This semester I’ve been working on an independent study project that can loosely be described as “impact and effectiveness of digitizing cultural heritage artifacts (namely photos) in creating a broader sense of culture, informedness, and community.” It’s taken many forms over the course of the last few months as I tried to figure out exactly what I wanted to tackle and what I could feasibly accomplish in one semester.
It came in baby steps. I knew I wanted to do something with the Flickr Commons so the first thing I did was write a few scripts that would pull a bunch of data via the Flickr API and stick it into a database that I could hopefully do something useful with. A few days and a lot of queries later I had a solid database of just about 19,500 photos, over 40,000 unique tags, and nearly 210,000 tag instances. I haven’t worked with that much data before and it was certainly intimidating!
We all know that I’m not a huge fan of having to code anything. And that part was difficult, but not nearly as difficult as narrowing down this giant concept into a few research questions and then devising a way to actually go about testing them. I read everything I could find on the topic in hopes of finding something that piqued my interest yet was specific enough to turn into an actual project – still no luck.
After meeting with Evan Earle, a masters student here who, it turns out, is also working on a project involving digitizing archived photos, specifically with the Cornell University Library (he showed me a bunch of photos from the archives and Ezra Cornell’s wedding socks. Yes, his wedding socks) I walked a way with a bit more direction and a few ideas for things that I wanted to test.
A few nights of coding away until 5AM and I had it: commonality. The system is intentionally vague to avoid priming and bias and all of that fun stuff, but it’s essentially a data collection tool that will hopefully help me to prove (or at least better understand) my hypothesis about tagging these types of images.
The system is pretty straightforward and I like to think that I picked a decent selection of photos that will keep people interested as they tag. It takes about 10 minutes to complete depending on how quickly you click and how thorough you are in filling out the survey afterwards. Have 10 minutes? Give it a shot!
There’s still a lot to do: I need to collect as much data as I can (read: get as many people to use the system as I can) and then I need to actually analyze the data and write a paper about it, but I’m already really happy with having managed to complete this much! Depending on my level of senioritis after finishing the paper I might post it here, or I might just go out and celebrate having finished college. Whichever.
Two finals, this project, and five days to go!
(please please please go try out commonality!)