The diversity of science blogs that emerged in the 2000s, their success and longevity, has generated a great deal of research. Blogging platforms have been studied more from the perspective of communication between layperson and expert than from the perspective of interdisciplinary collaboration and community building. The belief that blogs are personal objects hide the fact that a lot of them are in fact written by more than one hand. We will present here part of the work done on the platform “hypotheses.org”. Created in 2008, in France, today it hosts several thousand academic blogs in Europe covering all areas of the humanities and social sciences. Our corpus includes 389,089 posts written since the opening of the platform by 12,893 different authors on 3540 blogs. We studied the evolution of authors' co-presence in blogs and mapped clusters in order to understand the practice of collaboration or cooperation over three different four-year periods. Indeed, these practices vary. In some periods authors participate in several blogs gathered around a field of study or an academic discipline. But multi-author blogs can also exist around questions of researcher practice (what methodology to adopt for the survey, or in their thesis) because the authors share a belief in the potential of new digital tools for research. In some cases they also gather around the hope that blogs and digital tools will allow a transformation of research institutions (evaluation in particular) and of the positioning of each researcher in relation to their work. Hence the importance of academic blogs on reflexivity, methodology or epistemology in the early days. The study of these networks allows us to see how, in order to reach a wider public, researchers make themselves and their 'values' visible to their peers, and can thus come together and cross disciplinary boundaries.