constants, simplification, numerical evaluation



On 3/15/08, Richard Fateman <fateman at cs.berkeley.edu> wrote:

> if we prefer exp(x) to %e^x, I think there are lots of consequences like %e
>  becomes exp(1), and subst(2,%e,%e^x)  results in exp(x), not 2^x. And maybe
>  stuff like sqrt(x) becomes exp(1/2*log(x)).  Also sqrt(%e) has 2 values, but
>  exp(1/2) has only one.

Suppressing the simplification exp(x) --> %e^x doesn't imply any of this stuff.

The only one of the examples above that might come into play is
subst(2,%e,%e^x) --- if we put in place %e^x --> exp(x) then, yes,
that substitution would fail. I'm not losing sleep over it.

Robert