Getting Stuff In
Getting stuff into computers has always been more difficult than getting things out. Although there are several different ways to input text, the venerable QWERTY keyboard still is the fastest widespread way to do it (it’s claimed that the Dvorak keyboard layout allows faster typing speeds, but it never really caught on). Now that everything is becoming mobile and small, a new challenge has appeared – how to input text at the same speed and accuracy as a full sized keyboard on a mobile phone or PDA. We already have the multi-tap and T9 modes on mobile phones, and handwriting recognition, on-screen keyboards, or mini-keyboards on PDAs, but none of these approach the speed and accuracy of touch-typing on a full sized keyboard. The latest The Economist’s Technology Quarterly has an article on new approaches to text input (a subscription is needed to read it). Here’s a summary:
- Quikwriting – a New York University’s Centre for Advanced Technology project by Ken Perlin, who began working on the problem in 1997. It uses a stylus based system in which the words are written using a flower shaped area, running the stylus from the center to the “petals” and back again. Microsoft has licensed the technology, and is developing it for several platforms (under the name XNav), using prototypes that instead of a stylus use a flower shaped button.
- ShapeWriter – IBM is also developing a similar system, but using a hexagonal grid, for use on tablet PCs. It uses a pattern recogntion system, so the system is more tolerant of user inacuracy. The software can be downloaded here if you have a Windows tablet PC or a Wacom style tablet.
- EQx series – Eatoni, instead of moving away from the traditional QWERTY keyboard, is moving towards it. Basically it’s the same thing as the familiar T9 system, but instead of having an alphabetic sequence to the keyboard (2 – ABC, 3 – DEF, etc.), it maps a regular QWERTY layout on the keyboard space, be it a 3×4 phone keyboard, or a 6×4 bigger keyboard. The advantage, besides the familiarity of the layout, is that the system has less “collisions” between words when compared to T9, ie, a smaller number of words that share the exact same keystroke sequences. Which means more accuracy, and less input from the user. Eatoni claims that on mobile phones, T9 has an average collision rate of 1 in 27 words, and the EQ3 system has only 1 in 85 words. When used with a 6×4 keyboard, the collision rate drops to 1 in 1,800.
Although the more conceptual systems seem more elegant, the Eatoni concept looks to me the one with the most immediate future in terms of familiarity, a flat learning curve, and an almost zero cost in current hardware implementation (just changing the T9 software and the keyboard layouts).
