Thursday 18 December 2008

Consistent mental models and programming

A class of computer science students was given a programming quiz on the first day of their studies, before they were taught anything. I find it interesting that other programmers believed students who refused to answer the quiz would make the best programmers. The reasoning, apparently, is that those students have the sense to know that they do not know anything. I admit that humility of wisdom is highly prized by my people in our users.

But the highest performers in the subsequent years of the course were those who applied a consistent mental model (whether wrong or right) to the problems. Still others used inconsistent models to answer the questions, and these were expected by non-technical people to be the most successful programmers since they applied the most appropriate model for the situation. I may be unusual among programmers, but I did think the consistent ones would be best. The inconsistent ones would have to learn the same thing over and over, while those who simply refused to answer are more likely to be slack than humble. They were rebelling, not confused.

Mokalus of Borg

PS - Another study suggests that consistency might not be the whole story.
PPS - It's rarely that simple.

No comments: