On 5/29/06, Henry Rich <[EMAIL PROTECTED]> wrote:

Such a wide 'tolerance', which I called interval comparison, is an 
application-specific requirement that is not, and should not be, part of the 
general-purpose language.

Might be a bit extreme. J has been extended to include a lot of number
systems boolean, complex, extended... why not intervals and multiple
precision?
http://www.mpfr.org/
http://en.wikipedia.org/wiki/Open_interval

Precision (which differs from the host systems) is a pretty central
part of fields where there is a hardware design component, such as
graphics (which is what i was attempting in 94).

~greg
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to