> You are talking about running huge social experiments based on guesses about what might work better No, I'm not. But it's a good idea. > extremely difficult

> You are talking about running huge social experiments based on guesses about what might work better No, I'm not. But it's a good idea. > extremely difficult to convince people to participate Yes - easier to just stop tricking people into known-unhealthy behaviors first. > Hardly simple. The answer is simple. The implementation may or may not be [more simple than trying to replicate the lives of people in the past].