Lately I have been thinking about what big data could mean form my own work. This was sparked by a document I found while doing some archival work a month ago. As I have mentioned previously my Major Research Essay is about the development of a specific boulevard in the heart of downtown Philadelphia, the Benjamin Franklin Parkway. While looking at maps and plans for the parkway a helpful archivist offered me a digital copy of an old Works Progress Administration repot commissioned about the Parkway. In the ledger the addresses, owners and damages awarded to anyone owning property within the area cleared for the Parkway are listed. Although the document is only around 130 pages in all, I initially found it interesting but assumed it would be a little to time consuming to really use for my project. The types of questions and relations that I assumed the data could reveal were not necessarily the same ones I had conceived of and set out to answer in my own research.
After returning from my trip I quickly got back into the day to day of regular course work and the WPA ledger shifted to the back of my mind. Things may have stayed this way if we had not continued to explore ideas of topic modeling and data mining in my Digital History class. I could not help but wonder if any trends might be visible within the data I had collected. Although my document is short of text, the writing it does contain could hold very specific information. A simple frequency count could quickly tell me which streets appear the most often. This information coupled with a map of the area would give an idea of the density of property owners in different spots. A more interesting discovery would be how frequently different names appear. It would be possible to see if certain areas fell under one person’s ownership or if individual owners primarily populated the area. This could open some interesting avenues for future research into what involvement these people had in the development of the Parkway beyond the acquisition of their property. Along with discovering who these people where and where they were concentrated the ledger could also reveal where the most money was spent by the city in acquiring property.
Although it may be possible that programs like Overview could find trends I have not even considered my real interest in this data would be in the potential of marrying it to a physical location. Programs like CartoDB could help me to visualize the types of densities that I believe an analysis of the ledger would reveal. Although I am still not sure how these types of analysis will ultimately impact my overall project it has forced me to at least think about questions that I had not considered surrounding the Parkway. Although seeing things through a different perspective may not have caused my to completely rethink my work, it has caused me to at least become more away of the other stories and layers present in the development of the Parkway beyond the particular ones I am attempting to engage with.