This report is based on an empirical case study on content moderation, whose objective is to interrogate whether content moderation can be reclaimed by users as a practice they are partly responsible of, not only subjected to. We introduce a participatory method to social sciences that could contribute to research on social big data. Participatory methods can help mitigate the propensity to attribute to Machine Learning technologies and Artificial Intelligence the capacity of solving social issues, and, instead, can encourage public participation. The detection and classification of online content by social media users allow to discuss ethical issues related to automatic methods, as they surface in everyday life.