stdlib: Free Open Standard for Office Documents

As an initial focus area for the stdlib workgroup, we're looking into the issue of promoting Open Document Format (ODF) over Microsoft's Office Open XML (OOXML) and preventing the latter's approval as an ISO Standard.

Free Open Standards promote interoperability and avoid vendor lock-in, while proprietary standards make room for some parties to prohibit some uses of the standard or of implementations thereof.

Reliance on technologies or knowledge not covered in Free Open Standards, as well as patents that are not licensed for anyone to use in implementations of the standard, or derived versions thereof, render a standard proprietary.

Some standardization bodies have rules regarding patents, such as demanding disclosure, by participants in the process, of patents or patent applications that cover proposed standards. Some take a step further, in an attempt to promote Open Standards, and demand RAND ("reasonable and nondiscriminatory") patent licensing policies for all covered patents. And here lies a serious problem.

Charging fixed or per-unit royalties from anyone who wishes to license the patent is often regarded as RAND, as if the need for paying royalties wouldn't discriminate against the poor, or against businesses in (often underdeveloped) areas where an elsewhere-reasonable fee turns out to be unreasonable.

Even more seriously, per-copy or per-user fees prevent users in the jurisdiction where the patent is applicable from copying or distributing the software in freedom. Such fees are fundamentally incompatible with the very notion of Free Software. They are neither reasonable nor nondiscriminatory, because they discriminate against Free Software, and that is totally unreasonable.

Some discriminatory policies are even more subtle. Consider, for example, the Office 2003 XML Reference Schema original Patent License.

When it demands licensed implementations to require acknowledgment to potential Microsoft's patents in derived works, it discriminates against well-established licenses in the Free Software community, such as copyleft licenses, that reject any further requirements over those present in the license itself. One such copyleft license, the GNU GPL, is the license for more than 50% of all Free Software projects. Another, the GNU Lesser GPL, is the license of OpenOffice.org, the main threat to Microsoft Office's monopoly.

Microsoft eventually acknowledged the inconsistency of this approach with public-at-large and government demands for Free Open Standards, and published the Microsoft Open Specification Promise, but it still fails to permit Free Software implementations, not out of license incompatibility, but out of failure to permit full enjoyment of the freedoms to adapt and improve the software, without which the Software is not Free.

Meanwhile, the Office Open XML format specification remains available to anyone. As long as the MS-Windows-specific file formats in which it's represented can be decoded, that is.

After ODF, the Open Document Format, was approved as a Free Open Standard by an International Standard Organization, Microsoft submitted its own specification to ISO for FastTrack approval.

Setting aside the question as to why we should rush to approve, under so much pressure its proponent, yet another standard for the very same purpose, conflicting and redundant with other international standards while at that, we wonder if this 6000-pages specification is truly a Free Open Standard.

It carries a number of dependencies on earlier Microsoft decisions, not all of which are part of the already-huge specification, and Microsoft's promise covers only fully-compliant implementations. But Microsoft Office isn't fully compliant with the OOXML (Office Open XML) specification, therefore those who seek interoperability with Microsoft's software won't be covered by its promise.

And then, it might very well be the case that Microsoft licensed third parties' patents used in its standard, and its promise does not cover them. Even if anyone else succeeded in addressing, in an alternate implementation, all of the Microsoft-centric dependencies needed for full compliance, there would still be a threat that some third party, colluded with Microsoft or not, would demand patent royalties from this effort.

Microsoft itself wouldn't have a problem with this, but how about the rest of us? Why should the society at large, at a time when we already have an excellent Free Open Standard for office documents, accept the standardization of another format that would only bring us incompatibilities and risks, while furthering a monopoly convicted several times for anti-competitive practices?

It's not like stopping Microsoft's Office Open XML's standardization would somehow exclude Microsoft Office. It's already incompatible with this specification, and there are at least two implementations of ODF for it already, in addition to ODF implementations for many different office suites, applications and file format conversion programs. It's Open XML's MS-Windows-centric huge and incomplete specification that looks like a move to exclude, limit and delay the alternatives.

A number of governments, companies and organizations are concerned about the imminent danger of approval of the OOXML proposed standard, because it would void the promise of interoperability through a single international Free Open Standard. Meanwhile, most of the society remains oblivious to the dangers of proprietary standards.

We can't overemphasize their importance. Think Egyptian hieroglyphs without a Rosetta stone. And, to realize the inconvenience of multiple standards, think of the Rosetta stone itself, an official document that had to be published in multiple languages in order to reach all inhabitants of that region. Why would society benefit from having to publish documents in multiple encodings in order to ensure everyone can get to them? We already have one format that works well, it's already supported by all widely-used office suites, and anyone else is permitted to implement it. We are better off without another international standard, Open or Proprietary.

Please help us spread information, fight these threats to society. Work with us and your national standardization bodies to ensure that the ODF international Free Open Standard is endorsed as a national standard and gets wide adoption. This will contain and, in the long term, reduce the harmful effects to society of proprietary standards, such as the induction, preservation and extension of monopolies like OOXML.

We want your help to maintain in our Wiki information on what citizens, organizations, companies and governments can do to promote the adoption of Free Open Standards, and to ensure that only Free Open Standards, such as ODF, are approved and endorsed by national and international standardization bodies. You can also send us information and join us in the stdlib@fsfla.org mailing list.

Copyright 2007, 2008 FSFLA

Permission is granted to make and distribute verbatim copies of this entire document without royalty provided the copyright notice, the document's official URL, and this permission notice are preserved.


Last update: 2008-05-05 (Rev 3440)

svnwiki $Rev: 15576 $