My thoughts exactly! How the hell is flickr data useful unless you know what percentage of all photos taken are uploaded to flickr in the first place? If they actually *could* get that data, and it turned out to be something like 80%, well then that might be different.
FTA: "(Admittedly, the number of images on Flickr is about 5% of that on Facebook. It would be interesting to repeat this analysis using Facebook data, but there is no reason to believe the results would be substantially different.)"
BS article FTW!