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.

****************

Country: Germany

Criticisms: NOOOXML, Groklaw, Updegrove

Response:

The following statement comes directly from DIN – among the world’s most respected standards bodies. [English Translation]

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.”

***Update***

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. 

***Update***

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.

****

Country: Norway

Criticisms: Groklaw, Groklaw, Geir Isene

Response:

Standards Norway has released a statement  – here is the English version in PDF (note: the translation I had up before was not the Norway Documentofficial 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.

***Update***

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.

Country: Croatia

Criticisms: Groklaw

Response:

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.


Comments (17)

  1. Deși la votul din vara trecută nu a trecut, Open XML a fost de data asta ratificat ca International Standard

  2. Dr. Manhattan says:

    Jason: Either you are completely brainwashed by your employer or you are the most incredibly cynical of its minions. It is plain disgusting that after the "World Corruption Circus brought to you by Microsoft ™ " that all of us have witnessed you still have te gall to pretend to do the PR damage control for them. You keep trying to charge IBM of the very conduct that Microsoft has been developing from the beginning. I am sorry to invoke the Godwins law, but that is just Goebbels’ nazi propaganda tactics. Most people mad at the people like you at Microsoft are not stakeholders in IBM, Sun or any other of the business competing with Microsoft, we are just normal users who are in the know, fed up with the constant old lock-in and forced obsoleteness tactics that your company keeps trying to exert upon us through incompatible and ever-changing document formats. We are just users that value freedom above money,OpenOffice.org, KOffice, users, BSD and GNU/Linux users fed up of people like you in your company lying to us, trying to make us waste money and time, trying to force us to buy products we dont want because of the de-facto monopoly on the desktop computer market. We are just users that are fed-up with your company and its dirty tactics, your company are losing us as customers, but also our trust, respect and goodwill as citizens, by the thousands. And it will come the time when the people that could be Microsoft’s best customers if it behaved differently and delivered products that would be desirable on its merits instead of being forced upon us, will become its main competitors, because we have woke up and are starting to dennounce the dirty tactics of your company, and to reject ist products (Vista anyone?), but be careful: we are very tech savvy people and we also promote our choices: Our family, friends and employers rely on us for technology counsel, and we are counseling not to choose your company, neither trust it. It is very sad that thanks to the ISO being so naively shortsighted as to allow the abuses in the standards process and this fake standard to bear the ISO stamp, many public institutions and governments will fall into the trap and will keep wasting taxpayers money and subsidizing your employer. But they eventually realise that the emperor has no clothes. We will be there helping them to do so.

  3. arduino says:

    Go the the parliamentary level. Standards Bodies are not trustable, especially when the Standards Norway excuses are written by you know who.

  4. jasonmatusow says:

    Dr. Manhattan – I know there is little I can say that will ameliorate your concerns. I really wish you hadn’t brought Godwin’s law into this (making the comparison you have) because it is distasteful to me in the extreme. Aside from your clear lack of judgement on that issue, I think you both raise an important point and should also take a moment to really look at the broader context. First, to your point about trust of business practices. That is a fundamental issue for any business and MS is no exception to that. In fact, I hold myself and everyone I work with to both the letter and spirit of the ethics we espouse as a company. The steps taken by the company to work on openness and interop over the past 7 years have been transformative for the company. As for your taking a step back…no one is forcing you to use a Microsoft product. If you think there are other technologies that offer better value for you – by all means, use them. If though, you still believe that Free Software is not part of the commercial context of the software industry – and that real $ are drive real activities – then you are sadly mistaken. I recognize that there remains a segment of the developer and user population seeking to follow the central philosophical and licensing approach of Free Software – but you are standing on the shoulders of the billions of dollars invested by large commercial players who are making those investments for a reason. As for IBM and Microsoft’s conduct throughout the Open XML process – both participated with the intent to be successful. Both politicized the issue. Both lobbied. Both appealed to their partners and customers to join them in their approach. Etc. Etc. Etc. I do not view the world through the same rose-colored glasses that you do. I speak with customers of all sizes and in all vertical segments. They are tech savvy, they are value-minded, and they are highly critical of all vendors – holding all of us to a very high bar to perform for them. I’m willing to stand on the merits of our technology and business with those customers any day. I’m proud of the work I’ve been doing for a very good company. Sorry that doesn’t jive with your world view.

