On Tue, Jul 7, 2009 at 11:24 AM, rjf<fate...@gmail.com> wrote: > > > > On Jul 7, 10:25 am, William Stein <wst...@gmail.com> wrote: > ... >> >> I agree that having a component of Sage whose source code is visibly >> difficult for Sage developers to work with is a disadvantage of Sage, >> and one we should fix (by not using said component). >> >> It's hard to tell whether or not this is a disadvantage of Sage >> relative to the Ma's, which is what we're discussing. Perhaps closed >> source software also sometimes has major components that nobody >> currently at the component really understands and can >> modify/rewrite/fix. > > I believe this is the case for parts of Mathematica that were written > by persons who have left the company or died > (the unfortunate case of Jerry Keiper, > http://www.stephenwolfram.com/publications/other-pubs/95-keiper.html > ). > Presumably new people learn about / rewrite code eventually. > > There are, I think, only two "original authors" of the Maxima code > that look at it at all, and then rarely going over stuff they wrote > before. (Stavros Macrakis and me.) > > The solution to this particular issue you propose, (to rewrite Maxima > in python) is defective on at least 2 grounds. > 1. It would be an enormous task. > 2. The persons responsible for the coding might leave (or die), and > then you would have some version of the same problem: who would know > how to fix bugs. Presumably the solution then is to reprogram the > package using whatever is then the language du jour? > > One alternative, which you are following, is to report bugs to the > Maxima mailing list. > > Another would be to actually study Maxima. You would then be able to > understand the magnitude of the task, and maybe also fix the > occasional bug. > > I assumed that you also treated as black-boxes such libraries as > MPFR, QD, ATLAS, BLAS... If you have expertise on these packages, I > will keep you in mind if I find a problem in them!
QD is no longer part of Sage since it provided no real advantage for Sage over MPFR, and gave wrong answers on sparc solaris. Sage developers definitely don't treat MPFR/ATLAS/BLAS as black boxes. And, you're right -- as you say above, we don't treat Maxima like a black box either. - William --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
