Re: [office] spreadsheet table:cell/table:table (was: [office]OpenDocument TC coordination callminutes 2007-08-13)

[Kohei Yoshida <kyoshida@novell.com>]

On Tue, 2007-08-21 at 16:14 -0400, Florian Reuter wrote:
> I'll take a look at the formula spec.
> 
> I just want to make sure that we all have the sme understanding of the semantic of tables-in-tables and sub-tables for spreadsheets. 

That's essentially been my argument as well.  If the element is in the
schema, then the spec should give at least a rudimentary description of
what it does, and what it represents, to avoid allowing the implementers
incompatible interpretations of the same construct.

As Lars said, there has already been a discussion of using <table:table>
inside a <table:table-cell> to represent, for instance, in-line matrices
(which I think is a little stretch BTW, but I won't discuss it here),
while other people may use the same construct to represent something
else such as embedded sheet within a cell if an application implements
such a feature.  And these two applications can be considered
ODF-compliant because the ODF spec *intentionally* doesn't specify how
the nested-table is represented.  BTW, the schema suggests that the
nested tables are recursive.  How would that be used to represent
matrices?

And I picked the nested table issue just as an example.  There may be
similar cases with other elements.  I was just suggesting that we go
through those elements that are allowed inside <table:table-cell> to
make sure there aren't any other similar pitfalls like this.

If the goal of the ODF spec is simply to provide the schema and leave
out the specifics up to the implementers, then I can't argue with that.
But I would expect a little more than that from a file format
specification.

Just my opinion.

Kohei