A year ago, many words were written (including by me) on why Microsoft may have chosen Ecma to package Microsoft's Office Open XML formats as a standard. Now that Ecma has finished that project and adopted the result, there's additional data to examine that sheds some light on that question. That will be my topic today, and for the next several entries.
About two weeks ago I wrote a related entry called Ecma Approves OOXML – What Does it all Mean? In that entry, I tried to give an even handed overview (admittedly, as I see it) of how the Ecma approval of Microsoft's XML formats fits into the grand scheme of things. The bottom line was that I did not think that the Ecma action was very significant, given the circumstances under which it had been achieved.
That post elicited the following question from a reader:
Maybe I'm a bit naive . . . but does this mean that Microsoft is trying to get the various standards authorities eating out of their hands - i.e. uncritically approving everything pumped out by the behemoth?
I tried to give that question an even handed response as well – because in fact it's common practice for most companies to engage in the equivalent of what a lawyer would call "forum shopping:" looking for the court and judge they think will most likely rule in her favor. My response therefore read as follows:
Different organizations have different policies and attitudes towards accepting contributions or undertaking projects. Many companies (and not just Microsoft) that have a standards project they’d like to see launched typically shop for the organization that they think will serve their needs best. At one end of the spectrum, it can be the one with the greatest credibility and likelihood of getting the standard widely adopted, and at the other end can be the one in which they expect they will retain the greatest control over the project. Or, ideally, both. This has been going on for a very long time.
Sometimes there is negotiating behind the scenes, where the standards organization holds a discussion to determine in advance whether the contributor will guarantee in advance that they will not retain any intellectual property rights that might stand in the way of the standard being completely “open,” and sometimes the contributor tries to impose limitations. In the latter case, the organization needs to decide whether to take the contribution or not. If the restrictions are too great, then it risks damaging its reputation in the marketplace.
This isn’t just speculation on my part, as in my role as counsel to many standard setting consortia I’ve sometimes been brought into the discussion of whether to accept a given submission under the circumstances, or on the terms offered. Sometimes the answer has been “no,” or “yes, but only if you withdraw these conditions, and/or give these additional assurances.” No organization that sets open standards wants to gain a reputation as being a “rubber stamp,” as such a reputation, once it attaches, is hard to shake.
But perhaps that’s not always the case. With c. 500 ITC standard setting organizations in operation, all vying for members, projects and dues, perhaps gaining a reputation for being a rubber stamp might represent an attractive business plan?
I decided to take a look at the Ecma site to see what kind of evidence I might find one way or the other on that question. Here’s what I found, and you can form your own conclusions.
First, let’s take a look at the News section of the Ecma site, to see what Ecma wants the world to think, as demonstrated by its own carefully crafted press releases. Ecma has adopted over 365 standards and 90 Technical Reports since its formation in May of 1961. With that as a backdrop, let’s see how much emphasis Ecma has placed on this single standard development working group since Microsoft contributed the OOXML specification to Ecma in November of 2005.
Here are the titles of all of the press releases that Ecma has issued in the past six months:
December 7, 2006: Ecma International approves Office Open XML standard
November 29, 2006: CompTIA approves the Open XML Open Standard
October 23, 2006: White Paper on Office Open XML
October 9, 2006: Office Open XML Formats (TC45) – Final Draft
September 28, 2006: Last results about Office Open XML Formats (TC45)
August 23, 2006: Office Open XML Formats (TC45) – Intermediate Draft 1.4
July 25, 2006: Last results about Office Open XML Formats (TC45)
July 25, 2006: OSTA and Ecma Join Forces to Establish Industry-wide Optical Disc Archival Testing Standard
If you look at the next six months, you’ll find three press releases announcing adoptions of work product at General Assemblies, and four more press releases about OOXML. The final count? Even though Ecma approved 32 different standards during this time period, the Ecma site lists a total of eleven different press releases relating to the OOXML project, and only one relating to any other single technical project. So it would appear that either Ecma doesn’t think that much of what it’s doing these days is very important, or believes that there is a great deal to be gained from being associated with the OOXML project – or both.
This might reasonably lead one to conclude that Ecma placed a high value on acquiring the OOXML project at the outset, when the Programme of Work for the Technical Committee was written. That Programme, you may recall, was not to create a standard for office productivity software formats, but instead:
To Produce a formal Standard for office productivity documents which is fully compatible with the [Microsoft] Office Open XML Formats.
For further blog entries on ODF, click here
subscribe to the free Consortium Standards Bulletin
(and remember to Buy Your Books at Biff’s)
Simon Phipps called ECMA a coin-in-the-slot standards organization.
It is clear that Ecma is an Industry Standards organization, as opposed to an Open Standards organization, and is subject to influence. Nothing is more telling than their own website, which states (emphasis mine) “Ecma is driven by industry to meet the needs of industry, generating a healthy competitive landscape based on differentiation of products and services, rather than technology models, generating confidence among vendors and users of new technology.”.
Maybe i’m missing something, but logically, by your own analysis, you claim that “negotiations” between the standards body and the submitter happen all the time. Is it not possible that Microsoft negotiated that ECMA would release regular press releases on the status of Open XML? And the lack of press releases on other standards is simply that they aren’t inclined to release press releases unless required to do so (or requested to do so?) by the comittees?
That seems like a much more logical reason to me than conspiracy theory. Occam’s razor and all.
It just seems like you’re stretching to setup a condition where you can arive at the conclusions you’ve already decided you wanted to express.
This is a good question. Here are three thoughts:
1. First, I don’t want to suggest that negotiations happen all the time. Without writing a book on this, offering a complete spec has historically been unusual, especially on an unsolicited basis. Usually a TC is chartered to create a standard from the ground up. Some consortia (like OMG) put out the equivalent of an RFP for submissions of specifications of existing products to be used as a basis for a standard, so that people can immediately buy a compliant implementation. The kind of negotiation I refered to arises in the much less usual case when – as here – the spec is basically already done. This used to be rather rare, before the Men In Black (Microsoft, IBM and BEA) began offering completed specs to multiple standards organizations. So “all the time” would not be an accurate impression of what I meant; sorry if that’s the way it sounded.
2. I’m a big believer in Occam’s Razor. So what is the simplest explanation here?
3. Getting to your specific quesiton, I think (as noted) it’s a good one. But I think it may be that it comes out the same way if an organization that traditionally does no press releases on its standards – as appears to be the case here – suddenly changes it practices to accommodate a member that wants to lay down a barrage of press releases about its standard. Why should Ecma say yes, unless it thinks that it has something to gain, to offset the hit to its credibility from doing so? Any organization that has been around this long would be unwise to begin to kiss up to a single member, because of the obvious inferences that would be drawn, and the impact on its perceived objectivity.
So while you could very well be right, I think the inference comes out the same way, either way – either Ecma wanted to make the releases itself, or was willing to lend it’s name to a member that wanted to use the Ecma name to promote its agenda. The former would indicate that Ecma wanted the standard badly, while the latter would indicate the same thing — or that Ecma is willing to provide a rubber stamp, and that its press releases, as well as its process, are available when desired.
One of the attractions of ECMA to Microsoft would appear to be their status with ISO – and the
expectation that anything ECMA approves will then be rubber stamped via the “Fast Track”
process. This is of concern to those of us working on ISO standards (in my case C++) as it
erodes the credibility of these standards.
More in my recent editorial: http://www.octopull.demon.co.uk/editorial/OverloadEditorial200609.pdf