Comparison with OOXML function list

["David A. Wheeler" <dwheeler@dwheeler.com>]

Nobody knows what OOXML will look like, but I've taken the best information
on function names available to me, and developed a list of differences
based _solely_ on a comparison of function names.

Here is the list of function names that OOXML defines, that we do not:
-AVERAGEIF
-AVERAGEIFS
-BAHTTEXT
-CHIDIST
-CHIINV
-CHITEST
-COUNTIFS
-CUBEKPIMEMBER
-CUBEMEMBER
-CUBEMEMBERPROPERTY
-CUBERANKEDMEMBER
-CUBESET
-CUBESETCOUNT
-CUBEVALUE
-ECMA.CEILING
-ISO.CEILING
-NETWORKDAYS.INTL
-NORMSDIST
-NORMSINV
-PHONETIC
-RTD
-SUMIFS
-USDOLLAR
-WORKDAY.INTL

E.G., Note that our CEILING function has the capabilities of
both "normal" ceiling and "Excel weird ceiling".
Oddly, OOXML doesn't have any way of implementing
the mathematical FLOOR operator (which we do).

Note the "CUBE" functions.  Several functions, such as CHIDIST,
are currently in our spec but named "LEGACY.CHIDIST";
see the separate messages about that.

Unsurprisingly, we have many functions that OOXML does not,
because our work is based on examination of many applications
instead of being the documentation of a single implementation.
Here's that list:
+ACOT
+ACOTH
+ARABIC
+B
+BASE
+BITAND
+BITLSHIFT
+BITOR
+BITRSHIFT
+BITXOR
+CHISQDIST
+CHISQINV
+COMBINA
+COT
+COTH
+DAYS
+DDE
+DECIMAL
+EUROCONVERT
+FORMULA
+GAMMA
+GAUSS
+IFNA
+ISFORMULA
+ISOWEEKNUM
+LEGACY.CHIDIST
+LEGACY.CHIINV
+LEGACY.CHITEST
+LEGACY.FDIST
+LEGACY.FINV
+LEGACY.NORMSDIST
+LEGACY.NORMSINV
+MULTIPLE.OPERATIONS
+MUNIT
+NUMBERVALUE
+PDURATION
+PERMUTATIONA
+PHI
+RRI
+SHEET
+SHEETS
+SKEWP
+UNICHAR
+UNICODE
+XOR

I know that the OOXML folks copied our IFERROR function,
so that does NOT show up as a difference.  That's good!
They didn't copy our IFNA function, though, which is odd
because it solves similar issues.

Do not make the mistake of thinking that OOXML describes
an existing implementation, such as Excel.  As far as I can tell,
OOXML is still not out, and the latest drafts I know of do _not_
accurately describe what Excel actually does.  This is merely
potentially-useful information as we try to develop an
accurate specification for interoperability of spreadsheets.

--- David A. Wheeler