I've done some research in theoretical physics, and now I work as a programmer. And the two are very, very different in their reward structure. In research it

I've done some research in theoretical physics, and now I work as a programmer. And the two are very, very different in their reward structure. In research it was usual to go for a few months without any relevant feelings of success (and not just me, many of my colleagues felt the same). In programming I usually get some positive feedback a few times a day, at the very least every few days. That's a huge difference.