April 24th, 2006

At CHI 2006?

Confused by the schedule explorer? Tired of telling your Web browser to Find, Find Again on the long table-of-contents pages full of hyperlinks? Wish you could pull up papers and movies from the proceedings faster? You might want to try a script i wrote:

get it at http://zesty.ca/chi2006

Copy the contents of the conference DVD to your hard disk. Then run this script to rename all the files — from numbered names like "p1181.pdf" or "ea28.pdf", to names containing the title of the paper, the authors, and their affiliations.

If you're on a Mac, you can use the Finder to quickly find what you want. Type "berkeley" in the search box to instantly get a list of all the papers from Berkeley; or type in any title keyword or author name. Movies will come up with their associated papers, too.

If you use Windows, you can search for these renamed files using the file explorer (though it will take a few more clicks).

It's kind of silly that PDF provides a perfectly good way to include title and author information, and Mac OS X already knows how to search for it — but the metadata in the provided files is unusable. Let's hope we get reasonable metadata next year.


Yesterday i got on the BART to get from Berkeley to SFO, took a plane from SFO to Toronto, and changed planes to get to Montréal. I figured i was just going to keep to myself (and mostly sleep) for the whole trip, but i ended up meeting people all along the way.

On BART i sat next to an older woman who was a psychologist. She asked where i was travelling and we ended up having an extended conversation about how computers are difficult and frightening to use, and whether they isolate people or bring them together.

On the plane to Toronto i sat next to a high school student from Barrie who paints with acrylics. Next to us was a fellow from Pixar who was also going to CHI.

On the plane to Montréal i met a young French teacher named Candide. As we waited by the baggage carousel she hopped up and down to wave at her mother and boyfriend, who were meeting her at the airport, and her blonde hair flew everywhere.

Montréal is cold and grey and rainy. Lots of interesting people here at CHI, though.

This is a cute way to start a talk with an overview.

The talk is "Tinkering and Gender in End-User Programmers' Debugging" given by Laura Beckwith.

In the talk, the speaker cited previous evidence that women have a lower sense of self-efficacy when debugging; the low self-efficacy of women leads to low feature acceptance; and that men are more likely to tinker, whereas women are more likely to follow step-by-step instructions. So they added various supportive features to a debugging environment, intending to help women programmers.

But they were surprised by their results. The more women tinkered with a program, the more bugs they fixed. The more men tinkered, the less bugs they fixed. The high-support features helped the women fix bugs, but lowered their self-efficacy.

The men did more repeated tinkering, which harmed their understanding of the task. The women did more exploratory tinkering, which improved their understanding. The explanation they finally discovered was that the women paused to reflect more often than the men, which improved their effectiveness, whereas the men did more counterproductive tinkering.