Six Degrees

I just finished reading Six Degrees, and if I had it to do over again, I wouldn't.

Networks are the metaphor for our age. I am attracted to them as the moth to the flame. My gut tells me we've only glimpsed the start of something incredible. Hive mind on steroids. No, words can't even capture what humanity is headed into.

Smarty-pants that I am, I figured I'd get a leg up on most of the civilized world and formulate some rules for understanding the nuances of networks, social and otherwise. This book sucked me in with some great turns of phrase: "It is only slightly unfair to characterise sociology as a discipline that attempts to explain human behavior without the humans."

The author writes:

Regardless of whether we are discussing magnetization or the frezing of water into ice - procedures that involve completely different physics and ven completely different materials - it turns out that the nature of the corresponding phase transitions is the same!

The observation that very different systems can exhibit fundamental similarities is generally referred to as universality....

By knowing all the universality classes for a particular kind of model, physicists can make some very powerful statements about what can and cannot happen in different kinds of physical systems, again by knowing only the most basic facts about them. This is a tremendously hopeful message for anyone interested in understanding the emergent behavior of complex social and economic systems like friendship networks, firms, financial markets, and even societies.

Boy, I'll say. That's what I was after in reading the book. Wow. Let's get it on.

Sad to say, that was about all I learned. Geez. I waded through three hundred pages of a book by a guy with a Ph.D. in theoretical and applied mechanics and a professorship in sociology at Columbia, only to find nothing but theories that had no bearing on my reality. Oh silly me.

The science of networks appears to be similar to the science of economics. Supply and demand work if people are rational, and it's too bad they're not, for that rips the heart out of most economic conjecture. Similarly, network analysis usually assumes all nodes are equal or bandwidth is free or some other pie-in-the-sky notion. At least the scientists have figured out that a static diagram is not a very faithful reproduction of a dynamic network.

Let me save you some time. Here are the lessons of this book:

1. Things may be closer than they appear.
2. Little things can mean a lot.
3. History is an unreliable guide to the future.
4. A diagram of a network is not the real thing.

Oh, and if you didn't know it, that six degrees thing that Stanley Milgram wrote about in Psychology Today, finding that it only took six hops between acquaintances to link any two people in the world, it's crap. Bad data, flawed experiment. Didn't happen. Not a bad name for a book, though.

Posted by Jay Cross at April 15, 2003 10:18 PM | TrackBack