That, to me, is a kind of brilliant environmental ju-jitsu - using the energy of the market and the profit-motive to get businesses to invest in preserving and improving natural systems.

By focusing on teaching businesses about the ROI they can achieve by preserving and investing in nature, you're expanding the scope of the impact you can have.

In the food case in particular, one of the technologies that could help there - genetic technologies that could create better crops with higher yields and less need for water and fertilizer - is tremendously feared. Very little of that fear is scientifically grounded.

If you want to feed the planet and keep the forests we have, you need to be able to grow roughly twice as much food per acre around the world. How do you do that? New technology.

On almost every environmental issue I care about, in fact, I've been wrong at one point or another. I used to think that climate change was no big deal, that most environmental problems were massive exaggerations, that oil reserves were effectively unlimited, and more.

I decided five years ago that I wanted to truly understand, for myself, what the state of the planet was, and when I dug into it, what I found was quite different than I'd imagined.

There are really two kinds of optimism. There's the complacent, Pollyanna optimism that says don't worry - everything will be just fine and that allows one to just lay back and do nothing about the problems around you. Then there's what we call dynamic optimism. That's an optimism based on action.

Unfortunately, in the environment, I don't see as much willingness to invest heavily in R&D as I do in consumer technology. And that's a pity.

Ozone and climate are global issues, and it's hard to find a way in which the benefits of shutting down carbon emissions are going to pay for themselves for any given power-plant, say.

Often we need to use policy to level the playing field, or to be sure that a technology is managed in a responsible way.

You have to be willing to spend an awful lot in that R&D phase before you see the benefits. When you look at the companies that have really won customers over in technology - say, Apple and Google - you find that they spend billions of dollars on R&D each year, often spending that much on a product before they ever make a dime back in profits.

I think the environmental movement is now so large and diverse that it's hard to talk of it as a single entity.

You have to be able to generate usable energy without greenhouse gas emissions and you have to be able to do it cheaply if you want people to choose that approach.

We don't have good global policies in place for climate.

We have climate change caused by greenhouse gas emissions, mostly from human power and transportation infrastructure. At the same time, we have 2 billion people who live in energy poverty.

If your incentives are set up wrong - if for some reason you reward people for behavior that's actually bad for your customers or your organization - then you're going to encourage that behavior.

When you're managing a large number of people, you learn that incentives matter tremendously. You really want people to be rewarded for doing the right thing for the customers and the organization.