Paul Butler, a data infrastructure engineering intern at Facebook, was searching for a way to visualize the international relations being enacted on Facebook. With access to Apache Hive, the Facebook archive, Butler began by sampling relations to plot the relative strength of relations between cities around the world. The map he eventually produced is fascinating.
I defined weights for each pair of cities as a function of the Euclidean distance between them and the number of friends between them. Then I plotted lines between the pairs by weight, so that pairs of cities with the most friendships between them were drawn on top of the others. I used a color ramp from black to blue to white, with each line’s color depending on its weight. I also transformed some of the lines to wrap around the image, rather than spanning more than halfway around the world.
After a few minutes of rendering, the new plot appeared, and I was a bit taken aback by what I saw. The blob had turned into a surprisingly detailed map of the world. Not only were continents visible, certain international borders were apparent as well. What really struck me, though, was knowing that the lines didn’t represent coasts or rivers or political borders, but real human relationships. Each line might represent a friendship made while travelling, a family member abroad, or an old college friend pulled away by the various forces of life.
Later I replaced the lines with great circle arcs, which are the shortest routes between two points on the Earth. Because the Earth is a sphere, these are often not straight lines on the projection.
The map is fascinating on a number of levels. The presence of some regions in high densities of ‘friendships’ is not surprising (the US, the UK, France, Germany, Italy…). But I found the ‘dark spots’ in some places (Russia, China, Japan! — compare Japan and New Zealand, for example) and the ‘white rims’ running through certain parts of the developing world (Java, the Philippines, coastal areas of Venezuela and Colombia, parts of Chile and Argentina).
Some of this inconsistency is likely a result of simple patterns of urbanization; not only would Internet users (and Facebook users) be disproportionately gathered there, but Butler’s method seems to rely upon city-to-city weighted relations, possibly exacerbating what would already be a likely rural-urban disparity. So the first thing I’d love to see would be to run this data against simple population density maps, to see if there are disparities.
But even given this possible source of anomaly, some countries do seem to stand out even given their population densities (or be surprisingly dim). Butler suggests that borders may have something to do with it, likely affecting freedom of access in China, for example. But the pattern may also arise from high levels of out-migration, for example in the Philippines, Caribbean, North Africa, and Central America. I’d be intrigued to see if out-migration predicted what we might call ‘Facebook salience’ (which, remember, is tilted to highlighting international, inter-urban relations in Butler’s work).
Fascinating map though, and a particularly interesting way to think about the emerging virtual landscape in a social networking site in relation to geography.