Python vs Lisp: was RE: how to pass the name of a variable as argument?

On Wed, May 14, 2008 at 11:12:47AM -0700, Richard Fateman wrote:
>well, as has been pointed out before, you can always write a maxima
>top-level with a different syntax.
>
>A study pro/con of python vs. the Algol-60-ish top-level language might be
>appropriate.
>Given the examples of Maple, Mathematica, Axiom, Matlab, Mupad, Mathcad, ...
>it might make a worthwhile study to contrast them all and try to find either
>a consensus or a new superior design.  
>
> It seems to me that the main strength of python is the number of libraries
>(written in C?) that can be easily called
>from python. If this kind of access can be managed with the Maxima-python
>top level, then I suspect the same technique could be used to link the
>Maxima-Algol-60 top level to python libraries.  While it is not exactly
>equivalent, I think I have pointed out that I loaded the MPFR
>(multi-precision float/real) package into Lisp by loading the python
>library, so it is not implausible.  But are there other arguments?
>
>The original design of the Maxima language was supposed to be aimed at users
>of mathematics, not programmers.
>
>Originally, it seems almost inconceivable that non-trivial programs would be
>written in that Algol syntax.
>the PDP-6 had 1.2megabytes of memory, and a cycle time of 1 MHz or so.  Over
>1,000 times slower. 
>
>given all this, a better design might come up.

Would using Haskell for the top level be feasible?