spike1@xxxxxxxxxxxxx <spike1@xxxxxxxxxxxxx> espoused:
> Tim Smith <reply_in_group@xxxxxxxxxxxxxxxx> did eloquently scribble:
>> I have saved a spreadsheet in ODF. Looking at the content.xml file, I
>> see that the table:table-cell element for one of my cells contains this
>> table:formula="oooc:=PMT([.B3]/12; 360; [.A3])"
>> Looking at the ODF specification, available here:
>> I don't see any documentation as to what exactly the PMT() function
>> does. Could you point out where this is covered in that ODF
> Does a document FORMAT specification have to specify every single function
> predefined in the spreadsheet? It's a DOCUMENT FORMAT SPEC, it's not a
> spreadsheet SPEC. You know, document formats, those things that define how
> things will save to disk in such a way as they can load back in the same
> data next time? Functions in a spreadsheet don't affect the
> actual format of a save-file and in fact, spreadsheet functions can be
> expanded, added to, modified. Would you need to e-mail microsoft every time
> you came up with a new macro to add a new spreadsheet function so they could
> add it to OOXML?
Of course it doesn't, and indeed, it shouldn't, but this kind of deeply
broken layering is common to most Microsoft products. Their code is
designed to force cross-layer dependencies which should never happen,
thus, OOXML will never be implementable by anyone, probably even
Microsoft would not be able to do it, let alone anyone else.
I wonder how you design a test suite to check for OOXML compatibility?
>> If I had asked a similar question about an OOXML spreadsheet, the answer
>> would be easy, as PMT is covered in the 300+ pages the specify the
>> functions available in spreadsheets, in useful detail. Consider, for
>> example, the function ODDFPRICE. OpenOffice supports that function,
>> but, like PMT, it is not defined in the ODF documentation. In the OOXML
>> documentation, in comparison, there are *three* pages devoted to that
>> single function. (Similar for ODDLPRICE, ODDFYIELD, ODDLYIELD).
> Ahhh, so that's one of the reasons the OOXML "standard" is so fucking
> useless. They defined every single thing no matter if it affects the format
> or not, good way to make its 500000 pages totally unusable to outside bodies
Quite. It's entirely dependent on Microsoft's present and some previous
versions of Word, Excel and probably Powerpoint, and perhaps even
> Of course, they also made it patent encumberable and don't cover everything
> just to make it that leeeeeetle bit more useless too.
Software patents only apply to Nafta countries and Australia. Nowhere
else has shown the degree of corruption it would take to adopt the same
approach, ie., patenting the obvious and patenting maths.
>> Ironically, both of these, and most others that you might find in an ODF
>> document, have similar names and arguments as the corresponding
>> functions defined in OOXML, so if you had to implement ODF from scratch,
>> you could use the OOXML specification to fill all those holes in ODF!
> What holes?
The ones he just invented, of course. Classic FUD. He's turning into
Erik more and more each day.
> Why do you need to know what the PMT() function in a spreadsheet does in
> order to write a format in which to SAVE the spreadsheet?
> The function occupies a cell. So, save the contents of the cell with the
> correct cell address.
This should really be done by layering. One layer should be an index
into functions, and then cell contents should reference that index. A
smart format would only index the functions which were used, of course.
> Simple. Effective. NO need to know the meaning of 4000 built in functions
> for that.
At this, low, layer, it should not be necessary.
| Mark Kent -- mark at ellandroad dot demon dot co dot uk |
| Cola faq: http://www.faqs.org/faqs/linux/advocacy/faq-and-primer/ |
| Cola trolls: http://colatrolls.blogspot.com/ |
| My (new) blog: http://www.thereisnomagic.org |