Lab 5 Reflection
Just when I think I’ve got it, it seems I don’t. I am so thankful for Dr. Thomas’s help with the beginning of this lab. When I tried to submit my own csv. files into Gephi, my results were disappointing. I worked so hard on my excel spreadsheets with my data—to make sure everything listed was exactly right and for some reason, my graphs for my data did not resemble the graphs we constructed before. Hardships for Laurie’s journey into Digital Humanities.
In the graphs I constructed before with Gephi, the visualization was measureable. You could see how the nodes connected and those that were most heavily weighted. Graph 4 was the most interesting because it seemed to cover such a vast space. I believe this one was the Geographical layout. It is amazing that using data, the computer program can make all these connections and we can analyze the result to our benefit—for example, maybe seeing which characters interact the most in a single scene. The graph that measured centrality, which measures the shortest paths between nodes, is like a network of connections that would not otherwise be able to be measured. I don’t think this type of measurement could be graphed by hand.
This lab seemed to be the most relevant to making connections in literature, however, you need to be educated in technology to gather benefits. Doing the handwritten graphs and analyses really helped in my understanding, be it a very simple one. When I did upload my csv. files of data from “Why I Live at the P.O.”, the first graph looked very similar to my drawing—a contorted shape of lines that had nodes which connected simply. However, making these connections requires a lot of data input, just as coding into a corpus. I am not convinced that time invested into entering in pages and pages of data and interactions through characters or places or incidences will be justified by being able to get a visualization of what appears to be happening. And I understand that Gephi has many options and attributes that educated users can apply to make the data even more useable; however, a student like me is challenged in using the program to assess this data and its output. Knowledge of the terms used to navigate through the Gephi system was helpful, but the idea still seems very large and unreachable for me.
I am curious to know how literary scholars could tangibly use this in comparing works to each other. Is this tool only used in working with one text or is their overlap possibilities in combining several texts to make connections? For example, using my data from “Why I Live at the P.O.”, all of the women do most of the talking or dialoguing. Could you technically compare it to other southern short stories to see if this is, in fact, a pattern in southern dialogue? Or could you measure who asked the most questions in a short story—like who seems the most curious character. Maybe I’m reaching a bit. I am only trying to find examples in which a program like Gephi, when used appropriately, could result in some interesting feedback in various genres for scholarly use.