I also notice that pi is defined as an operator, so that pi() returns the
result %pi.  I wonder how hard it would be to include useful constants in
Axiom, and to be able to enter (for example)

solve(integrate(1/x,x=1..t)=1,t)

and obtain the answer %e.

Alasdair

On 25 May 2007 06:42:37 +0200, Martin Rubey <[EMAIL PROTECTED]>
wrote:

"Alasdair McAndrew" <[EMAIL PROTECTED]> writes:

> I find the behaviour of constants rather inconsistent.

Yes, certainly.

> And it's odd that %e::Float doesn't work, but %e+0::Float

look at the type of %e+0.0 and you will understand.  Note that it is *not*
Float.

> (and any other arithmetic involving %e) does.

No: sin(%e)::Float does not work.

> Also, the Euler-Mascheroni constant should be in there!

Yes, certainly.

Martin


_______________________________________________
Axiom-mail mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/axiom-mail

Reply via email to