The data for this chart was derived from my interpretation of the Facebook Terms of Service over the years, along with my personal memories of the default privacy settings for different classes of personal data. The population sizes are statistics from Google, the Facebook Data Team, and wild guesses based on what seemed reasonable to me.
Then Facebook decided to turn “your” profile page into your identity online — figuring, rightly, that there’s money and power in being the place where people define themselves. But to do that, the folks at Facebook had to make sure that the information you give it was public.
The rest of this paper is organized as follows: We ﬁrst derive the multi-image PIFS algorithm by generalizing the single- image PIFS algorithm. We also describe a way to optimize said algorithm using DFT lookup tables. Afterwards, we take on the problem of combining the “right” images into groups, by ﬁrst describing efﬁcient ways to compute a distance function between two images, and then, in the next session, comparing a number of clustering algorithms working on such a distance. The ﬁnal algorithm is evaluated by compression runs over a photo database consisting of 3928 images