It's been interesting watching visualization (and similarly, intuitive usage of data mining) struggle through what appears to be a very long and very awkward in

It's been interesting watching visualization (and similarly, intuitive usage of data mining) struggle through what appears to be a very long and very awkward infancy stage. Academics and developers continue clinging to things like hyperbolic trees and treemaps as flagship tools for understanding large data sets. Most of these have never seemed particularly useful to me, from the first time I saw a presentation of "cutting edge" visualization techniques in 2001. Similarly, people have this seemingly endless fascination with tag clouds, which are often of dubious utility. This article mentions Wordle, a tag cloud generator that adds typographic design to the equation, and the audience is wowed anew. It's always amazing when excellent, conscientious design meets large datasets, and when the end result promotes understanding and insight. I'm blanking on current examples at the moment, but two chestnuts are from information-laden maps such as the New York City MTA map and Gapminder's world data visualization.