This paper talks about an interface technique that allows users to perform operations by spatially accessing predefined location in the air. The virtual shelves are located on the theta and phi planes in front of the user and different regions imply different interactions (e.g. a button, a shortcut, etc.). Two studies were conducted – the first one being measuring the general kinesthetic accuracy of one locating a region on the virtual shelves and the other one being a proof-of-concept prototype on a cell phone with which participants were asked to operate by orienting the phone and reaching certain spatial locations. Results show that users can perform such spatial operation quite accurately and moreover, faster in tasks than conventional interaction (e.g. pressing buttons).
Limitations of mobile devices (due to physical properties)
- Operations require many key presses or screen touches;
- Eyes-free interactions are usually unavailable.
- Further readings:
- Why would virtual shelves work? A simply explanation is: spatial memory works. But it seems that there is still differences between, say, a driver knowing where the radio button is and a participant remembering (1 o’clock, 10 o’clock) is the recent calls button. One of such differences is the degree of being eyes-free. In virtual shelves it is almost 100% eyes-free as the authors assume spatial memory requires almost no seeing. However, I doubt this assumption. Try typing in the dark and you will find the error rate is higher. And I am sure even an experienced driver will miss the radio button sometimes if there is no light at all in the car. Maybe eyes-free should mean eyes-peripheral other than being totally free. Maybe BM should be able to compensate such shortages.