Vote: Should I go to the SAC Winter Formal? I ain't much for formal occasions, but on the other hand ODP will be DJing a room with blues and swing and other good dance music.

Anyway: interdisciplinary thinking rocks. I just read a paper titled "Data Crystallization Applied for Designing New Products" by Horie, Maeno, and Ohsawa that describes a data analysis tool that (among other things) algorithmically detects and graphically displays "meaning clusters" and connections between them in text. The on-screen graphics of ideas and their relationships look like little molecules bouncing around.

When they were talking about how they modified the display to make it easier to understand, they described how they were inspired by the "molecular" look to take on a "crystallization"metaphor. Idea clusters are like snow falling from the sky, with a few key words acting like "dust motes," nucleation points that cause other words to cluster around them. In order to make the clusters larger (and thus fewer in number and easier to understand), they decreased the "temperature," making it easier for the ideas to clump. Increasing "temperature" yields a more dynamic and complex picture of thoughts bouncing all over the place.

I realize this is mathematically, computationally, and chemically inaccurate - plenty gets lost in the cross-discipline bridging - but that the essence of ideas can make their way across boundaries like that is pretty awesome. Incidentally, the tool described in the paper just mentioned is great for picking out interesting connections between idea clusters thought to be previously unconnected.

Man, academia is cool.