    Thanks

    Jason

  5. jasonmatusow says:

    arduino –

    You should read the Norway section again in this blog post. The Standards Norway position was written by…yes, you guessed it…Standards Norway. The process challenge is a legitimate part of the lifecycle (you can even find it well documented in the JTC 1 directives). I’ll keep my mind open to it if you will.

    Jason

  6. quux says:

    Dr. Manhattan: You say "You keep trying to charge IBM of the very conduct that Microsoft has been developing from the beginning."

    I don’t think that was the insinuation at all. What I saw was more along the lines of ‘We’re a black kettle. We’ve always said so; there are no rules about being black when you spend your days in the fire. But my my, isn’t it ironic to see you, a black pot, calling us a black kettle without achknowledging your own blackness?’

    And yes, invoking Godwins Law was just tacky.

  7. Dennis Byron says:

    Good idea.  How about adding information about Microsoft’s protest in Inida so as to be fair and balanced?

  8. Yes ..really thanks everybody, people who were "for" and people who were "not for". I believe that everything

  9. Kimo says:

    One place leads me to another. Just an open question: why are Microsoft employees to active writing blogs and articles about this matter? Actually I’m amazed to see such activity in support and defence of a process that probably has damaged ISO for a long time to come. Why celebrate? Or isn’t Microsoft concerned about the condition of such an important organization?

    I’m very concerned since nobody outside of the Microsoft’s inner circle knows what a possible approval would mean. Will it mean what it has in the past? Then we’re damned. Or has Microsoft suddenly made repentance and changed its policy towards openness so suddenly? I wish so, but even the employees and supporters of Microsoft can’t deny the history.

    I’m working every day with administration of Microsoft software, so you might expect me to celebrate a possible OOXML approval. However, frankly speaking, I’m so sick and tired of the mess Microsoft and now Trend Micro (just examples) is doing. It’s hurting progress a lot, don’t you see that? I can’t get a pure working horse of an operating system from Microsoft any more, because it’s so stuffed with all kinds of DRM and other similar junk. But please, what does that have to do with my working stations? Anyone?

    Alright, our company has improved because of this. Today we’re at least using a mix of systems: FreeBSD, Linux and Windows. Will this OOXML vote strengthening Windows’ position in my working environment? It has just helped me to understand the importance of making the right brave choices: next time I set up a company it will be Microsoft free from scratch (and that says an administrator of Microsoft systems)!

    Thanks Microsoft and all your bloggers and supporters for making this clear to me!

  10. Open XML says:

    Analyse intéressante de Jan van de Beld ex secrétaire général de l’Ecma à propos de l’hypocrisie de certains

  11. Scott B says:

    Where can we expect challenges?

    1) ISO rules not being followed.

    1.a) Voting rules on ratifying a standard.

    Past ISO standards were approved by consensus, so it is little wonder the Byzantian ‘break the deadlock’ rules had to be invoked.  Some blogs have recounted theory on how documents say it works, but obviously this process had never been tested before and arguments abound.

    1.b) Failure to meet needs

    It has been argued that OOXML fails to meet its own criteria for judgement as a standard. i.e.: One cannot use this document to create text that will guarantee proper compatibility with Microsoft Office.  This immediately bogs down in ‘interpretation’.

    1.c) Failure to provide required documents

    It has been noted there exist rules requiring a release of all changes in a final document after the BRM meeting.  This document did not arrive in time for the vote and still hasn’t arrived.  Hopefully there is a rule allowing ISO administrators to ignore this fact as needed to equate editor instructions with a final document.

    1.d) Suspicious Rules Adjustments

    At several points, ISO acceptance rules were changed in a manner that allowed OOXML to get the support it needed.  I consider this a wash.  Nobody ever had to push a doc through this way before, so of _course_ they had to adjust rules along the way.  Remember, ISO was built for consensus standards, not voting one in.  Such irregularities are to be expected the first time they had to put these rules in practice.

    2. Corruption of Standards Bodies

    2.a) Stuffing of technical committees

    The explosive growth in standards committees worldwide with MS affiliates and partners is suspicious in the extreme, but isn’t against any rules.  It merely discomforts the people who worked there before the influx.

