Wednesday, November 28, 2007

Complex Systems

I've always known that systems science (aka systems theory and cybernetics) is in a sorry state. But this was driven home to me this morning when a senior professor in the field commented that "systems mean different things to different people". Hmmm, does it, really?

Systems science is supposed to be abstract, so it doesn't matter whether the system you're talking about is biological or industrial or urban. And indeed, all three of those systems are examples of matter-value systems. Well, what other kinds of systems are there?

It turns out there are 2x2 kinds of systems. There are information systems and matter systems. And each of those can be either pure or valued. So physical systems are pure matter, software are pure information systems. And industrial systems in general are matter-value.

What do I mean by a valued system? I mean one which processes the value of its elements. Or put more baldly, a metacircular system. One that redesigns itself to suit an internalized conception of its own purpose.

An FPGA-based computer whose CPU reprogrammed itself on demand would be an example of an information-value system. But there are other, much more common, systems that fit this category: political systems.

So ignoring chimeras like health systems, we have software, political, physical and industrial systems. And that is all.

No comments: