[Edu-sig] Math + Python: reviewing some themes (long)

[David MacQuigg]

michel paul wrote:
> Recently I've found Sage <http://sagemath.org> invaluable for the 
> purpose of getting computational thinking into the math curriculum.  
> I've spent the last year figuring out how to harness Sage in class, 
> and it is paying off.  The difficulty with a pure Python approach has 
> been that it seems so foreign to everyone from kids through 
> administrators, it doesn't look like anything that gets tested on 
> state standards, and it seems like 'hard work' when we already have 
> these nifty hand-helds that graph any function you want.  However, the 
> power of Sage blows any graphing calculator, even the new Inspires, 
> out of the water.  Simultaneously, you can program in pure bare-bones 
> Python within Sage.  So I have found it invaluable to capitalize on 
> the power of Sage to serve as a way to introduce into math classes the 
> value of the ability to think in pure Python.
Nice graphics is definitely a key requirement for any tool I would 
consider in an introductory course.

I'm not familiar with Sage, but I wonder if adding a few packages to 
"pure Python" would do the same.  I'm looking now at NumPy and 
MatPlotLib in a proposal for "Introduction to Scientific Computing", 
currently taught using C with some addons for plotting.  The class is a 
joint effort between our Astronomy and Physics departments.

The advantage of Python/Numpy/MatPlotLib is that what students learn of 
Python will be useful beyond just math and science.  I think of Sage as 
just a replacement for MatLab, not something I would use in programming 
my mail server.

Anyone with experience using these tools?

-- Dave