Writing a common lisp system in python would solve your problem and be more feasible, in my opinion.
If you want to write a program to do 80-90 percent of freshman calculus problems, copy the program in Norvig's book which implements the derivative-divides method (and is based on a program I wrote with a student). It is often the case that one can, with only 10% of the code, solve over 80% of the problems by using fairly weak methods. William Stein said.. I so wish the years of use that Maxima has seen meant it were nearly bug free. I certainly don't mean to ignore bugs and code quality in this discussion. ... Writing new python code might be beneficial in eliminating some sorts of bugs, but if the code were constructed by copying each subroutine from lisp into python, it would probably have many of the same bugs, and likely would have more, unless that top-10 python coder were also extremely good at reading Lisp. Writing new python code from scratch, reading the literature would be another approach, but not likely to produce good programs. A friend once compared programming to making waffles. The first one never looks right. As for what percentage of users of Sage use it essentially as a front end to access Maxima's calculus program vs. (say) other material that is not encountered short of graduate-school math, I suggest you try to gather some statistics. When I was running a table-lookup integration program on the internet, I was surprised by what percentage of the inquiries were syntactic nonsense, and of those that were syntactically acceptable, how many of them were sin(x). The world out there is very different from the world I expected with respect to mathematical interests and sophistication. As an alternative feasibility proposal, I have, from time to time suggested that the core part of Maple, that implements its programming language, be implemented in Lisp. Then all the programs written in the Maple language could be freely executed within a Lisp environment. (Naturally this would require a Maple license). But programs written natively in Lisp could interoperate somewhat with Maple programs. Also Maple language programs might be compiled to Lisp and to assembler. Also Maple would be able to benefit from the (possibly better) implementation of garbage collection in a good Lisp. Also Maple would (except for front end) run on all computers running the same Lisp, reducing the effort of the Maple people. You might consider rewriting that core in Python, though it would not absolve you of licensing issues, it might be fun to load all of Maple into Sage to see what it could do. Rewriting Sage in Lisp is like the algorithm for removing a chocolate stain from your clothes. The final step, if nothing you have tried removes the chocolate, consider dunking the entire garment in melted chocolate and turning it all into chocolate color. RJF --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
