Introducing I/O

By setting up this research blog, I'm finally getting around to something I've been planning to do ever since starting my postdoc at the beginning of this calendar year. I'm calling it I/O, not just because I deal a lot with inputs and outputs in my day-to-day work as a researcher, but also because I hope to provide a bit of a peek into the ins and outs of doing the kinds of research work I'm currently doing. It involves collecting, handling and analyzing novel kinds of data.

At irregular intervals I will be posting snippets of code and other solutions to research problems I've come up with in the course of my work. My work relies heavily on the Python programming language and a number of excellent libraries in its ecosystem, including pandas, numpy, igraph, shapely, and fiona. I'm mostly dealing with social media data, most of which I've collected from Instagram using my research tool, kijkeens. There's a good chance I'll be touching on other issues and tools as well. The criteria for what I post will emerge over time, but my idea is that everytime I figure out how to do something that didn't seem obvious or trivial to solve, I'll try to share something about that here.

I've benefitted immensely from what others have generously posted around the web, so at the very least I can repay the favor by sharing some of my adventures. I also know from previous academic blogging experience that writing up an idea every once in a while and just putting it out there can be enormously beneficial simply as a form of low-stakes writing assignment. Since I really don't know what I'm doing much of the time, it's good to get additional eyeballs on my work every once in a while. Of course that's what peer review is meant for, but I'm honestly not sure whether a peer reviewer in sociology would be able to catch errors in my SQL code or a poorly conceived lambda expression. (No offense!)

Finally, a few notes on this site. It is based on Daniel Rodriguez' excellent site and uses his IPython Notebook plugin for Pelican. I'm grateful to Daniel for making all parts of his site, including the beautiful theme, available under permissive licenses. Without his firm foundation, I would have put this off even longer.