Scientific Libraries in Python

[Fernando Pérez]

> 
> Would it be possible for one to absorb the other so we have a single
> standard scientific library for Python?

Indeed. SciPy seems to have a very good framework, into which probably others 
could be plugged in. Currently there's a ton of half-baked systems, none of 
which is fully satisfactory. I truly think that with a good, comprehensive 
system of packages for numeric/scientific work, which includes a *real* 
visualization framework, Python can beat as a tool MatLab and IDL by far. 
Mathematica/Maple are a tougher question, because symbolic computing is still 
a bit of a black art. But there's Ginac which has (had?) Python bindings, and 
in the meantime one could call Mathematica via the Mathlink protocol.

The way I see it, we need a solid, integrated set of libaries with:

1) python-level access to all routines

2) c/c++ implementations of the low-level stuff for speed.

3) A true visualization *framework*, which has both high-level plotting 
routines and access to basic primitives. Mathematica's plotting system, while 
not the most convenient to use, is a good example of this idea. You can just 
'plot' stuff, but you also have low-level access to the graphing primitives 
to program more complex stuff yourself.

4) A convenient environment for interactive work.

1/2 are being addressed by SciPy/Scientific, so some integration would be 
nice.

3: there's a lot out there, but nothing convinces me yet. VPython is great 
for 3d work but still in its infancy, gnuplot and grace have python access 
but don't provide low-level access, scigraphica doesnt' seem ready yet, ...

4: we have the prompt, and improvements exist. This is the easiest of the 4.

F.