TED Session 11: Nate Silver, Alex Tabarrok
1. Nate Silver, statistician
Election, only 90 days ago. If you look at states, just about everywhere became more blue. 20-30-point swings in highland regions for Obama compared to Clinton, it’s remarkable. What’s the matter with Arkansas. 14-20 percent of exit polls said race was important and said they voted for McCain.
Is racism predictable? There are strong predictions, when you look at education — low education = race-based voting. Also, how rural you are leads to race-based voting. So yes, racism is predictable.
Are there any black people in your neighborhood. Racism is predicted by a lack of interaction with people of other races. Cities facilitate such interactions. So we could think about smart street grids — makes you interact a lot with your neighbors. Alternative to cul-de-sacs — he thinks they made the country more conservative. Another idea is intercollegiate exchange program — the networking experience you get when you get to college.
When something is preDictable it’s Designable.
2. Alex Tabarrok, economist
The first half of the last century was terrible — wars and other events threw up political walls, trade walls, transportation walls, communication walls that divided people. In the second half, we took down tariffs, we used container ships, trade has increased. China is the world’s greatest anti-poverty program.
One world, one market.
Larger markets increase the incentive to produce new ideas. World GDP growth is increasing. Growth can wash away even what appears to be a Great Depression — growth went up at a faster rate after Depression than previous rate or adjusted rate would predict.
We need to keep globalizing, keep extending across boundaries, keep improving education.
3. Pete Alcorn, iTunes podcast director
Used to be Malthusian, but a declining population will have beneficial economic effects. Take away land speculation, price of land drops, that lifts burden off world’s poor. Declining population means more jobs. Europe after the plague shows this.
If we can make it through the next 150 years, our great great grandchildren will be planning for the next Enlightenment.
4. Bruce Buenode Mesquita
It’s a great thing if you can predict the future, because then you can change the future. Game theory. People are rationally self-interested. People have values and beliefs.
We don’t need to know very much to predict. We do need to know who has a stake in the decision. What they say they want (not what they want in their heart of hearts). We need to know how focused they are compared to other issues. And we need to know how much clout they would bring to bear if they chose to engage on the issue. We all care about two things: outcome and credit.
Uses this method to produce that Iran will not build nuclear weapons.
5. Nicholas Negroponte
I have presented 14 times at TED. OLPC proposed three years ago here, and some people predict netbooks will be half of the market in 12 months. You can thank us for that. But OLPCs are more hardy — he throws them around the stage. Commercial markets are interfering with non-profits. Half the kids with our laptops in the field are teaching their parents how to read and write. The only complaint I hear from parents is that they get too much email from the kids. In another three years, we will have 5-6 million a month going to children, and it will be totally open source, including hardware.
6. Dan Ariely
His interest in irrationality came when he was in the hospital and badly burned. The nurses took the bandages off him by ripping him quickly. He had 70 percent of his body burned so it took about an hour. He begged them to go slower, but they said they knew what they were doing, and that the word “patient” did not mean to ask questions or fear.
Left the hospital three years later (!) and studied at university. So he sudied it: originally used carpenters’ vice to crunch people’s fingers, then moved to sound and electrical shocks, added a pain suit to add much more pain. Dark sense of humor.
Because we don’t encode duration in the way we encode intensity, he would have had more pain if it had been more slow. Also start from the top, give him breaks. But nurses didn’t know that.
He did experiment where he invited people to cheat by self-reporting. We saw a lot of people who cheat a little bit. But cheat less if they think they would get caught. Is there a level of cheating we can do at a small degree that doesn’t change our impressions of ourselves? We call this the “fudge factor.” What can we do to reduce the fudge factor? People asked to recall Ten Commandments as a test cheated less. So we got people to sign an honor code — which was particularly interesting because MIT doesn’t have an honor code. Distributed Coke to refrigerators at MIT. Left plates with money in fridges, and those were never taken. Being a step removed from cash increases cheating — giving tokens to exchange for money instead of money doubled cheating.