In the past month, after several weeks of despairing tinkering, I finally got the features in my GeoJSON files to appear as points on my map. I now feel much more comfortable playing around with Javascript to see what my map can do. Currently my map looks like this:

Now that the basics of the map are there — markers for each testimonial advertisement in a newspaper, clusters for concentrated areas of markers, and a sidebar with click-able information — I want to add more complexity to the map to represent more information about testimonials and the history of consumer culture in South Africa. One of the things I’m interested in is highlighting differences between urban and rural writers of testimonials. Since I haven’t been able to find an appropriate historical map of South Africa from the time period, Ethan Watrall gave me the idea to create different-coloured markers for urban or rural locations. Now, I’m working on this task, using population information from the 1936 census. My goal is to represent urban or rural status using different coloured markers, with different shapes representing the different newspapers that the information comes from.

Adding this additional complexity to my map will force me to consider exactly how to define “urban” and “rural”. One metric I’m considering is whether or not a town is listed in the 1920 census list of “towns and villages possessing some form of urban local government.” But this is perhaps a problematic category, as the structures of “urban local government” might have been granted differently depending on the racial make-up of a locale. In that case, my best option may be to pick a certain number that I consider to represent the border between urban and rural, and simply trust that the 1936 census gives a fairly accurate population count.