First day of IEEE Visualization 2007

Hi there gang!

I’m spending the week in the Hyatt Regency Sacramento (that’s in California, US and A!) at the IEEE Visualization 2007 conference. I’ll try and write more posts if anything with significant bloggability comes up. Until Monday, there are mostly tutorials and the InfoVis conference and on Tuesday Vis itself starts. The difference between InfoVis and (Sci)Vis is a hotly contested topic about which I’m not even going to try to comment here. However, this is a brilliant topic to discuss in bars with visualisation people, as is the difference between visualisation and visualization.

The InfoVis keynote was given this morning by Matthew Ericson (download his slides here), deputy graphics director of the New York Times. Ericson is responsible for any graphics and illustrations (visualizations, you could call them) used in the paper and online versions of the New York Times. Besides being a beautiful example of visualization used daily by more than a million readers, what struck me was the visualizations also have to be made against serious deadlines. One of the cool examples was their visual analysis of political speeches: transcriptions of these are often available minutes after the speech itself has been made. The NY Times runs this through a bunch of scripts, and creates really cool bubble plots of the terms occurring most often in various politicians’ speeches, also over time. One year, Dick Cheney ended a speech at 23:15, and the paper’s press deadline was at 23:45! Crazy.

Another topic worth a mention is the social visualization site many-eyes.com. You can upload yer dataset, run a number of Java-based infovis techniques on it, and publish the results. Other users can discuss these visualizations, blog-comments style. Go look, it’s pretty slick.