Links

Input Device Research Links

Bill Buxton has a input device link collection available on the web.

The HCI lab at University of Oregon has made their Fitts paradigm test software available on the web.

William Soukoreff has made the Generalized Fitts' Law Model Builder available for download. This software has been used in some of the pointing device experiments by Soukoreff, MacKenzie, and others, so it is a good idea to study it if detailed understading of the experiments is needed.

Notes on input device work in Windows:
Pointer Ballistics for Windows XP is a good read before starting pointing device experiments under Windows. Windows API documentation does not stress many of the important details enough. For example, for a beginning input device researcher it is difficult to find out that special procedures are needed in order to receive all mouse movements. WM_MOUSEMOVE is no good. I used to use DirectInput and the CPN driver, but both are a little painful. For a modern developer Raw input may be preferrable.

Text Input Links

Scott MacKenzie has done a lot of work in text input. His homepage has a good collection of articles on the subject.

Shumin Zhai keeps a nice collection of his publications online. This includes work on Steering Law, the Atomic keyboard layout and many other things.

Although, most of William Soukoreff's publications can be reached through Prof. MacKenzie's pages, his academics page may be worth visiting because of some othe text entry related stuff he keeps available for download.

Jacob Wobbrock is working on EdgeWrite and other text entry issues.

Ken Perlin has made a Quikwriting Java demo available. Versions for the Palm Pilot and Pocket PC have been made available. Quikwriting is a word-level unistorke system for pen-based user interfaces.

Textware Solutions sells Fitaly keyboard and Instant Text abbreviation engine.

Cirrin constructed in the Georgia Institute of Technology PenDragon project is another word-level unistroke system for pen-based systems.

Palm Computing uses Graffiti in their products. Graffiti is a character-level unistroke alphabet suitable for pen-based interfaces. Graffiti characters are closer to Roman characters than the original Unistrokes by Goldberg and Richardson (1993). Which seem to descend from work by Gabriel Groner at RAND Corporation in 1966. Note that this chain of descendance seems obvious in hindsight, but may not be how the ideas travelled.

Fonix Sells Allegro handwriting recognizer, which in addition to printed and cursive handwriting recognizers has a special unistroke alphabet. The Allegro alphabet is even closer to Roman character shapes than Graffiti.

ART Makes a handwriting recognition engine. It is available both as a consumer product and as a sdk. It is also used in several OEM products.

CIC has another handwriting recognition engine. It too has consumer versions (jot for Palm and Windows-CE) and OEM versions for OEMs.

The Gadgeteer has some light-weight articles on text input and other PDA gadgetry related subjects.

Tegic Licences T9 text input method for OEMs and T9 soft keyboard for Palm Pilot users. T9 uses a database to disambiguate words. T9 is built for the 9-key telephone keyboard, but the idea can be applied to other devices as well.
You might want to check your favorite patent database for similar work by Motorola (US5952942). While you are at it, do a couple of searches for fun. "Keyboard" is a good keyword for bringing up large number of some of the most amazing inventions that you have never heard about.

e-acute Sells Octave, which is a combination of quikwriting-like area-based segmentation and T9-like word completion (I am not making claims on which came out first. The likenesses were formed in my head when I read the description). The e-acute site seems to be down (in June 2002). If anybody knows what happened to them, please let me know.

Dasher is an interesting text input system for 2D pointing devices. They say it works with eye trackers too.

FASTAP is a clever hardware hack for shrinking the keyboard size. Sirisena and Cockburn have tested a prototype. The results are reported in Sirisena's Honours thesis and in a paper in the 2003 British HCI conference.

Some of the reasoning to support the usefullness of thumbscript. Seems very familiar to me. They probably have not heard about my work on MDITIM, but still it is nice to see familiar thoughts to be believed in. It is very unfortunate if the thumbscript patent prevents people from using MDITIM-like systems without licensing thumbscript as I have intentionally published MDITIM to make it unpatentable (and thus free to use for everyone).

HoriKeys is a combination of a tilt sensor and a small number of keys. The idea is that the small keyboard (3x4) is a window to a QWERTY-keyboard. When kept horizontal the keys represent the keys in the middle of the keyboard, when tilted left, the window is over the left end of the keyboard and when tilted right, the keys become the keys of the right end of the keyboard.

Seppo Tiainen is trying to revive the idea of chord keyboards with his GKOS system. The pictures of his vision on how to integrate the keyboard in a mobile device look very nice. It might be fun to run a user study on the system.

OTM Technologies is working on text input solutions based on their optical translation measurement sensors.

EXideas is working on MessagEase which is a modified multi-press system for telephone keypads. It also has a simple mapping for stylus use on small displays.

Eatoni Ergonomics develops and licences disambiguation systems that can give good performance with rather minimal hardware. They have made a number of research and other papers on their systems available.

Vitaly Gnatenko proposes a touch-and-flick kind of an interface to be used with an input device that is a kind of a hybrid between telephone keypad and a touchpad.

Mike Coleman has implemented an interesting system for using joysticks for text entry in X. The system is called Weegie. It is available for download at SourceForge.

The people at Senseboard are working on virtual keyboards. Unfortunately their site tells very little about the technological side of the prototype (Oct. 2004).

For some entertaining reading about the history of the QWERTY keyboard check out The QWERTY Connection

Päivi Majaranta has collected a list of known eye typing systems.

UniTap is a technique for building small keyboards by chording miniature keys. This technique seems similar to the one described in the digitwireless Fastap pages listed above.

Co-operwrite Limited has a product called Mytext. It allows writing with a joystick through motions that resemble handwriting characters.

Canesta Inc develops a keyboard where the layout is projected onto a surface and the usert's finger are tracked using a separate IR emitter and receiver.

Ventris Inc. has a two keys per character mapping called Handyscript for numeric keypads. Their demo works on Windows, but the concept can be applied on many kinds of devices.

Microth Inc. has a site that describes a number of text entry systems including Claviature, InPad and KeyWheel. Trial versions seem to be available for evaluation.