The unique universe by Lee Smolin

In his very entertaining article , Lee Smolin argues against the "timeless multiverse" concept (i.e. that our universe is merely one of the many probability-driven universes) and explains why Newtonian physical schema fails when applied to the study of the universe as a whole. In Newtonian schema, there is a notion of "configuration space", i.e. of the set of possible states of a given system. To find a law, we set some initial conditions, run some experiments, change the conditions, run the experiments again and thus distil the laws that are independent of these conditions (and in some sense independent of time since time in this scheme is being used only as a "parameter on a trajectory in configuration space"). But in case of the cosmology that tries to find laws for the whole universe, there is no such thing as initial conditions, and there is no possibility to repeat experiment to find out what belongs to "law" and what is conditional. Thus, history of universe as a whole becomes not one of the possible trajectories of configuration space but essentially laws of physics itself.
Funny how it reflects McLuhan's "medium is the message" concept, btw.
Discussion below the article is quite interesting, too.