I believe this is somewhat similar to how Snippets work in Lift, where each snippet renders by itself and coordination is managed by actors.