The Zeitgeist

Lately, in preparation for any number of projects on my plate now and into 2008, I have been trying to brush up on my data visualization skillz’ and so I’ve been turning to a few books that seem to be the core of the interactive visualization movement that been dominating ‘web2.0′. (You can see what I’ve been reading lately in the left side bar of this blog). My latest read is Visualizing Data by Ben Fry which was just released last week by O’Reilly Media.

Fry is the co-creator of the “Processing” platform, an Open Source, Domain Specific Language (DSL) that he and Casey Reas created to showcase various computer graphics constructs including data visualization techniques. Fry’s new book is really a great read in that is very cleanly laid out and its exposition is logical and flows very nicely even if you are reading it as someone without a deep programming/IT background.

One of the biggest problems with writing a practical book on the art and craft of constructing data visualizations is always trying to find a way that your read can actually do the things you are talking about.

Until “Processing” became available just setting up a graphics platform on which one could experiment was a daunting challenge (and I say this as a programmer with almost 30 years of pretty deep experience); the books that preceded Visualizing Data set the stage by presenting their Processing framework and presented great info on how programmers can explore and learn a lot about practical computer graphics, but didn’t give a lot of guidance on the craft of how to transform “data” into “information” that can make your point through rich visual presentations.

With Processing, which will run on OS X, Linux and even Windows, you can experiment with really advanced concepts without having to spend weeks compilinge software or knowing anything about the guts of graphics drivers. As a language it’s Java-like, but is stripped down to the bare essentials needed for drawing, simple I/O and controlling drawn-objects.

When looking though his and other people’s (like Neoformix by Jeff Clark) visualizations it strikes me, that just like a good UI in general, the more elegant the final presentation of a data visualization, the more intensive the work required to create it; and most of the time that work is not about the graphics itself …it’s all about the data.

I’m about 1/3 through Visualizing Data and I am really impressed with Fry’s ability to demystify the steps needed to take a pile of data and parse it into useful information. He presents good, clear recipes for deciphering what’s important in your data and how to display it so it has both impact, and can be interacted with by your users.

I’d highly recommend Visualizing Data even if you don’t have to create the kinds of visualizations fry describes and demostrates, but even if you’re someone who might need to just betting understand what’s important in data you have to work with.

Comments RSS