Directives tell robots everything. What to do. Where to go. After some hundred years though, some robots might wander away from their directives, while others insist they follow the directives to the dot. Why is that? What are the boundaries? Do breaking points really break walls, or do they really reinforce them?