For many years now our research team has been involved in an effort (both theoretical and technological) that can be labeled as an attempt to investigate the notion of cooperation from the 'participation' or 'contribution' perspective. From our perspective, it encompasses a set of situations in which different actors identified or unidentified, ratified or not, distributed in space and time, contribute to a sometimes ill-defined collective goal, using most of the time low-overhead web-based technologies. Doing so, people participate to a collective design that aims at generating a bunch of perpetually moving collective knowledge and decisions submitted to discussion, negotiation and sometimes dismissal. To avoid the design of services as a repeated "one-shot" process we have gradually built a transverse software infrastructure that can be used in various projects aiming to design participatory services using complex knowledge and cooperations. Thanks to this "non-disposable" infrastructure, the participatory services designed take profit from the scientific outcomes of each previous project.