On Sat, Mar 15, 2008 at 11:29 AM, Robert Dodier <robert.dodier at gmail.com> wrote:
> On 3/15/08, Stavros Macrakis <macrakis at alum.mit.edu> wrote:
>
> > exp(x)-%e^x => 0
> > log(exp(x)) => x
> > exp(x+y)/exp(x) => exp(y)
> > factor(exp(2*x)-1)
> > integrate
> > radcan
> > limit
> > etc. etc.
>
> Whatever simplifications are already implemented for %e^x,
> can also be applied to exp(x), with sufficient programming effort.
Yes, of course. But beyond the pure waste of effort, the consequences
of representing closely related concepts in more than one way are
inconsistency and confusion.
And what is the advantage? Some obscure cases of ev(...,numer) are
somewhat more consistent?
-s