Monday, December 26, 2005

Software Notebook: Designing people-friendly systems

At the University of Toronto in the 1970s, Bill Buxton made a point of choosing a very technical name for his computer science initiative -- the Structured Sound Synthesis Project.

"It was getting funded from science money, so we couldn't call it a music synthesizer," he explained, laughing. "It was a really awkward way of saying music. But it sounded very scientific."

Given the outcome, the people who funded the project would probably forgive him. The creation of that early digital synthesizer turned Buxton from a career in music to one in human-computer interaction. He became a leading expert in that field, working with Silicon Graphics, the Xerox Palo Alto Research Center and others.

Now, he'll be applying his expertise to projects at Microsoft, where he was named a senior researcher last week. The company, which has lost some key talent to Google and others during the past year, trumpeted the hiring of Buxton as an important addition to its research unit.

Buxton said he had opportunities to work for other companies -- "some of the obvious candidates" -- but decided that Microsoft was the right place.

Among other things, the breadth of Microsoft's research efforts suits Buxton's expertise. His work over the years has dealt with a variety of novel and intuitive ways of using computers.

For example, one of his projects, a two-handed method of controlling a PC, arose from his frustration at needing to move a mouse back and forth repeatedly between the text of a document and the scroll bars to the side.

As part of the project, Buxton and a student equipped a computer with a traditional mouse on one side and a touch pad on the other -- letting users move a finger across the touch pad to scroll the document vertically, horizontally and diagonally, while using the mouse in the other hand to point and click.

Buxton is still hoping the technique catches on more broadly. He called the common one-handed mousing approach an example of how modern computer design often falls short.

"My litmus test for what's good design or bad is how well the technology reflects me as a human," he said, citing motor-sensory, cognitive and social skills in particular.


Post a Comment

<< Home