Touch Usability

"Bar of soap" uses touch sensors to recognize how you hold it

The "bar of soap" is a research project by Brandon Taylor and Michael Bove of MIT and was presented as a work-in-progress at CHI 2008.

Their idea was to make a handheld device that could recognize the manner in which it is being held so that it can enter the right mode of operation.  They built a prototype that, for a single user, was 95% successful at recognizing five different types of grasp (pictured below) -- remote control, PDA, camera, game controller and phone.

This prototype used 24 capacitive sensors on its faces and a machine learning algorithm for training.  They've built a subsequent prototype that has 72 capacitive sensors and an accelerometer.

You can read more about it at their project page.  You can download the paper here if you've got an ACM membership: The bar of soap: a grasp recognition system implemented in a multifunctional handheld device.

This was reported on a few gadget sites back in December as well.

Rubbing as a Zoom Gesture

At Alex Olwal's web page you can watch video and download a CHI 2008 paper describing research on rubbing as a gesture for zooming.  Rubbing up and to the right zooms in, up and to the left zooms out.

Their evaluation showed that it was better than other single-finger methods for zooming, though fatigue can be a problem if the screen surface isn't smooth.

They also tested two variations on combined two-handed tapping and zooming techniques (e.g. one hand points or zooms and the other can tap to select).

Their focus was on touchscreens that report only a single finger for applications like public kiosks.

The paper is "Rubbing and Tapping for Precise and Rapid Selection on Touch-Screen Displays" by Alex Olwal, Steven Feiner, and Sanna Heyman.

User interface history at CHI

At CHI 2008, Anker Helms Jørgensen and Brad Myers organized a SIG meeting on user interface history, with the goal of launching a concerted effort towards creating a history of user interfaces and human-computer interaction.

The organizers made the point that efforts at HCI history so far, while very useful, have been sporadic and done by HCI insiders (e.g. Ron Baecker, Jonathan Grudin).  As the field matures it could profit by engaging with historians and related groups, such as the Society for the History of Technology.  Another common observation was that HCI history so far tends to focus on a few highlights (PARC, GUI, mouse, sketchpad) and neglects a lot of the variety that used to exist.

Slides, a short bibliography, and other materials from the SIG are up at a new UI History blog created by Anker Helms Jørgensen.

I attended the SIG and found it to be a really interesting discussion and effort.  There was also an HCI history course at CHI, taught by Jonathan Grudin, but I didn't manage to attend this.

Related links (pulled from Anker and Brad's slides):

• UIST conference 20-year history archive by Michel Beaudouin-Lafon and Wendy Mackay
• The Interaction Museum
• Open Video Project (University of Maryland)
  
• Computer History Museum (in Palo Alto, CA)
  

Mobile Computing as a Discipline: CMU's new Mobility Research Center

From Government Computer News:

It’s a sign of having arrived when a new technology gets its own academic discipline. It happened more than 25 years ago for computer science. And some schools launched certificate programs in geospatial studies a couple of years ago.

Now Carnegie Mellon University is launching a new Mobility Research Center at its Silicon Valley site that will be a locus of research and teaching focused on mobile computing.

“There are billions of cell phone users all around the world, and their introduction to the use of computation and the Internet is going to be through use of this handheld platform, not through their desktop or laptop computers,” James Morris, professor of computer science and dean of Carnegie Mellon West, told GCN.

“The United States needs to have that perspective as we look at a global market for computing devices on the Internet.”

The multidisciplinary program will focus on context-aware applications and services, serendipitous collaboration and rich semantic information to enable novel data and media management, visualization and access.

“We have probably 30 faculty members who work in various areas — anything from antenna design [to] anthropology and psychology — and we’re getting a lot of these people together into teams to perform research to look at the way people are going to use mobile devices in the future,” Morris said.

Link: Mobile computing gets academic,

via ACM TechNews.

CMU page: CyLab Mobility Research Center.

Video of Eee Multitouch Functions

YouTube video, posted by touchpad maker Elantech, shows swipe, pinch, rotate, two-finger scroll in various applications:

Via Mac Rumors, Internet Siao, and Google Alerts.

UniGest: Text Entry using Wii Gestures

"UniGest" is a proposed scheme for text entry using a Wiimote.  The user holds a button while making one or two primitive motions with the controller.  It was presented as a work-in-progress paper at CHI by Steven Castelluci and Scott MacKenzie of York University.

They have constructed an alphabet and performed an initial user study to measure the time taken to perform the basic gesture primitives.  From this they derived a theoretical upper bound on text-entry speed of 28 words per minute.  Real performance is likely to be slower, but probably better than previous efforts at text entry using game controllers that achieved around 7 words per minute.

The paper is "UniGest: Text Entry Using Three Degrees of Motion" by Steven J. Castellucci and I. Scott MacKenzie.

You can download the paper from Steven Castelluci's web page.

TWEND: Twisting and Bending as Input Gestures

"TWEND" is a student research competition entry (and runner-up winner) from CHI 2008 by Gero Herkenrath, Thorsten Karrer and Jan Borchers of RWTH Aachen University in Germany.

They embedded optical bend sensors into foam and plastic prototypes.  Users can bend the object to perform actions like scrolling, zooming, and page flipping.  They see this as being most useful for mobile devices -- sort of like Microsoft's force sensors, but in this case you can really bend the device.  This would make for a pretty slick soft e-book reader.

Their project page has more pictures and details: TWEND: Twisting and Bending as new Interaction Gesture in Mobile Devices.

Microsoft Twist Control

The BBC's dot.life blog reports on work at Microsoft Research to use force sensors as input on handheld devices.  Excerpt:

Forget multi-touch, or tilt control accelerometers... Microsoft researchers have been working on force sensing technology that would let you bend/twist/stretch/squeeze your handheld device in order to control it.

The research has been carried out at Microsoft's Cambridge lab by James Scott, Lorna Brown and Mike Molloy. You can read their research paper here (pdf). All the images in this post come from the research paper.

The technology allows users to apply force to their portable device in order to carry out on-screen actions, such as flip a page in a document or switching between applications.

Link: May the force be with you,
via Engadget.

Video Navigation by Direct Manipulation

At CHI there were two papers demonstrating this new navigation technique.  The systems pre-process your video to isolate objects and compute optical flow.  The user can then navigate by dragging objects directly along their paths.  In usability studies the technique outperforms the typical navigation slider.

The first paper is "Video Browsing by Direct Manipulation" by Pierre Dragicevic, INRIA & University of Toronto and Gonzalo Ramos, Jacobo Bibliowicz, Derek Nowrouzezahrai, Ravin Balakrishnan, Karan Singh, University of Toronto.  Project webpage.

The second paper is "DRAGON: A Direct Manipulation Interface for Frame-Accurate In-Scene Video Navigation" by Thorsten Karrer, Malte Weiss, RWTH Aachen University, Eric Lee, Apple Inc., Jan Borchers, RWTH Aachen University.  Project webpage.

This doesn't have anything directly to do with touch, but it's a feature that would be pretty handy on touchscreen mobiles.  Navigation through long videos or podcasts is one of the few usability failures on the iPhone/iPod Touch, in my humble opinion -- you can't manipulate a slider very precisely with fingers.

Reviews of Asus Eee 900 with Multitouch

I'm a bit late on this... Asus's new version of the popular Eee mini-notebook is out in the UK, sporting a slightly larger touchpad with multitouch functionality, which they're calling "FingerGlide."  It has pinch and Apple-style two-finger scrolling.  Engadget has a roundup of the reviews.

See also: Asus press release.