Last modified: 18 October 2006
No, I’m not into pseudomathematics all of a sudden, but I have just finished working on a short piece that will debut at the upcoming ASTC conference, thanks to the folks from E&S Spitz, who are setting up a fulldome system for the meeting. My motivation was to visualize non-science data in a manner that would initially seem simply aesthetic but then (I hope) allow for some slight recognition of the underlying numerical statement. Below, I offer both playback and interactive versions of the data.
I have a QuickTime of the rough cut available (28MB), which manages to give a general sense of the piece. N.B. that you’re looking at it in “dome master” format, with the circular aperture representing the hemispherical surface of a planetarium dome with the front toward the bottom of the image. Thanks go to Blake Misajet for donating the soundtrack.
To describe briefly what you’re seeing in the movie… We’re flying through data, and the coordinates represent national debt, inflation, and unemployment (as x, y, and z) for each of 116 countries for which I had enough data. As the piece begins, we zoom toward decreasing national debt, with the cluster of most countries lying ahead of us; the explosion icons are scaled by each country’s military expenditure and color-coded by the percentage of GDP spent on the military. The icons cross-fade into lightning bolts, sized by electricity usage and colored by electricity production. As we enter the central core of countries (characterized by low national debt and small- to mid-sized unemployment and inflation), the icons shift once again to read “HIV,” where the size of the characters is proportional to countries’ HIV infection rates and the color is a function of the number of AIDS-related deaths. As we pull away from the main cluster of datapoints in debt-inflation-unemployment space, the icons become individual flags, sized once again by military expenditure. We pull away along the same route from which we initally approached, passing by countries’ flags with increasing national debt, until we fly through the United States flag, symbol of the world’s greatest debtor nation.
Honestly, the above serves only as a first draft of what I see as a somewhat more extended piece. The flight path is incredibly rudimentary—a hyperbola in the x-y plane that keeps the camera pointed at the origin, and I didn’t toy with the aesthetics as much as I wanted. So it goes.
I created the above animation using Partiview, and I provide the general set-up files below, along with the images used to represent the icons. All work was done on my (c. 2003) Apple 12-inch PowerBook.
The flightpath, as noted above, is calculated using a C program that outputs a shell script for creating the movie. The shell script includes a little-known Partiview feature that allows for multiple cameras:
subcam bc0 0 90 0 45 45 45 45 1.0 snapshot t_filename subcam bc1 0 0 0 45 45 45 45 1.0 snapshot f_filename subcam bc2 90 0 0 45 45 45 45 1.0 snapshot l_filename subcam bc3 180 0 0 45 45 45 45 1.0 snapshot b_filename subcam bc4 270 0 0 45 45 45 45 1.0 snapshot r_filename
I then stiched my cube faces together into a dome master using Paul Bourke’s tga2dome program (’cuz I don’t use Windoze and therefore can’t take advantage of the glom program from Spitz.
Because I’m lazy (and was using a kludged flightpath), I also modified the Partiview set-up files to display the flags upside-down relative to the set-up files you can download below. I then rotated the final dome masters by 180° to put most of the action near the front and center of the dome.
I rendered out separate layers for each dataset (military, electricty, HIV, and flags), then I took everything into Adobe AfterEffects for editing, compositing, and adding credits and sound.
All the data comes from the World Factbook, prepared by the United States Central Intelligence Agency. I collated different countries’ information into a single table that became the basis for the data used in the movie above and the Partiview files below.
Partiview runs on MacOS, Linux, and Windows; all three versions are available below. I’m not going to go into a lot of details about running Partiview, so you’ll need to find that information elsewhere. I highly recommend Brian Abbott’s extensive documentation. You can also join the Partiview Googlegroup to get in touch with other users.
Download the data and Partiview executable here:
All data are mapped into the same coordinate space: the coordinates represent national debt, inflation, and unemployment (as x, y, and z, respectively) for each of 116 countries for which I had overlapping data. There are four data groups:
A note on flying… To move around and through the data, hold down your right mouse button (or on a Mac, your single mouse button and the option key) and move the mouse; you will move toward and away from the origin. To rotate around the data, hold down your left mouse button (or single mouse button on a Mac) and move the mouse; you will effectively “orbit” around the origin. Again, for more details on Partiview and how to use it, please see Brian Abbott’s extensive documentation.
I encourage you to play with the data, fly around, and just try to get a sense of how some of the factors are correlated and anti-correlated. For example, note that many of the countries with high HIV infection rates tend to have high inflation and high unemployment; they certainly aren’t spending a lot on the military. N.B. that I’m not trying to suggest causal relations or specific connections by displaying the data in this way. That’s why I call the piece “Numerology.” I just want to provoke an opportunity to look at the data in a different way and to consider the meaning and ramifications.
Email Ryan if you have any questions, comments, concerns, or gripes about this page.