Mobile Technology

Researchers create an app that makes you tilt and roll to type

View 3 Images
This tilt-typing mechanism solves the problem of entering text on small or large screens
This tilt-typing mechanism solves the problem of entering text on small or large screens
Researchers say you can get up to a speed of 32 words-per-minute with just 90 minutes' practice
A keyboard app incorporating SWiM is on the way
View gallery - 3 images

Scientists from the University of St Andrews in Scotland have developed a new text input system called SWiM, or Shape Writing in Motion – and it's designed to make typing on the smallest and largest touchscreen devices that little bit easier. Whether anyone is willing to put in the time to learn it, though, will be another question.

Rather than pecking away at a keyboard, SWiM lets you roll a cursor around the screen to select letters (just like the classic game of Labyrinth), which doesn't look the easiest method to master.

It's certainly something different though, and SWiM's developers say it can help as a backup to normal keyboards if you've only got one hand free and are tapping out a message on a tablet or a wearable.

Bearing in mind both Android and iOS already have built-in options and third-party apps for making it easier to type with one hand (usually by shrinking the keyboard to one side of the screen), we're not sure this is any more intuitive for most users – but it might help on the biggest or smallest of screens if you're willing to put in the practice.

Until we give it a go ourselves we're not sure the speed benefit would outweigh the awkwardness of rolling a pointer around, but the researchers say their test subjects were hitting typing speeds of 15 words per minute (wpm) almost straight away and 32 wpm after around 90 minutes of practice.

SWiM is built on the existing idea of gesture keyboards that use swipes rather than taps, but you rotate the device instead of your hand, and tap to start new words. Combined with some smart autocorrect suggestions, it could have promise.

Eventually, the team behind the tech think it might make its way to game controllers and VR as well as mobile devices, but we'll leave it to you to decide whether trying to carefully balance a keyboard one way then another is any easier than pointing at letters in virtual reality air.

"The design of both wearable and large mobile devices assumes both hands are free, however one-handed use of devices, regardless of their size, is now commonplace," says Professor Aaron Quigley, Chair of Human Computer Interaction at the University of St Andrews.

If that name sounds familiar it's because Quigley and his team are also working on teaching computers to recognize objects from drawings, as we reported last year.

A keyboard app incorporating SWiM is on the way

Quigley points out that mobile screens are getting bigger, but we're still often using them with one hand, with the other hand busy holding a coffee, an umbrella, or a grab rail.

"As more and more computation is woven into the fabric of our lives, fluid interaction with such systems and devices requires us to examine our thinking about such interaction and challenge our basic assumptions," adds Quigley.

The software is controlled by the user's dominant hand, which is able to produce the most precise movements, according to the SWiM team.

For testing purposes, the team used a 5.7-inch Samsung Galaxy Note 3 device, though with both Android and iOS now accepting third-party keyboards, getting SWiM up and running on any smartphone shouldn't be too difficult – not as difficult as convincing large numbers of users to give it a try, anyway.

An application is indeed in the pipeline, which will be integrated into a gesture keyboard, but first the researchers are publishing a paper on their innovative typing technique at the ACM CHI 2017 conference.

You can see SWiM in action in the video below.

Source: University of St Andrews

View gallery - 3 images
  • Facebook
  • Twitter
  • Flipboard
  • LinkedIn
1 comment
vbillc
I agree it sounds tricky especially for those with large (or awkward) hands who would probably have the most need of this. However using a similar system to "roll around" adjacent letters when you hit the wrong one might be more useable - I always seem to hit an incorrect letter a number of times before I manage to get the correct one (often hindered by dumb autocomplete interventions).