Paper Reading #12: Disappearing mobile devices

Comments
 Shena Hoffman - http://csce436-hoffmann.blogspot.com/2011/02/paper-reading-12-d-macs-building-multi.html
Joshua Penick - http://jip-tamuchi-spring2011.blogspot.com/2011/02/paper-reading-12-detecting-and.html

Reference Information
Title: Disappearing mobile devices
Authors: Tao Ni, Patrick Baudisch    
Presentation Venue: UIST 2010: 22nd annual ACM symposium on User interface software and technology; October 4-7, 2009; Victoria, British Columbia, Canada

Summary
In this paper the researchers explore the evolution of the mobile phone, focusing on how it has gotten smaller and smaller over the years. They then discuss how the mobile phone will one day be small enough to integrate into surfaces or on our skin once hardware linked to human constraints (finger size, eye sight, etc.) has been removed from the equation.

Focusing on motion and touch, the researchers explore technologies that would make it possible to reduce the size of the mobile phone down to its smallest. The devices they discuss are a motion scanner, touch scanner and a direction scanner (a scanner that reacts to gestures). They explain how the devices would use Morse code, gestures and the path of gestures as the language they would interpret. 

They explore the technique of marking, a language of eight directional gestures, in a user study. The users could gesture to the left, up, down, up and to the left, etc. In their study they found that the error rates were below 5%. They also found that making the ‘left’ gesture was one of the more difficult gestures because the sensor would pick up their gesture even after they finished it. To compensate for this the users had to pull their hand out of the sensor’s range or twist their hand around to get it out of the way.

They also explored text entry on motion scanners using two different interfaces, the Graffiti condition and the EdgeWrite condition. Depending on which condition the users used, they had to gesture the letters (like writing in the air) differently in front of the sensor. The users operated the device with their whole hand and wore a glove to improve tracking. EdgeWrite proved to be a good option for a small mobile device, but Graffiti had error rates as high as 66% due to its relying on the relative position of the hand at the start.

Note: In both studies, the users went through a training session first.

Image taken from the paper

Discussion
When reading the abstract and seeing the art of a little device on someone’s wrist, I laughed. While I’m not sure I would ever want my mobile phone to be that small (I just know I would lose it), the ideas they explored in this paper were interesting and unique. I especially enjoyed the part about motion scanners and how the users wrote letters through gestures. I think I would prefer that to texting if one day it’s an option in mobile phones.

As mentioned in the paper, the biggest area of future work will be to study how to incorporate visual output from something as small as what they envision here. Their main focus in this paper was inputting information into the device.