Title: Trade-off between virality and mass media influence in the topological evolution of online social networks

Abstract: Nowadays, available comprehensive datasets of online social networks (OSNs) have enabled researchers to study the social universe at a previously unattainable scale. The worldwide impact of OSNs and the necessity to sustain their rapid growth emphasize the importance to unravel the laws governing their evolution. However, the co-existence of interacting OSNs competing for the same users makes the identification of such laws challenging. Here, we study an OSN with a unique set of properties --a very high population coverage, a language-induced isolation, and the absence of strong competitors-- which allows us to gauge the fundamental mechanisms responsible for the growth of a quasi-isolated OSN. Our findings suggest that the coupling between the real pre-existing underlying social structure, a viral spreading mechanism, and the influence of the mass media play an essential role in the evolution of OSNs. The empirical validation of these ideas allows us to draw the following conclusions. First, the viral effect is four to five times stronger than the mass media effect and, second, individuals have a higher tendency to participate in OSNs by invitations from weaker social contacts, in agreement with the "strength of weak ties" paradigm.