"sum" quotes its arguments -- what's up with that?

> The reappearance of certain topics over a period of years
> might be considered evidence that, in fact, something is wrong.

Yes, something *is* wrong -- Maxima doesn't understand variable scope
in sum, integrate, plot2d, or even lambda in many cases.

> are simple workarounds -- quote i or use j instead.

I don't agree that quoting i is a good solution here, though I *do*
agree that autoquoting the argument is *also* a poor solution.

> There is an analogous and equally strong argument for
> making "integrate" an mspec. Shouldn't "sum" and "integrate"
> both be mspecs or both mfuns?

Yes, please read my earlier postings (and bug reports) on precisely
this issue.  Integrate and sum are inconsistent -- in fact sum is
internally inconsistent.
 
           -s