Learning the ropes of Palladio
To be 100% frank, when using Palladio, I ran up against a bit of a wall: while all the nodes and edges and such in the readings on network theory made perfectly good sense to me, I had absolutely no idea how that translated into something that a computer could actually read. As a result, my first efforts looked like this:
Kind of like Koosh balls–neat, but not super informative. My neighbor in class, Sally, kindly informed me that I needed to change the target data set, so I did that, but honestly, I still didn’t really get what I was looking at. (I also learned that I couldn’t upload the .svg files you can export from Palladio as a graphic on WordPress–and this, on the very day I learned how to get rid of all the funky error messages from my iframe embed plugin! WordPress giveth, and WordPress taketh away.)
So I decided to try again with data that was a little more meaningful to me. In Googling “Palladio data,” I came across a dataset that was squarely up my alley: on this tutorial page from the University of British Columbia Libraries, I found a link to this dataset:
Les Miserables: coappearance weighted network of characters in the novel Les Miserables. D. E. Knuth, The Stanford GraphBase: A Platform for Combinatorial Computing, Addison-Wesley, Reading, MA (1993).
Hot dog! This is my jam. I’m interested in tracing character arcs and interactions in medieval fiction, but the principle holds perfectly well for Les Misterables as well. Looking at the dataset, I could see that it mapped out how often characters appeared in the same chapter as other characters, which also gave me a clearer idea of what Palladio wanted when looking for a “source” and a “target.” This, I thought, was a dataset that could really help me see how Palladio could be useful to me.
So here’s what that looked like:
Looking at this, I could see that a few things jumped out, including clusters of characters (most prominently les Amis de l’ABC) who primarily interacted with each other, and interacted with each other a lot. Another thing that stood out were outliers like Cravatte and Mother Plutarch who are only loosely connected to the rest of the characters. Perhaps even more interesting for me were characters who served as linkages between clusters–Eponine, unsurprisingly, was one of these, and of course Javert and Valjean, but I guess I hadn’t really been thinking of Fantine as a linking character, or Gavroche. That could be useful for me in tracking characters’ narrative roles in linking different parts of a text.
Messing around with the settings, I tried out the “size nodes” option. Now this could raise some interesting research questions–are the people who appear a lot in the text the same as the people who connect different groups of characters? In Valjean’s case, seems like yes–in the case of the Amis, seems like no. I’m fascinated by the size of the cluster over by Fantine–I don’t remember those characters at all, but then, it’s been a long time since high school when I read Les Miserables!
In any event, this is something I definitely want to mess around a bit more with. After all, as our readings on visualization and network theory in the class make clear, there’s some intellectual value in playing with and exploring these tools!