Registered: Oct 2003
Posts: 38

There was an article "A New Kind of Ecology?" by Jane Molofsky and James D. Bever in the May 2004 issue of BioScience.

At least for the moment, the full text article can be freely downloaded at
http://www.ingentaconnect.com/searc...000005/art00011

The article basically uses NKS as an excuse to review CA models in ecology, so it is pretty useful as a bibliographical starting point.

Then it tries to make some comments about NKS. It gets a few things right, and a few things wrong of course. For instance, it seems to assume that NKS === cellular automata. It also wrongly states that probabilistic rules are not capable of class 4, and makes the wrong deduction that class 4 behavior probably doesn't happen in ecology.

It does get other things right - that complexity is easy to generate and we might be better off looking for the simplest explanations.

In any case worth a read.

__________________
Everything is an expression.

Report this post to a moderator | IP: Logged

Registered: Feb 2004
Posts: 8

Nice article, and it certainly is a net positive for the NKS crowd, but I do not think that the author fully understands the point of the book or class 4 automata.

For example, the paper mentions one of the premises of the book which is that simple rules may yield complex behavior, but it is as if she did not get to the part about the universality of rule 110 and the concept of compuational equivalence. Infact, she says it is "difficult or impossible to distinguish class 4 behavior seen in the field from random stochastic fluctuations".

I personally believe that this article misses many key points, and that if what was noted in NKS was truly understood, then she would not conclude by saying that "..given the redundancy in the processes that control ecological dynamics, determining a priori which interactions are essential to system behavior seem unlikely. This leaves ecologists back where we started, measuring multiple processes, projecting their dynamics and interdependence, and trying to digest generality from the output".

That last part makes me think she surely did not read the section on perception and analysis.

Report this post to a moderator | IP: Logged