Thursday, September 3, 2009


Innovation is the friend of speculators and the enemy of investors. -- F. Street

Below are excerpts from an interview with W. Brian Arthur, an external professor at the Sante Fe Institute for Complex Studies and the author of a new book, The Nature of Technology: What It Is and How It Evolves.

"There isn't any deep understanding of innovation out there," Arthur said in an interview with "And I think you can see that because the way innovation is described is very hand-wavy, and-then-something-creative-happens. All societies want to be innovative, but in the absence of any deep idea about innovation, governments and companies tend to run after what seems to be the latest idea; that if you somehow have, 'creativity,' or invest in R&D, or set up industrial parks, that's going to work."

The book's basic argument is that new technology is just combining old technologies in new ways. And all technology is, at its core, simply the harnessing of nature and its manifold phenomena for human needs.

Historically, innovation starts with very simple technologies like mastering fire and forging tools and developing wheels. These then become the building blocks for the next generation of technologies, which then, once perfected and made simple, become the building blocks of the next technologies, and so on and so forth, right up to things like the modern hand-held computers and genetic engineering.

Often innovation arises from improving and deepening current technologies, using existing tools to find cheaper and more efficient ways to do old things. Sometimes, innovation arises from borrowing ideas from different domains and applying them in new ways. Occasionally, a radically new innovation like electricity or the transistor comes along, making a whole generation of previously unthinkable technologies newly possible.

The key implication of what Arthur calls the "combinatorial evolution" view of technological innovation is that these innovations do not come out of nowhere. "There are not magic wands or bright ideas in bathtubs," Arthur said.

Rather, they emerge in what we often tend to think of as a more mundane way, from something that Arthur calls "deep craft" — that is, from a really thorough understanding of the existing technologies and comprehensive knowledge of a domain. "What you really need in invention is a superb command of the pieces in a toolbox," he said. "What really counts is a mastery of some vocabulary."

Since technological innovation arises from combinations, it follows that the more tools one can command, the more potential combinations one can produce. And the more potential combinations, the more likely one is to find the truly innovative solution — the new technology that neatly solves the problem at hand, and perhaps makes all that came before it obsolete.

Such an understanding has important implications because it means that innovation is not necessarily something that follows just because the president of the United States announces $2.4 billion in grants for electric cars, as he did in his Wakarusa speech (incentives can help, but Arthur argues it is dangerous to target them too much because it's so hard to predict where the next innovation will be).
Nor is it productive to wait around for some elusive moment of genius. ("I don't believe there is any such thing as genius," Arthur wrote.)

Read the rest