>
> PS. It occurs to me a larger issue is whether Maxima should
> be comfortable w/ Lisp complex numbers; I'm not sure, although I'm
> leaning towards yes.
That would be fine with me, but we would of course have to run a fine-tooth
comb over the Maxima codebase.
For example, some Lisp and Maxima modules look for "syntactically complex"
expressions by seeing if the symbol %i is literally contained within them.
Of course that doesn't catch cases like log(-1) anyway.
Some other things to resolve:
We would in any case have to continue to support the form a+b*%i for numbers
where a and b are bfloats.
Another problem: in GCL at least, (sqrt -1) = (sqrt -1.0) => 1+6e-17 + i,
not 0+i exactly.