Hi Kirill
There is a difference between bug (which is deviation from spec) and
feature (where product behaves by specification). As has been
repeatedly noted, Excel's behavior is feature because it is in
specification. Whether you like specification or not, you can not call
it bug.
You're quite right. I was the one who first used the word "bug", and that
was before I realised it was an Excel specification.
To be absolutely accurate, though, it is a bug. A bug in Lotus 123 or
Visicalc or wherever it came from, not in Excel. Whoever it was who made
the initial error didn't do it on purpose. (There's probably some
interesting history there.)
Excel's "specification" was for compatibility reasons. When Microsoft
wrote Excel, there was already innumerable Lotus 123 (and other)
spreadsheets, and some of those contained formulas which would produce
wrong results unless Microsoft followed suit. Now, a decade and a half
later, there are even more spreadsheets that will be broken if the formula
is actually fixed. Maybe it would have been better for them to bite the
bullet back then, but who knows?
I've been wondering about how many spreadsheets this problem actually
affects:
1. Only spreadsheets that contain date formulas (is it only that one
function, or all date functions?)
2. And only when those date formulas refer to a date before 1 March 1900.
That's probably a small percentage of all spreadsheets, but I'm sure there
are some specific areas of interest where they would come up all the time.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
