Sometimes, I think that more of Maxima should be written in Maxima.
The speed advantages of CL over Maxima aren't all that great for
symbolic work. But the bugs in the Maxima to Lisp translator are
discouraging.
When I wrote hessian and jacobian, I was thinking about getting Maxima
to work with matrices with indefinite size such as
(i,j) -> kron_delta(i,j) + kron_delta(i,j-1). I never pursued the
idea, but I thought that using genmatrix would be the way to go to
make this work. Anyway, that's my excuse for the redundancies in hessian.
Barton