I thought this article was a good pick because it describes two aspects of network theory. One is that it can be useful for a wide range of topics. In this case, network theory is applied to zebra herds. The second aspect is that as useful as network theory can be, it has certain limitations when it comes to studying changing systems.
Dan Rubenstein, an ecologist, noticed that the differences in network structure might explain why one plains zebras were thriving and Grevy’s zebras were dying out. The plains zebras tend to form groups of several mares and babies with one male. The Grevy’s zebras form groups too, but they aren’t as rigid and they break up frequenly. While trying to determine if this could be the key difference, Rubenstein realized network theory wasn’t perfect for his type of data.
The biggest shortcoming of network theory applied to problems like this is that it is hard to properly account for the effect of time on networks. The data on the zebras was gathered over a three month period, but if two zebras interacted just once in that period, they had to be connected. Rubenstein realized that he needed a method to measure the changes in zebra networks, so he contacted Tanya Berger-Wolf, a computer scientist. Obviously, if they can account for the way time affects connections, their work will be applicable to a wide range of topics, not just zebras. This problem and search for a solution is similar to the article posted by Mr. Hartwell about the need for “long data”, or data that changes over time.