On 04/15/10 18:31, Regina Henschel wrote:
The problem is, that Excel 2010 has a lot of _new_ functions. Therefore
the user gets a lot of #NAME? errors. He can see in the input line, what
formula causes this error, but the bad thing is, that the user is not
able to see the value, that has been calculated in Excel.
Just like with imported xlsx files.
Yes, a feature to show the stored results from a file could be useful.
But I don't see a sudden urgency because of Excel 2010. Improving the
import, even in smaller steps, seems like the better way.
No, you cannot use xls, because the functions are not available in xls.
To make it clearer. Up to Excel 2007 there are only small problems. Most
major functions have a counterpart in the other application. But with
Excel 2010 we get more than 50 new functions. A lot of them can be
mapped to Calc function. But there are at least AGGREGATE, CONFIDENCE.T,
CEILING.PRECISE, F.DIST, MODE.MULT and NETWORKSDAYS.INTL (I have not yet
examined all the others) where a simple mapping is impossible.
You wrote:
"When saving, we would have to know if the file is meant to be opened in
Excel. In the end, it would boil down to choosing a different file
format."
That would be the choice C of my initial posting?
Keeping only the new functions in Excel's namespace would mean that both
Calc and Excel can interpret some of the formulas in the file, but
neither can interpret all.
Otherwise, it depends on what you mean by "offer". Loading a file that
came from Excel does not mean that the user wants to open it in Excel
again after saving. On the other hand, someone might want to create a
new file and send it to an Excel user. So any automatic solution where
the user doesn't get to choose won't work and only cause confusion. If
the user has to choose, that choice could as well be the existing xls
export.
Where a mapping is possible, the result can also be saved to xls, so
again, improving the import is what helps interoperability most.
Niklas
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
