https://bugs.documentfoundation.org/show_bug.cgi?id=80017

Andrew <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|DUPLICATE                   |---
     Ever confirmed|0                           |1

--- Comment #6 from Andrew <[email protected]> ---
Sorry but I don't agree with the resolved status due to duplication with bug
67026. I understand that both explain the problem in terms of floating point
representation but bug 67026 goes on to talk about multiplication and division
whereas 80017 talks ONLY about adding and subtracting.

Maybe I'm being overly picky and obnoxious but subtraction and addition CANNOT
- I REPEAT CANNOT - be concerned with rounding error. How Excel and Calc choose
to represent numbers should take this very basic mathematical axiom into
account. NO MATTER what I do something like the example in the Wikipedia
article
(http://en.wikipedia.org/wiki/Numeric_precision_in_Microsoft_Excel#Subtraction_of_Subtraction_Results)

should never have to be checked, especially as we're talking about simple
numbers with 3 decimal digits.

Maybe - I am being facetious - Calc and Excel need to introduce an option to
represent numbers within a certain range in a real number notation which
guarantees reliable representation and manipulation of simple numbers.

Why am I so uptight about this subject? As an ex-programmer and systems analyst
(Assembler, C & C++) I cannot accept that basic, axiomatic mathematics be
totally ignored.

-- 
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Libreoffice-bugs mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs

Reply via email to