Joint Gesture m 12 1 1.5 1-n High Medium 0 100 + no
Thumbcode d/m 12/96 1+4 1.5 1-n High Low 0 70 + no
Chording Glove d/m 5 5 1.5 1-1 Medium Low 0 80 o no
FingeRing d/m 5 5 1.5 1-1 120ms Low 120ms 120 + no
Touchpad d 52 10 1 n-n Low High 0 250 o fixed possible
TouchStream d 52 10 1 1-1 Low High 0 250 o yes
VType d/m 10 10 1.5 1-1 Low Medium 0 100 o with HMD
VKey d 52+ 10 1 1-n* Low High 0 250 o yes
VKB Projection d 52+ 10 1 1-n* Low High 0 250 o yes
Scurry d 52 10 1 1-n* Low High 0 250 o no
Senseboard d 52+ 10 1 1-n* Low High 0 250 + no

interfaces, or virtual keyboards. We found that the trend goes

7. FUTURE WORK towards retaining the original keyboard metaphor as closely as
We have provided a qualitative analysis of different virtual possible.
keyboards. The logical next step is to conduct user studies to
obtain quantitative measures on the usability and efficiency of Our conclusions are that while the keyboard is often regarded as
these methods and devices. Based on these experiments we will an antique method that is unsuitable to modern computing
try to draw conclusions on how each of the metrics and devices, a number of characteristics are inherent in the way we
characteristics from section 5 influences usability and efficiency, use it that make it preferable over alternative methods. Input
independent from device artifacts. with keyboards is and will be an important user interface
modality for computers for decades to come.
8. CONCLUSIONS
We first gave an overview of the range of input devices and 9. ACKNOWLEDGMENTS
methods for alphanumeric data. We then had a closer look at We would like to thank Andy Beall for helpful references and
touch-typing as input method and highlighted its benefits. This valuable comments.
was followed by a survey of the state of the art of touch-typing
10. REFERENCES [17] Norman, D. A., Fisher, D. Why Alphabetic Keyboards Are
[1] Alden, D.G., Daniels, R. W., Kanarick, A.F. Keyboard Not Easy to Use: Keyboard Layout Doesn’t Much Matter.
Design and Operation: A Review of the Major Issues. The The Journal of the Human Factors Society, vol. 25, 5, pp.
Journal of the Human Factors Society, vol. 14, 4, pp. 275- 509-519, 1982.
293, 1972. [18] Noyes, J. The QWERTY Keyboard: A Review.
[2] Blinkenstorfer, C. H. Graffiti. Pen Computing, 30–31, Jan. International Journal of Man-Machine Studies, 18, 265-
1995. 281, 1983.

[3] Daniel, K. W. Murphy Was a Captioner. [19] Perlin, K. Quikwriting: Continous Stylus-based Text Entry.
http://www.robson.org/gary/captioning/murphy.html ACM Symposium on User Interface Software and
Techology 1998.
[4] Davis, J. R. Let Your Fingers do the Spelling: Implicit
disambiguation of words spelled with the telephone [20] Plant, S. On The Mobile. Motorola Report, 2001.
keypad. Avios Journal, 9, 57-66, March 1991. http://www.motorola.com/mot/documents/0,1028,297,00.d
oc
[5] DSI Datotech Systems Inc.: Multi-Point Touchpad;
http://www.dato.com [21] Pratt, V. R., Thumbcode: A Device-Independent Digital
Sign Language. http://boole.stanford.edu/thumbcode
[6] Dvorak, A. There is a Better Typewriter Keyboard.
National Business Education Quarterly, vol. 11, pp. 51-58, [22] Rosenberg, R.: Computing without Mice and Keyboards:
1943. Text and Graphic Input Devices for Mobile Computing;
PhD Thesis, Dept. of Computer Science, University
[7] Evans, F. and Skiena, S. and Varshney, A.; VType: College, London, 1998.
Entering Text in a Virtual World, submitted to
International Journal of Human-Computer Studies [23] Samsung Scurry; PCWorld article at
http://www.pcworld.com/news/article/0,aid,70568,00.asp
[8] FingerWorks, Inc. TouchStream Stealth.
http://www.fingerworks.com [24] Senseboard, http://www.senseboard.com
[9] Fitts, P. M. The information capacity of the human motor [25] Shannon, C. E. Prediction and Entropy of Printed English.
system in controlling the amplitude of movement. In Bell System Technical Journal, 30, 51-64.
Journal of Experimental Psychology, vol. 47, pp. 381-391. [26] Shneiderman, B. The Limits of Speech Recognition.
1954. Commun. ACM, 43, 9, 63-65, Sept 2000.
[10] Fukumoto, M. and Tonomura, Y.: "Body Coupled [27] Silfverberg, M., MacKenzie, I. S., and Korhonen, P.
FingeRing": Wireless Wearable Keyboard; ACM CHI Predicting text entry speeds on mobile phones. Proceedings
1997. of the ACM Conference on Human Factors in Computing
[11] Goldstein, M. and Chincholle, D.; Finger-Joint Gesture Systems, pp. 9-16, 2000.
Wearable Keypad; in Second Workshop on Human [28] Tegic Communications. T9 Text Entry. http://www.t9.com
Computer Interaction with Mobile Devices, 1999.
[29] Virtual Devices, Inc., VKey;
[12] Klockenberg et al. Rationalisierung der Schreibmaschine http://www.virtualdevices.net/press_release.htm
und ihrer Bedienung. Springer Verlag, Berlin, 1926.
[30] VKB Ltd., http://vkb.co.il
[13] Liebowitz, S. J. and Margolis, S. E. The Fable of the Keys.
Journal of Law & Economics vol. 33, April 1990. [31] Zhai, S., Hunter, M. and Smith, B.A. The Metropolis
Keyboard – An Exploration of Quantitative Techniques for
[14] MacKenzie, I. S., and Soukoreff, R. W. A model of two- Virtual Keyboard Design. UIST 2000.
thumb text entry. Proceedings of Graphics Interface 2002,
pp. 117-124. [32] Zhang, Z., Wu, Y., Shan, Y., and Shafer, S. Visual Panel:
Virtual Mouse, Keyboard and 3D Controller with an
[15] MacKenzie, I.S., Soukoreff, R.W. Text Entry for Mobile Ordinary Piece of Paper. ACM Workshop on Perceptive
Computing: Models and Methods, Theory and Practice. User Interfaces, Nov. 2001.
Human-Computer Interaction, 2002. To appear.
http://www.yorku.ca/mack/hci3.html
[16] Mankoff, J. and Abowd, G. D. A word-level unistroke
keyboard for pen input. ACM Symposium on User
Interface Software and Techology 1998.