Waldek Hebisch <hebi...@math.uni.wroc.pl> writes: > Martin Rubey wrote: >> >> > In the past I tried to keep Expression as mathematical as >> > possible and conseqently had serious doubts about functions >> > like 'abs'. However, now I think that we have no chance >> > for mathematical purity in Expression. >> >> This makes me very curious: why do you think so?
We probably should define what "purity" means, and I admit I forgot what you wrote about the mathematical model for EXPR. In any case, wouldn't it make sense to split EXPR into several layers? In particular, might it make sense to have a expression-like domain with reasonably nice definition which works provided a testing-for-zero oracle? ---------------------------------------------------------------------- One thing I learned to love in sage is that it is quite disciplined concerning "good domains". For example, it has QQbar, a reasonably well working analogue of AlgebraicInteger. It has problems with speed (which may be inherent) if you are interested in the minimal polynomial of a number, but it is still quite useful. (I used FriCAS twice to guess the algebraic equation and solve it, then Maple's gfun & equivalent to get the asymptotics automatically, then sage's QQbar to find the minimal polynomial for the leading term.) Another very good domain sage has are symmetric functions... However, sage still doesn't have the "good" domains concerning differential equations and recurrences. Also, its support for lazy sequences (in particular Taylor series) is still very lacking. All the best, Martin -- You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group. To post to this group, send email to fricas-devel@googlegroups.com. To unsubscribe from this group, send email to fricas-devel+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/fricas-devel?hl=en.
