Gary Hamel: So a few months back and you had you know this kind of almost instant meltdown on Wall Street from all the program trading that exploded. I mean where does all of this go wrong and what do we need to worry about as more and more we embed these decisions in our systems.

Tim O’Reilly: I think there’s a lot that can go and we’re going to discover it the hard way. There’s no question that algorithms are programs, are parts of programs and you know, programmers know that you got to do debugging. A friend of mine once defined debugging brilliantly. He said, “Debugging is the art of understanding what you really told your program to do instead of what you thought you told it to do.” And of course, we don’t know often what we really told our computer to do until it goes off and does it. And so we have to be aware. We have to have human tripwires, and we need to be able to correct when things go wrong.

And of course you now, the Web 2.0 companies know this. They are pretty good at updating their software, updating their algorithms on a continual basis because that’s another thing that’s going to radically change the nature of corporate IT. You know the idea that you can have a project that takes 18 months or two years or three years to fix a problem, just goes away. In this Internet-connected era, if you’re not able to roll out a new version on a daily basis or even an hourly basis sometimes, you’re not up to the game because this is a living system.

But there’s another aspect that I think we can learn from Wall Street and I think it’s really, really critical because it’s where, not just where you know companies go wrong over time. It’s where the whole ecosystem goes wrong.

When Wall Street started, it was really a set of companies that were assisting customers to build liquidity for the market. And at some point, the firms started trading against their customers. And I started thinking about this. I did a conference called Money Tech which I was trying to figure out, you know I said, “Hey, the Internet is becoming this network economy. Wall Street’s a big network economy long before that.” What will financial markets tell us about the future of the Internet? And that idea that the companies start trading against their customers was one of my big takeaways and since then, I’ve been watching it on the Internet.

Gary Hamel: And the Congressional hearings as well right?

Tim O’Reilly: That’s right.

Gary Hamel: Because there’s a few companies that gotten pulled up short on that.

Tim O’Reilly: Yeah you know Google used to be all about creating value for other people on the Web. You know we’re sending you off to somebody else’s page. And increasingly more and more search results point back to Google. We’ll give you the data directly. And they’re developing more and more products that are designed to capture the eyeballs rather than to feed the customers. And in each case, they’re able to tell themselves, “Well this is really a better user experience.”

I think that the mistake that the companies make is that that’s not good enough because even though you can say, “Oh yeah, this is really a better user experience.” You have to think about the health of the entire ecosystem. And it was pretty clear that when companies stop creating value for the ecosystem, stop pushing energy out so that other players can benefit, creating what Umair Haque calls "thick value" and instead focus on creating thinner value where they’re effectively trading against their customers, trading against their partners, then the whole ecosystems starts to go awry and innovation slows down. You really become more of a monoculture. And monocultures tend to fail.

We saw this so vividly in the computer industry with Microsoft. Microsoft was a huge ecosystem builder for the first decade or decade and a half of their existence. And at some point, the ecosystem's growth had slowed down and they started getting growth by consuming the ecosystem at which point, innovation went somewhere else over to this funny thing called the Internet where nobody thought it was possible to make money. And all the crazy people were kind of hanging out having a good time, and the new innovation economy grew up there, somewhere else because they stopped feeding the ecosystem.