    2.b) Placement of Dead Weight on ISO

    There is no question that members added to push OOXML through the process are now causing trouble.  They don’t do _anything_ else!  Much work is now at a standstill because these new members fail to respond to repeated requests for input.  If these members get removed for continued apathy, their votes might wind up annulled.  I consider this a greater concern than most other problems as it directly challenges the very existence of ISO as a working body.

    2.c) Negating Input from Technical Committees

    Several instances have been noted of Technical committees that normally provide the recommendation on such standards being slandered, ignored, sidestepped, or removed from consideration.  This may look bad, but in the end the ISO only responds to the official NB and any internal irregularities are the responsibility of that country to deal with, not ISO.  I don’t foresee a challenge here succeeding.

    2.d) Bribery

    Currently the few instances where technical partners were given concrete considerations in the form of discounts or favors are too few and too difficult to prove.  Like 2.c I see this as irrelevant to ISO unless it resulted in the NB representatives themselves being challenged on this score.

    3. Trade Restrictions

    3.a It has been noted the WTO has rules about what is allowed in the form of standards to support competition worldwide.  The passing of OOXML to favor a single corporation in favor of existing standards could cause trouble.  I think however that MS forms a strong enough lobby worldwide to keep this conflict from being acted on.

    4. Anti-Trust

    4.a Much has been made of the fact a convicted monopolist is allowed fewer actions than other industry players.  Activity on OOXML may raise eyebrows, but seems unlikely to change votes on the ISO directly, leaving the standard intact.

    Overall.. MS has a good chance of keeping OOXML as an international standard, provided that they prod lazy partners to _continue_ working on the _other_ standards that ISO is supposed to be completing.  Failure to do so is likely to earn ire at the highest levels in the ISO administration.

  12. Wayne says:

    Jason,

    Here’s Andy Updegrove’s most recent blog posting, if you haven’t read it yet you should:

    http://www.consortiuminfo.org/standardsblog/article.php?story=20080405104109438

    Wayne

  13. Brett says:

    So,

    You’ve laid this whole thing down like this type of "To-and-Fro" legal style of rule and process jigging is a normal part of the ISO Standards Ratification.   All this talk of subversion by IBM and how "They Did It Too" so it’s all good is the biggest load of tripe I’ve heard.  Can you point to me any other ISO DIS that didn’t have Microsoft involved but had NB’s stacked by a single vendor’s motivated interest followed by as much quantifiable opposition in these NB’s particularly by the long running memberships (most notably the academic and government organisations) and so much vocal opposition in the form of legitimate technical concerns that still exists in a barely passed standard that we’re still yet to see?

    Just one??

    I have no doubt Microsoft will prevail in this standard ratification and likewise the next million it’ll choose to pseudo-standardise after this one.  It’s just that it won’t be implementable in any true open way in it’s current form.  Even if it weren’t so open to interpretation, the patent encumberances are just unforgivable. I’ll also not hold my breath on seeing Microsoft Office Anything ™ fully implement any current ISO version of it’s OOXML Ever though I’ll pleasantly retract my statement if I’m ever proven wrong.

    A bit about me:   I don’t do anything for anyone involved in this mess, I work for Aussie Defence and only know about this largely because of the open source stuff we’re working with at the moment.

    Don’t mind my Australian spelling please, it isn’t wrong where I come from… 🙂

  14. Fiery Spirited says:

    >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.

    You claim IBM raised most of the opposition. Yet there was 3500 comments on the DIS text with 1100 unique issues raised, that means an overlap that is at most around 30%. WIth IBM using the same objections in many countries we can with minimal effort deduct that IBM must have a played a pretty small part of the opposition.

    The 90% overlap Microsoft are so fond of retelling is caused by a NB that failed to do critical review besides the issued raised by IBM.

    If Microsoft had been correct about IBM causing most of the opposition to OOXML the percentaqe of unique  concerns compared to comments would been far lower.

    Btw the reason so many issues was "resolved" at BRM are that limited time prevented that Ecma:s suggestion would be examined. Every country could only raise one or two dispositions  for discussion. During the final block vote most countries abstained, but a few voted yes on everything…most probably more would have voted no if they had time to hear the numerous objections.

  15. When you tell a lie often enough, it takes on a patina of truth each time it is uttered, and after a

  16. When you tell a lie often enough, it takes on a patina of truth each time it is uttered, and after a

Skip to main content