The link discusses the research done by Cannarella and Spechler. The author of the post, P.J. Lamberson, expresses skepticism about the method and the conclusion.
As Lamberson explains it, Cannarella and Spechler use a modified infectious disease model to describe Facebook viral adoption and abandonment.
There is good reason to be skeptical about the conclusion. As Lamberson notes, just because the model fits data from MySpace does not mean that the model captures the user dynamics of Facebook (or even MySpace).
Of course I’m sure that the conclusion to draw from the research is that Facebook users are going to suddenly develop an awareness of, and a desire for, privacy. As a result, they’re going to abandon Facebook en masse and join nderground :). We built nderground so that it will scale. We’re ready!