UIST2009: ARC-Pad: Absolute+Relative Cursor Positioning for Large Displays with a Mobile Touchscreen

Authors: David C. McCallum & Pourang Irani from University of Manitoba
Summary
     This paper covers ARC-Pad which is short for Absolute+Relative Cursor pad.  ARC-Pad is a method for controlling the cursor of a large screen display from a mobile phone with a touchscreen.  The authors claim that ARC-PAD has several benefits when compared to other techniques.
    ARC-Pad works by combining absolute and relative cursor positioning methods.  Absolute cursor positioning is when the cursor on the big screen corresponds to the exact location the user touches on the touch screen.  Relative cursor positioning is when the cursor moves a certain distance depending on how far a user moves their finger on the touchscreen.  In previous methods the user had to explicitly switch between relative and absolute modes.  ARC-Pad achieves this by going into absolute mode if the user taps the screen lightly and goes into relative mode if the user drags their finger on the touchscreen.  This saves the user time by not having to explicitly switch between modes.

     The authors also talked about clutching.  Clutching is the amount of time the user drags his finger on the touchscreen.  One thing ARC-Pad does to reduce clutch time is it makes use of Control-Display gain.  Control-Display gain is when finger movements on the touchscreen are amplified resulting in large movements on the big screen.  This will decrease clutch time but it will also decrease accuracy when the Control-Display gain is too high.  Another technique used to reduce clutch time is cursor acceleration.  When the user drags their finger across the touchscreen faster the cursor will travel further on the big screen.
     All in all the ARC-Pad has two main benefits.  One, it reduces cognitive load by not requiring the user to switch between relative and absolute cursor positioning.  Secondly, it reduces occlusion of the screen by not taking up any of the display like other techniques do(radar view). 

Discussion
     This paper was significant because it was a practical solution to improving cursor positioning on displays from mobile touchscreens.  The authors found that ACR-Pad reduced clutch time and the amount of time it took to move the cursor across the screen.  The differences were much more noticeable the longer the distance the user had to move.  However, the authors only mention performing one experiment with ACR-Pad leading me to believe their could still be some faults in their plan.  In the experiment they use a HTC TouchDiamond mobile phone with a 640x480 and a 52" monitor with a 1360x768 display.  Since the display ratios are different, the speed at which the phone could move the cursor had to be adjusted.  This raised a few questions in my mind.
    First, even though the experiment went smoothly with their display ratios, what if the display ratios were much different?  If there was say a business presentation on a large screen with very different dimensions than the phone, would this technique work as smoothly as the others?  I would think so, but I'm sure once the difference in ratios between the screens was so extreme this would eventually present a problem.
    Second, how does the HTC TouchDiamond touchscreen compare to other touchscreens today.  Is it more or less accurate than an iPhone?  If it is a lot more accurate than other phones to the touch of a finger, then absolute cursor positioning could become a problem with other phones.  Also, what percent of the time does the phone believe a quick touch is a drag and vice versa?  Having experience with an iPhone, I know touchscreens sometimes are more difficult to work than intended.
    Future uses of this technology could be widespread.  First, they need to try it with many different devices and monitors.  The authors also mention in the paper this could eventually be used on laptops to control the cursor when it is hooked up to a monitor.  Controlling the mouse on a large monitor from a laptop touch pad can sometimes be very frustrating.  Another future use they mention is they could implement dragging and dropping icons much quicker.  The user could start dragging the icon with their finger then tap the screen once to go into absolute cursor positioning and move the icon anywhere on the screen with the click of a finger.  The would greatly reduce clutch time when dragging icons.