Research

Social Software

My research focused specifically on the photo sharing website Flickr.com. We were interested in learning more about how people use the cite - how they find new photographs, which users they add to their social networks, and so on. Using the Flickr APIs, I wrote several scripts in Ruby to collect data on users, photos, and groups and store it in a MySQL database. I wrote supplemental web scraping scripts to obtain data not available via the APIs from the website in compliance with Flickr's Terms of Use. Over a one week period, we were able to assemble a rich dataset of information about approximately 1500 photos and 55000 users. Our findings suggest that the primary way users discover new images on Flickr is by perusing the latest images submitted by their contacts, a process we call "social browsing." The social network data is still being analyzed.

A paper describing our results has been submitted to the 2007 International Conference on Weblogs and Social Media. A draft of that paper is available here. Since it hasn't been published yet, please do not cite it without permission.