In American math classes, we teach a lot of concepts poorly over many years. In the Asian systems they teach you very few concepts very well over a few years.

It’s useful to compare our preparations for epidemics with our preparations for war.

People are building the software and so having the pieces be such that a single person understands all the tradeoffs and everything that's going on in a piece is extremely valuable. It avoids getting into an experimental mode where you're just trying things out. That never works.

When I was trying to figure out why lives have improved so much in the last 300 years, where we've gone from a third of kids dying before 5 to - by 1990 it was down to 10% - now it's down to 5%. And saying why, over all history, there were smart people, but that number didn't change. Average life span didn't change. What's magical about what's been deemed the Industrial Revolution? It's really energy intensity.

Countries which receive aid do graduate. Within a generation, Korea went from being a big recipient to being a big aid donor. China used to get quite a bit of aid; now it's aid-neutral.

In software you can't really add people and expect to get more done, because their ability to understand the program and what's going on it would require so much investment and all their work would require so much review that you'd be more likely to slow things down.

So on the demand side [for energy], there have been a variety of policies that globally have been way over $50 billion a year of tax credits, raising the price of electricity through things like renewable portfolio standards, so the total amount of money that's gone into sending a price signal to push up demand versus what would happen without it has been gigantic.