The Process Challenge - A Predictable Path
***Updated April 7, April 8***
In light of the approval of Open XML as an international standard (see the ISO press release here), there are a few countries in which process issues are being raised. This blog post will act as repository of the challenges and information about each of them.
For the past few weeks I’ve been saying that no matter what the outcome of the Open XML standard work, the specification would be improved and that was good for the industry. As an ISO standard it moves to SC 34-led maintenance and continues to see significant take-up via independent implementations and extension of existing implementations. If it had failed to become an ISO standard it would continue with Ecma TC 45 maintenance and they would decide how much of the post-BRM spec to adopt into the Ecma spec – and it would continue to see significant take-up via independent implementations and extension of existing implementations. Of course all of this is in light of the fact that if you don’t like the specification….then you don’t have to implement it. (I have always been opposed to technology mandates.)
What has remained consistent during all of this is that product competition between MS Office, OpenOffice, Lotus Symphony, GoogleDocs, Acrobat, WordPerfect, etc. etc. etc. has remained firmly in place. Just because IBM is implementing Open XML (which they have announced) and ODF (proving my point about multiple formats) – they will still argue that their product offers greater value than everyone else’s product. Good – and each of the other vendors will do the same.
Industry participation has been driven by commercial interests – essentially acting based upon their own analysis of what they think their customers want from solutions acquired from that vendor. Everyone needs to take a step back and remember that having commercial motivations both in-favor or against the adoption of Open XML as an international standard is a good thing and is completely normal in standardization. It is completely legitimate for companies who are Microsoft partners (or who have an independent interest in using the specification) using Open XML in order to deliver functionality to their customers to participate in the standards process. They choose to invest engineering and business resources in support of Open XML as a means of supporting their own business interests. This is just as true for a business partner of IBMs (or who have an independent interest in document formats) who joined big blue in the work against Open XML because they felt it was in their best interest to do so. Standards work is meant to be inclusive rather than exclusive.
Following the ratification of ISO/IEC IS 29500, those with commercial interests who have opposed Open XML are now pressing a concerted process challenge which is to be expected. Strangely enough, it is ultimately healthy that they are doing this, but I think it is important to look at this next step in the Open XML saga.
The Process Challenge
It is common in litigation that after a decision goes against a party, one of the first things they look to in their appeal is to challenge the legal process itself (and the actions of their opponents) rather than the merits of the case. This is usually because the arguments made regarding the merits of the case resulted in an unfavorable outcome. So, they go after the process itself in hopes of getting an appeal heard and ultimately to turn the decision in their favor.
Ultimately, it is better for the system to have this avenue of appeal always remain open. It means that “the system” (in whatever context you are talking about) has hygiene built into it. It also means, that if the rules were not followed, that the parties involved need to be accountable for their actions. However, you have to make your case that your appeal has merit under the applicable rules and regulations, so attacks for the sake of attacks don’t achieve much beyond generating sound and fury that signifies nothing.
So we find ourselves in this situation with Open XML. IBM/et al worked very hard to oppose the Open XML standard. They committed millions of dollars to the effort, lobbied hard, worked with their partners, but did not get the outcome they were hoping for. The thing that makes the world of standardization different than litigation is that working on a specification (as supporter or critic) results in the specification getting better. IBM was the #1 contributor of comments worldwide. In fact, they were so pleased with their comments that they made sure they were replicated and echoed from national body to national body. In some countries, IBM was responsible for more than 90% of the submitted comments, and everywhere they added essentially the same comments to their work in national bodies. They got their wish – their comments, indeed, 87%+ of all comments, were resolved in the disposition process and the national standards bodies felt that the process had been successful. The standard had improved, and many of them modified their votes (completely in line with the goals of the BRM process) in favor the specification. Ultimately over 85% of all voting countries approved IS 29500. Comments that may not have been addressed during the BRM to a participant’s satisfaction, or any other issues that national bodies can think of going forward, can now be raised in the maintenance process that will be established under ISO/IEC SC34 control. That’s exactly how the system has been designed to operate so that specifications may continue to improve over time.
