Thoughts on Python

[Barny]

We all know Python is great for scripting, but what are people's thoughts on its use in scientific computing?

For instance, I see little use for Matlab now considering the numpy/scipy/matplotlib libraries for Python. Similarly surely R will become obsolete with Python's stats libraries? It's more convenient to use just one language rather than jumping between matlab/R/Octave/etc.

Agree/Disagree?

 

[Andy Nguyen]

I don't think R will die. Just look at FORTRAN. If you crave out a niche, you won't die.
have you seen http://repl.it/

 

[Yike Lu]

I like python+numpy+scipy+matplotlib. I am making an effort to ditch R and MATLAB for this new stack.

That said, I am making a serious effort. Remember that programming languages are habits of the mind. People use what they are familiar with. If it's a suboptimal choice, they will often keep working with it until it becomes excruciating, simply because the cost of switching to something new is so high.

R and MATLAB also have their own value. MATLAB for its toolboxes, R for its large array of stats packages. One thing I may do as I go forward is translate those R packages that I find useful.