In the public sphere as in academia, the frenzy over big data has been met with equally high levels of enthusiasm and criticism. “Unprecedented empirical opportunity” vs. “poor data”; “methodological revolution” vs. “fascination for large N”; “scientific revolution” vs. “debasement in the production of knowledge”: the positions are polarized. Based on a review of the debates in the social sciences, this paper shows that this situation is likely to persist as long as the discussion is organized around the fuzzy term of “big data.” Instead, this paper proposes to distinguish between different types of data, each of which raises specific questions. By so doing, it shows that big data is but one aspect of a much wider transformation –the massive availability of digital data– which in turn raises new questions for our disciplines. Four aspects are subsequently explored: disciplinary reorganizations, the evolutions of statistical methods, the access and management of data, as well as the objects of social sciences and their relationship to theory.