> when I suggest they stop, step through the creation of the bug, and just think about what algorithms and data changes are occurring at each step, they often h

> when I suggest they stop, step through the creation of the bug, and just think about what algorithms and data changes are occurring at each step, they often have an epiphany and realize where the problem must lie. All the prior exploration probably often lays the foundation for that epiphany, though.