Given that you are talking about enriching the experience for the physics student, be advised that they use computers in two very different ways. The first way works fine living inside a closed system such as SAS, Matlab or whatever. You have your data, and you take it into the world-of-matlab, and then you do whatever it is that the designers of Matlab thought would be cool and useful things to do. Courses such as 'how to get the most out of Matlab' are extremely useful and justifiably popular. But the other thing to use computers for is to build our own models, libraries, worlds and what-have-you. This, be definition, cannot be done in sombody else's world. The skills to do this have to come from somewhere else -- and this is where a course in Python might be most useful. But they are different skills. And the baby-step things they have in common mostly don't have to be taught at all -- physics students can learn these things by reading the appropriate manuals. My person al belief is that most physicists would get more joy out writing programs and get better results if they learned to do test driven development when they wrote their own code, so if your goal is to improve the experience for the physics major, I would start there. Laura