https://bugs.documentfoundation.org/show_bug.cgi?id=37271

--- Comment #13 from [email protected] ---
Not being an Excel user I hadn't realised that Excel also deviates from
mathematical convention, but I agree that compatibility should be maintained as
far as possible.

My question is, would a gnumeric-like solution of background-bracketing -x^y to
(-x)^y would at least give users a clue that the result may not be what they
intended, if they understand the maths behind what they are perhaps trying to
achieve.

Would such an approach break anything?

I have difficulty in accepting that an inaccurate result in a spreadsheet from
a formula observing conventions of operator precedence should not be considered
potentially a major problem.  Also, in maths, for something so fundamental,
surely this "convention" is a standard in all but name?  Other approaches
(postfix etc) might be considered somewhat non-mainstream and I think are
likely to be used by those who are mindful of such issues and know what to
check for and against (eg. scientists, mathematicians) rather than, say, a
doctor or pharmacist calculating a dosage and expecting the spreadsheet to work
like maths does...

-- 
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