Skip to main content

Is life too complex?

The complexity of modern life may have given birth to the science of 'complex systems'. But that hardly justifies spending a lot of money on research. So there may be something in the argument that because we now understand some of the simpler issues of science, we can start to devise grander theories.

Science spent centuries breaking complex problems down into ever-simpler parts, and then trying to understand each part in as great a detail as possible. The argument goes that we can now begin to reassemble the knowledge of those simple parts in ways that explain the behaviour of more complex systems.

A pioneer in complexity research is Professor Stuart Kauffman of the Santa Fe Institute http://www.santafe.edu/. Around 10 years ago he wrote: 'How do we use the information gleaned about the parts to build up a theory of the whole?'

As well as details of Professor Kauffman and his work, the Santa Fe Institute's web site has an archive of working papers that go far beyond research on complex systems. These show the myriad links that researchers make between complexity and evolution, business and culture, to pick a few from the list. There is even one report on 'The Computational Complexity of Sandpiles'.

Talking of business, any scientific theory these days inevitably prompts someone to dream up ways in which it can be brought to play in business. Yes, complexity theory has its advocates. Read about them at CIO Enterprise Magazine http://www.cio.com/archive/enterprise/041598_qanda.html

I landed on this one not because I know the people interviewed, but because I found the link at Brint.com, 'The BizTech network'. This has a page on 'Complex Systems Articles, Papers, Books & Bibliographies' http://www.brint.com/Systems.htm, which is well worth your time.

As this site demonstrates, research into complexity is not the same as 'complexity theory', which is not so much thinking about complex systems as a mathematical discipline, sometimes linked to chaos theory, albeit one that is important when considering complex systems.

To begin to understand complexity theory, you might think of starting at http://www.complexitytheory.com/. But contrary to the title, it isn't a commercial web site but the opening page of 'a graduate course on the theory of computational complexity'.

This site, an offshoot of the School of Computer Science at Carnegie Mellon University, tells us that 'Complexity theory is the study of how much of a resource (such as time, space, parallelism, or randomness) is required to perform some of the computations that interest us the most.' This definition is far more restricted than the work on complex systems, and probably a pretty narrow definition of complexity theory.

Carnegie Mellon is into work on the wider field of complex systems. For example, one group works on project Rainbow, which is about 'Architecture-based Adaptation of Complex Systems'. Read more at http://www-2.cs.cmu.edu/~able/rainbow/

The architecture here is computer architecture, and networks of computers, one of the classic complex systems that researchers ponder. They demonstrate the classic issue with complex systems. Computers and networks usually consist of many components that are in themselves simple enough. But put a whole bunch together and they interact in relatively simple, but non-linear, ways. And that makes it hard to predict what will happen.

Another example of such a system is the electricity grid, which fell over in a big way in the USA in August. This is why organisations such as the US Defense Advanced Research Projects Agency (DARPA) worry about complex systems and support some of the work at Carnegie Mellon and elsewhere http://www.darpa.mil. (I once spent weeks trying to get into this site and started to get paranoid about being kept out because I am a security risk, but suddenly one day the curtains moved aside and I could see into a fascinating treasure trove.)

While it is always worth visiting DARPA's web site in search of interesting science - the US military reaches into mind-bogglingly strange nooks and crannies - a better place to investigate complex systems and defence is at the Information Directorate of the Airforce Research Laboratory of the USA http://www.if.afrl.af.mil/. Step back and be amazed by the offer of a 57.7 Mb PowerPoint presentation on what they get up to.

The home of much of the work on complex systems is the programme on Dynamic Assembly for Systems Adaptability, Dependability, and Assurance (DASADA). This has a site of its own, with numerous links to groups working on complex systems http://www.if.afrl.af.mil/tech/programs/dasada/

One link goes out to the University of Oregon where there are details of a project whose goal is to 'Enable provision of software "gauges" that indicate a variance from predicted behavior before it leads to a violation of acceptable behaviour'. They haven't updated it for a while, but the subject sounds interesting and the contacts may work http://www.cs.uoregon.edu/research/perpetual/dasada/

As you would expect for a rising research subject, there are research institutes dedicated to the subject. One place to visit is the New England Complex Systems Institute http://www.necsi.org/. This has sections showing its work on such things as health care, education, corporate management, the military and Third World development.

NECSI is in Cambridge, Massachusetts, which is also home to MIT. This world famous lab probably works on complex systems, but it gets enough publicity not to need any more. Wander, instead, to Serendip 'a gathering place for people who suspect that life's instructions are always ambiguous and incomplete'. http://serendip.brynmawr.edu/complexity/complexsystems.html. That introduction would make me stop and look even if the site did not have a good page of links on complex systems.

This site, loosely connected to Bryn Mawr College, has a nice way of encapsulating the issues: 'Simple things interacting in simple ways can yield surprisingly complex outcomes... Brains too consist of relatively simple things interacting in relatively simple ways.' Trouble is, those brains, mine at least, do have problems coming to grips with complexity. But somehow I doubt if a load of research programmes will help me much.




Topics

Media Partners