"JE McGimpsey" wrote...
....
There's no ambiguity that negation and subtraction are different. The
fact that the typography is ambiguous means that you need to check your
assumptions.

That's the real problem here. The character - is being used both as a sign
character and as an operator. The apparent goal is consistency of results
for, say, -3^2 and -A1^2 when A1 == 3. Excel's results are consistent: both
return 9. FWLIW, Lotus 123's results are also consistent: both return -9.
The consistency of results is what's important, not the precedence.

While this isn't a bug, it was at best a questionable design decision.

Those who insist that a computer application must conform to *their*
standard have never programmed in APL.

The only language I know for which this isn't ambiguous because different
characters are used for negative sign on the one hand (part of the number
token) and monadic/diadic minus on the other. PITA to use if you cling to
any preconceived notions of left-to-right evaluation, but NEVER ambiguous.

Actually that's unfair to C and Logo (and probably most other functional
languages). C and Logo don't have exponentiation operators, just power
functions, so again no ambiguity.



Posted Via Nuthinbutnews.Com Premium Usenet Newsgroup Services
----------------------------------------------------------
** SPEED ** RETENTION ** COMPLETION ** ANONYMITY **
----------------------------------------------------------
http://www.nuthinbutnews.com

In article , "Harlan Grove"
wrote:

PITA to use if you cling to any preconceived notions of left-to-right
evaluation, but NEVER ambiguous.

Yup - had to learn APL for my high school math classes. To be
successful, it was hard to avoid learning the math, rather than just the
formulae.