We now see IBM/et al driving an orchestrated process attack in the hopes of overturning the ratification of Open XML, or at least to discredit what has come out of this long, global process. While I certainly hope they are not successful in this attempt, I do respect the right of anyone to raise questions about the process. National standards bodies have been saying that they voted in favor of the standard because of the importance of the technology, the fact that the process was successful, and particularly the fact that the comments that they submitted (meaning the issues they cared about most) were satisfactorily addressed (properly within the process) during the BRM process.
If there are aspects to the JTC 1 process that need to be changed (and there are some aspects that could be improved to help all future standardization), then that will occur through the mechanisms allowed for the modifications of the JTC 1 directives. Ultimately, the lessons learned from the unprecedented activity around a single specification will be good for JTC 1. We look forward to being a part of a productive dialog about what modifications, if any, should be made to the JTC1 directives so that it can be well prepared for the future.
***Updated*** Anticipating that these process challenges will continue, I will collect information on my blog to help people put the big picture together. To be clear, my belief is that the overall process leading to the ratification of the standard was solid - and that it is extremely unlikely that the result will be changed due to the issues being raised by those opposed to Open XML. I will keep updating this blog post over time. It is not my intent to speak for the national bodies, they will make their own choices about what to say regarding the adherence to their defined processes. The same is true for ISO/IEC (parents of JTC 1).
At the end of the day, a successful procedural challenge needs to walk a fine line between rational debate and personal insult for those responsible for the process. Casting aspersions on respected members of the standards community is not the best way to achieve change. The community attacks on folks like Patrick Durusau (toward the bottom of the screed) are ridiculous. The personalization of the process claims by IBM fall into a similar category in my opinion (if you haven’t seen Rick Jelliffe’s post in response to a Bob Sutor post – it is really worth the read). The outright attempts of character assassination on respected standards professionals around the world by some of the more acerbic voices in the blogosphere are extremely distasteful to me. (Apparently this is true for others as well.)
From the Microsoft perspective – we are going to be part of the international standards community for a long time to come. In the past our participation was not as consistent as it should have been. There is no question that international standards will likely increase rather than decrease in importance over the next few years. My sincere hope is that the global community both moves on to work on achieving interoperability between document formats in the market (more on that in another blog posting) and works to understand what the Open XML process can teach us about how standards development efforts in the rapid development/innovation environment of information technology should be undertaken going forward.
Current reports in the Internet, on the decision procedure in the DIN committee on ITC with regard to ISO/IEC DIS29500 are wrong and misleading.
The vote in the working committee responsible for the expert evaluation on ISO/ IEC DIS29500 has been cast on March 11th. The question was whether the -vote “yes with comments” from September 2007 should be maintained after the insights from the ballot resolution meeting or should be changed to a “No” vote. This was a vote of the experts in the working committee. DIN as such had no vote in the working committee.
This ballot has approved ISO/IEC DIS 29500 with 15 to 4 , this has been correctly reported from non authorized sources. The experts in the working group have voted. DIN as such doesn’t have a vote in a working group.
The steering committee did not have the mandate to review or lift the expert decision. This is not its responsibility. The only question to decide was whether the ISO process has been flawed with regard to formal criteria according to the opinion of the steering committee. Because the decision of the steering committee did not deal with technical issues but with the rules of the JTC1 fast track, i.e. with compliance issues with the rules DIN saw a need to take a position. Therefore the DIS employee took part in the voting and did not abstain as he would in technical issues.
On March 27th the voting members of the steering committee on ICT standards had to vote not on the acceptance or non acceptance of ISO/IEC 29500 as standards but on the compliance of the process. The steering committee has accepted the process to be compliant and therefore seen no reason to lift the yes vote of the working committee. The steering committee has accepted the process as compliant with the rules with a majority of 7 to 6 and therefore it has seen no reason to lift the decision of the working group. If the majority of the working group would have been convinced that the process of dealing with and voting were non compliant to the rules then the German vote would have been changed to abstain.”
