Dear categorists and process algebraists, I'd be very grateful for any feedback on my current work, summarised at http://lama.univ-savoie.fr/~hirschowitz/papers/dcats-confs.pdf and more fully developed in the drafty long version http://lama.univ-savoie.fr/~hirschowitz/papers/double-cats.pdf . The reason I'm writing to both lists is that it is about a semantics for CCS, which uses rather sophisticated category theory (to me at least). For process algebraists: the semantics may be viewed either as a presheaf semantics with innocence, or as a game semantics with the possibility of accepting a play in several ways. For categorists: the involved tools include pseudo double categories, polynomial functors, and 2-categorical limits. Here's an abstract: We introduce a new algebraic structure called playground. From any playground D we construct (1) a language S(D) and its operational semantics (a labelled transition system (LTS)) and (2) a game (or presheaf) semantics. We then construct an LTS G(D) for strategies in the game semantics, and a translation from terms to strategies, which is shown to be a strong, functional bisimulation. We consider a particular playground, D_ccs, which makes previous work with Pous (2011) on Milner's CCS an instance of our framework. Finally, we exhibit a functional, weak bisimulation of standard CCS into S(D_ccs), which shows that the game semantics of CCS given with Pous is a weak bisimulation. Weakness of bisimulation here is not w.r.t. synchronisation, but w.r.t. a kind of administrative reductions close to heating in the chemical abstract machine (Berry and Boudol, 1990). Thanks in advance for any comments, Tom [For admin and other information see: http://www.mta.ca/~cat-dist/ ]