Universal Interoperability Council

When did IBM learn that Microsoft would not implement Excel spreadsheet formulas in OpenDocument Formats ("ODF") v. 1.1 the same way OpenOffice.org does? And why does that timing matter?

According to IBM vice president Bob Sutor, he learned about it at the October 9-10, 2008 Pretoria ODF Workshop:

About the least positive thing I heard was that in Microsoft's implementation of ODF 1.1 they are planning to use the ECMA OOXML definition of formulas instead of the more logical and forwardly compatible OpenFormula definition in ODF 1.2.

Bob Sutor, Pretoria ODF Users Workshop, Bob Sutor blog (13 October 2008).

Why then did IBM wait nearly seven months, until May 3, 2009 — after Microsoft's ODF 1.1 native support was coded in Office 2007 SP2 — to mount the Big Blue attack on Microsoft's ODF 1.1 implementation of formulas? See Rob Weir, Update on ODF Spreadsheet Interoperability, An Antic Disposition (3 May 2009); reprinted on Groklaw, (4 May 2009).

If the IBM goal were in fact interoperability via ODF 1.1 between Microsoft Office and other ODF implementations, would it not have been more timely for IBM to raise its formula stink before the Office 2007 ODF support was coded so that Microsoft management might have been persuaded to do formulas as OpenOffice.org does, back when there was still time to influence the decision?

And why the preposterous IBM claim that the differing formula support made Microsoft's ODF 1.1 non-conformant with the ODF 1.1 standard?1

Why for that matter is IBM not equally as concerned with other implementations of ODF 1.1 that limit interoperability? For example, OpenOffice.org 3.x warns that selecting ODF 1.1 formats as the default save formats and "[n]ot using ODF 1.2 may cause information to be lost."2 Surely Big Blue is aware that compatibility modes for particular formats can be programmed so that data is not lost when saving to ODF 1.1?

Might the IBM reticence to criticize OpenOffice.org on such grounds stem from the fact that Big Blue is recycling the OOo 3.x code base in its own proprietary programs and was unwilling to assign one of its 35 engineers working on OOo to implement a compatibility mode for ODF 1.1?3

One might reasonably infer from such circumstances that ODF interoperability was far less important to IBM than was preserving its ability to attack the quality of Microsoft's ODF support after it was hard-coded. Or put another way, IBM appears in this instance to be more committed to double standards than to ODF interoperability.

1. 1.

   See Weir, supra (quoting language from ODF 1.1 that — like most of ODF — is not worded as a requirement); see also ISO/IEC Directives Part 2: Rules for the Structure and Drafting of International Standards (5th Edition, 21 December 2004), Annexes H.1, H.3 ("shall" and "shall not" are to be used to express requirements; "[d]o not use 'must' as an alternative for 'shall[;]' "[d]o not use 'can' instead of 'may' [to express permission]"). See also this author's proposal to the ODF TC to repair the ODF specification in relevant regard.

2. 2.

   See in OOo 3.x, Tools > Options > Load/Save > General.

3. 3.

   IBM press release, IBM Commits to Future of ODF With Symphony Roadmap (5 November 2008) (use of OOo 3.x code base); Andy Updegrove, IBM and OpenOffice.org: An Interview with IBM's Doug Heintzman (11 September 2007) (number of developers).