Perhaps my illustrations were unclear
I am not asking that J display real numbers at higher precision than
it has them stored
but exactly the contrary
I am stating that it would seem ideal to me and not necessarily to
the rest of the world, when dreaming in colour rather than in actual
implementations of J
that one considers the actual algorithm and its actual ability and
limits and display and store numbers in the most precise form but in
no higher precision than it is capable of producing. Thus if there
or for example only 3 significant digits, store and display
1.23En
and not
1.23000En
if there are 7 significant digits
1.234567En
and not
1.23456E(n+1)
if data entered is
1234567890123456789012345678901234567890123456789012345678901234567890
and it is only stored and used as
1234567890123456En
that is how it should be denoted
to reflect the digits that are significant
to omit digits that are spurious because of the algorithm
and not to arbitrarily truncate significant digits at any point by
default
or default to a display of a number in less precision than it is
actually stored altho it could be explicitly requested to be formated
in less precision
this is not a request to change the implementation of J but just some
musings as to what this might mean
Donna
[EMAIL PROTECTED]
On 25-Jun-06, at 12:05 PM, John Randall wrote:
Asking J to display real numbers at higher precision than it has them
stored will give correct but misleading results. This is at the
root of
the confusion with 8! formatting.
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
