Over the last year, my ongoing cellular automata research (discussed in the large thread below) has convinced me that what we have is not so much a boundary between opposites (order and chaos) as a zone of synergy between complementary roles for the same iteration rule. Now I see, hopefully more clearly, that the extreme sensitivity to initial conditions which defines chaos is also the method of rapidly exploring combinatorial possibilities from which unexpected order/organisation occasionally emerges and that such orderly/locally-predictable processes in turn accelerate the spread of chaos. (This all sounds far too much like the long view of Jedi v Sith, so I'll leave discussing my very recent rediscovery of Mirek's Star Wars rule for another day.)

But none of that overcomes my basic concern with the 1993 paper that what was basically a failure to obtain data needed to reject a null hypothesis was greatly over interpreted. It took until half way through Mitchell's new Complexity: a guided tour for the quote that could not have expressed my own feelings more accurately: "(...) whatever its other merits might be, we cellular-automata addicts can all agree that (Stephen Wolfram's A New Kind of Science) provided a lot of welcome publicity for our obscure passion."