Patrick Durusau wrote:
> Besides, if the formula group wants a defined set of results, then the
> results aren't by any means "implementation defined," we are simply
> offering a choice to implementations. They cannot chose, for example,
> the weather report that David suggested.
>
> Yes?
I would say: "no".
Which suggests that we need to define what the term
"implementation-defined" means in the document.
I'm with Rob Weir on this one; Rob says:
> It is perfectly legitimate to say "implementation defined" and also
> specify additional restrictions. For example, ISO C++ says that the
> length of a character is implementation-defined, but it
> must be at least 8-bits long.
I would say that the result of calculating 0^0 is
implementation-defined, but must be one of short list of values (1 or an
Error, at least).
Obviously, this means that a spreadsheet that uses 0^0 may produce
different values on different spreadsheets. Such variation is
undesirable, but it's better to document and limit that variation, as
compared to requiring a specific result that we can't get agreement on.
In particular, a spreadsheet document creator can work around this,
once they know that it's an issue.
Thankfully, there are lots of spreadsheets that don't need to compute
0^0 :-).
--- David A. Wheeler