Here is a press release from Fraunhofer FOKUS - with quotes from the chair of the DIN SC 34 mirror committee, Gerd Schurmann.
Once again the international collaboration of experts from different countries within the framework of the ISO has proven to be a suitable forum for the development of international standards that can also meet market timeline demands through cooperation with other standardization organizations like Ecma and OASIS.
This is the English press release from TÜV, a respected German institution. Their statement is in favor of the OpenXML ratification by ISO and supporting the ISO and DIN processes. TÜV NORD Group (http://www.tuev-nord.de/), with its staff of over 8,000, more than 6,300 of whom have technical or scientific backgrounds, is one of the biggest technical service providers in Germany.
Standards Norway has released a statement - here is the English version in PDF (note: the translation I had up before was not the official English version. I have pulled that text from the blog in favor of the doc from Standards Norway.) Standards Norway has provided a very thorough discussion of the process leading to their decision. It is worth reading.
This was supported by coverage in the International Herald Tribute, and the NY Times. Ivar Jachwitz, deputy managing director of Standards Norway is quoted as saying [lifted straight from the article]:
"We had an initial vote back in 2007 of nearly 50 people and the vast majority were in favor," Jachwitz said. He did acknowledge that 21 members of the group last week submitted a letter asking for Norway to oppose Ooxml. "Our vote reflected the majority opinion," Jachwitz said. "I do not see that it was improper."
And Roger Frost of ISO in Geneva said in the article,
We have received background information from them and have no reason to question the validity of their vote.
Additionally, coverage in the Norwegian press (Aftenposten) has a quote from the Ministry of Trade and Industry. Here is the unofficial translation of the quote from the paper:
The Ministry has nothing to criticize Standard Norway’s handling of the Norwegian voting procedure regarding the Microsoft-standard Open XML and the ISO-approval. – The Ministry does not engage in standards processes. It is the responsibility of Standard Norway to execute standard processes in a proper manner. We have now received a brief from Standard Norway on how they handled the process, says State Secretary Anne-Lene Svingen (labor party).
(My analysis) I have a colleague in Europe who is tracking this closely as well - Stephen McGibbon's post. The situation in Norway has emerged as one of the more contentious around the world. When I look at the situation there are a few key factors that jump out at me:
First, Standard Norway, at the beginning of the process, was clear in that the decision process would not happen based on a majority vote in the committee. In spite of this fact, the anti-Open XML participants inflated the committee with their representatives just prior to the Sept. 2 vote seeking to influence the decision based on a majority vote - Standard Norway immediately rejected that proposal. The curious part is that as long as the decision went their way (for the anti-Open XML community), they made no complaint. Following the Sept. 2 vote of “No with comments,” there was no process concern from the anti-Open XML community even though the process was identical to what has happened following the BRM process.
Second, after all of the “committee stuffing” claims from IBM, it seems that both IBM and Google joined the committee just two weeks before the final March 28th meeting. Ironic, no?
Finally, Standard Norway repeatedly stated following Sept 2nd that the key issue in deciding the final Norwegian vote is to what extent the 12 Norwegian comments were sufficiently met. The fact that the comments were addressed then drove a decision to move to yes.
There is a call for a protest march on the Oslo SC34 meeting. I'm not sure of this, but I'm guessing that this is the first time an SC34 meeting has been protested.
Croatia was one of the countries that voted yes for DIS 29500 on the September 2nd ballot. A month after the ballot resolution meeting, Croatia maintained the yes vote, reflecting the understanding and approach of the members of technical committee. There was an effort by Open XML opponents to change this outcome. During that effort, several policies and procedures were violated when the Open XML opponents sought to organize an electronic ballot at the end of the voting period without the consensus of all TC members. Very late in the process a large number of additional members were added to the committee to vote no (without having been part of the process up to that point). The attempt to overturn the vote and not following the policies and procedures defined by NB’s rules was declared invalid by the National Body and is currently under investigation.