Do they not have entrance exams for Princeton?
There is no reason to believe that epidemiological models apply despite the desperate sounding crap supposed to justify it - the reference to MySpace is so poorly thought out it makes me wonder if you need to do abstract thought to get into Princeton these days - if you think about it for more than a knee-jerk second MySpace cannot be looked at as a precedent for a methodology analysing the future success or otherwise of Facebook.
"The precedent for applying epidemiological models to non-disease applications has previously been set by research focused on modeling the spread of less-tangible applications such as ideas"
That's it - something as massive as the dissemination of ideas has a superficial resemblance to some epidemiological models therefore it applies to FaceBook which is not an idea but an internet service with an actively changing relationship with its clients. Oh - and apparently when no-one is actively espousing an idea it is because they lost interest in it (after all what else could have happened to an idea) which is the equivalent of becoming immune to the idea.
If all that is not enough they then admit that its all based on a Google mashup anyway.
FaceBook may well be getting stale in a digital world moving at a tremendous pace and personally I don't think they would be any great loss. But this paper is just trolling bollocks.