June 25, 2011
Follow me on Twitter

Conference Chairs

It’s been a long time since I’ve linked to Torill Mortensen, but she’s back: AOIR 11 is over and her stint as program chair has been successfully completed. Torill blogs eloquently about the experience.

This conference removed that already elusive concept "spare time," and even cut into the sacred raiding time. Rather than settling to blog, play with photos, write friends or chat, when I picked up the computer I'd start responding to emails or work on some conference detail. I am not whining. You see - I liked that. I got to write a lot of people I have not seen for years, people I have only read (about) and some total strangers who turned out to be extremely interesting scholars. What's not to like!

However, I did get tired. The last weeks before the conference I just had to give up on raiding all together, because with teaching and administration and the conference, I was working 10-12 hour days. Not quite that long weekends, but still, it's not like I followed regulations. And so, as the conference approached I dreaded it more and more, as the workload just kept increasing.

Running a conference is a lot of work, but in computer science this work is indispensable. It’s how things get done, and crucially how we connect isolated facts and projects to people. The results, of course, are what matter. But to understand the dynamics of the field, to know what people are trying to do and what they hope to do next, you’ve got to know more than what’s on the page.

One tension, evident in my recent travels, is that multidisciplinary research can go to so many conferences that it’s becoming hard to encounter work that is not quite right. Instead of appearing as a short paper or a poster that might inspire controversy, correction, and collaboration, the investigators just send the work to another conference until it is eventually accepted and becomes another entry on a bunch of CVs. A related dynamic tends to fill programs with safe, straightforward work that cannot fail to bear results in the 2-3 year window allowed by the European Ph.D.

This is bad for conference programs and also bad for scholarship: in reviewing papers, I’m starting to be tempted to let lapses in scholarship and methodology slide because I want to buy the authors a beer and argue with them, and if we don’t publish the paper they won’t come. (Web Science had 70 posters – 70! – many of which were presented by very senior people and many of which were very fine research. This is terrific, but we need a better way to reward good posters.)