I think the article in Scientific American was probably aiming at the
Santa Fe Institute itself. Unfortunately, John Horgan (the author) is
almost absolutely right about current complexity research. Except for
a few number of people at the SFI (like Kauffman, Mitchell, Holland,
etc), most researchers there do work on the "Gee, isn't this
simulation reminiscent of a biological phenomenon" pradigm.
I am now preparing a course on
Artificial Life, and it is very hard to find textbooks that don't make
you want to throw up! They usually do not understand the differences
between simulations and realizations, nor do they know (or try to
know) any work done before 1986 in Cybernetics and Systems Research
(e.g. Von Foerster, Pask, Maturana, VArela, etc), or even -- more
surprisingly -- theoretical biology (Pattee, Conrad, Rosen, etc.). Not
everybody in ALife or complexity studies behaves like this, but the
most vocal people do, so Horgan is mostly right.
I do believe that complexity is somehow in the eye of the beholder,
and what it amounts to is interesting behavior. However, ultimately
everything is in the eye of the beholder, so we must work on
definitions of terms like complexity BEFORE claiming that something is
complex. This is of course where Rosen's work is of great interest; he
proposes a workable definition of complexity. It is not easy to
approach his work though...
Luis