On 3/5/07, Barton Willis <willisb at unk.edu> wrote:
> It's my fault. I've been working on a replacement for newton,
> newton1, and mnewton. My code constructs a lambda form for
> evaluating the Jacobian.
Well, I think there is really no need for apologies. I bumped into
this problem a while back in an exactly analogous way
(automatically constructing a function from an expression).
As that seems like a pretty natural thing to do with a symbolic
computation system, I was surprised to find that it didn't work.
I have also bumped into this when I wanted something like
block([sigma[x]], ....).
FWIW
Robert