how can I make a variable local inside "product"?



> I don't think it's possible to automatically distinguish between
> the i which you want to be a dummy variable and the i which
> is something else.


It appears that the new code does *two* evaluations of the summand.  This is
bound to lead to problems like this....   I thought we had discussed this a
long time ago, and in fact discussed this particular kind of problem.

For better or worse, Maxima semantics makes it the case that
eval(eval(<expr>)) is NOT the same thing as eval(<expr>).