As a physicist I'm really glad that Nature talks about IPython. Most of my teachers are only familiar with Matlab or maybe Maple or Mathematica.
I think the main problem with them is not reproducibility. The main problem with these software that you can't really verify the correctness of the results. You will never know the exact algorithm being used for a given simulation or data evaluation. Especially Mathematica uses complicated adapting methods for numerically integrating, solving differential equations, nonlinear regression and so on, but you will never know the exact algorithm being used.
Perhaps that's what they use to teach because those languages are reasonably easy to use from a math/physics point of view, even if they're somewhat limited from a CS point of view. I taught myself to program early on, but for others who went through honours in maths with me it wasn't so easy.
Don't be surprised if your lecturers use other languages on their own work. At my university, everybody in the math department uses Matlab and Mathematica. However I know that a lot of the guys who do smooth particle hydrodynamics use Python and Fortran, and my honours supervisor said he once programmed in PostScript (which isn't so much practical, but it's impressive and shows he knows how to program in languages other than Matlab).
Comments
As a physicist I'm really glad that Nature talks about IPython. Most of my teachers are only familiar with Matlab or maybe Maple or Mathematica. I think the main problem with them is not reproducibility. The main problem with these software that you can't really verify the correctness of the results. You will never know the exact algorithm being used for a given simulation or data evaluation. Especially Mathematica uses complicated adapting methods for numerically integrating, solving differential equations, nonlinear regression and so on, but you will never know the exact algorithm being used.
Perhaps that's what they use to teach because those languages are reasonably easy to use from a math/physics point of view, even if they're somewhat limited from a CS point of view. I taught myself to program early on, but for others who went through honours in maths with me it wasn't so easy.
Don't be surprised if your lecturers use other languages on their own work. At my university, everybody in the math department uses Matlab and Mathematica. However I know that a lot of the guys who do smooth particle hydrodynamics use Python and Fortran, and my honours supervisor said he once programmed in PostScript (which isn't so much practical, but it's impressive and shows he knows how to program in languages other than Matlab).