By the way, if we are starting to implement nontrivial, high
performance random algorithms, we should start thinking about making
it easy for the user to supply advanced random number generators.
quasi-random vs pseudo-random, gnu scientific library generators,
numpy, etc.

And if the code is to be useful for research, we may want to add
optional cythonized modules.

All that I am saying is that for competing with established software,
we should really think about performance.

Most of my comments are probably applicable only to the group theory
module, but anyway, I am mentioning it.

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/sympy?hl=en.

Reply via email to