Countries Are Saying Yes To Open XML


***Updated April 2*** – ISO has just confirmed the approval of IS 29500 in a press release. ***Updated end***


It has been a few days since I’ve had a chance to blog. I’ve been on the road and that tends to make it harder to find the time to put up posts. I have been pleased to see the fact that countries around the world are either sustaining their “yes” votes or moving to “yes” from either “abstain” or “no.”  Again, that is the point of the BRM process (not just the 5 days in Geneva, but the whole 5 month process) – address the concerns raised so that the spec is improved and move on to maintenance (further improvement) over time.


I am cautiously optimistic at this point that the final outcome will be the approval of DIS 29500 as an ISO/IEC standard.


I have to say, it is impressive how much diligent, thoughtful conversation is going on around the world about Open XML. This is based on the fact that technical work on the comments from the Sept. 2 ballot has resulted in a high-quality standard .


Another key factor is the fact that people recognize the broad use of Open XML in the market as seen by the hundreds of independent implementations of Ecma 376. And the literally thousands of customizations of Microsoft’s implementation in Office by independent software providers and services providers. Good standards are used…and are used broadly. So the big question about the future of Open XML becomes one of future work on the spec. The reality is, it is in the best interest of the industry and users of implementations of the specification to have DIS 29500 be maintained through the ISO/IEC processes rather than at Ecma alone. (ODF editor Patrick Durusau has made this point as well.)


I had a conversation recently in Japan that reflected this point. There remain concerns in Japan about future interoperability issues particularly for a single Japanese software provider. Okay, valid point and one that is best addressed by having a strong voice in the future of the spec – which will happen through their position in SC34 than if the spec were to fail the ISO/IEC process and fall back to the sole stewardship of Ecma TC 45.


I have not spent time blogging the back and forth of the past few weeks. The FUD campaign and process attacks from the anti-Open XML crowd have become increasingly shrill and not reflective of the truth nor reason. (For example the fear-mongering misdirection on the IP issues.) The whole idea of the standards process is to move forward and improve technology specifications – not rip them down. Open XML is an important document format for the industry. Driving interop through the ISO/IEC process is better for ODF, PDF, CDF, UOF, and any other document format.


Frankly, I’m looking forward to moving beyond this discussion. I started writing about Open XML 2 years ago and have watched the progression carefully. Open XML has been about moving forward, brining the promise of XML into the Office ecosystem and well beyond it through standardization. I am amazed that the anti-Open XML crowd have spent so much time trying to stop something vs. build up their own technologies. (Miguel de Icaza wrote on this recently) Imagine if the time/money put into the anti-Open XML campaign had been put into improvement of ODF and/or OpenOffice? It would seem a better investment.

Comments (17)

  1. Andre says:

    I believe that an approval would not stop the discussions about the legitimacy of the format and in summa it would turn out to become less beneficial for your side than you might expect.

    "The reality is, it is in the best interest of the industry and users of implementations of the specification to have DIS 29500 be maintained through the ISO/IEC processes rather than at Ecma alone."

    We both know that nothing hinders Microsoft to continue the ISO/IEC process. No party will get any advantage when ISO members adopt the half-baked specification now. And for your partners the party would end quickly.

    A DIS 29500 with a slight majority due to P-supporters as Jamaica and Cote d’Ivoire is poised for adoption by the public sector.

    Furthermore your posting reflects an arrogance towards those parties that worked very hard to improve the format in a process that is designed to make these improvements damn difficult. It is not campaigning against the standard but rather your campaign for the adoption of suboptimal results that is detested.

    All we got in the BRM process is a direct result of the disapproval vote in September. You worked hard to avoid it although you knew that the format had issues. The BRM process is a proof that disapproval was the right choice and your campaign for adoption was wrong.

    The same applies to the vote now. Only disapproval would make your company to get real and turn out beneficial for all parties including your own company. It is a negotiation process and only hard negotiations work, no other party benefits from presents to your side.

    Your company does not care much if it damages the ISO institutional framework. Whatever you decide to do in an immoral fashion, it would backfire: a homostasis of the political economy underlying the ISO process.

  2. lori says:

    I want to correct you:

    "Imagine if the time/money put into the pro-Open XML campaign had been put into improvement of ODF and/or Microsoft Office? It would seem a better investment."

    Dual standards add costs to consumers.

    VHS Vs Betamax, remember?

  3. jasonmatusow says:

    Andre –

    I have blogged before, and stated in public panels, etc. that I’m appreciative of the work done by ALL parties to make the specification better. IBM is the single largest contributor of comments to improve the specification. That is very helpful to all implementers of the spec – full stop. To ignore the fact that the work done on the specification has come from literally hundreds of other organizations is really a testiment to how important the specification is in the first place. People have wanted the document format for Microsoft Office to be an open specification for a long time – now it is. The fact that the specification was worked on and improved by so many hands is a good thing – it is about community working on a technology – so I’m not understanding your argument.

    "All we got in the BRM process is a direct result of the disapproval vote in September." RIGHT, right, right, right, right. I am very confused by your problem with that. The process did exactly what it was supposed to do. It made the spec better and now you are seeing national delegations recognize the value of that work and the improvements to the specification.

    Microsoft has shown nothing but respect for the ISO process and has stuck to the letter of the directives. There is a reason that you are seeing the long-time standards professionals (think Jelliffe and Durusau as two good examples) who were against the specification orginally, come out in favor of the work being done. The teams at Microsoft and in the TC 45 membership companies have done incredible work to be responsive to the comments and adhere to the process.

    If DIS 29500 passes, I predict you will see a massive process attack by the anti-Open XML community. They will claim foul in every way possible and effectively attack the process with a mind to tear it down. I think you should look carefully at who you are accusing of damaging the ISO institution. I could make a very compelling argument that Open XML has done more to revive the reputation of international standards than any other standard before it.

    Sorry – I completely disagree with your analysis.

    Jason

  4. jasonmatusow says:

    Lori –

    There is a huge difference between the physical world and the software world. That said, I think you should go back and consider networking protocols for a moment. The fact that there were competiting standards were a VERY good thing for everyone. Pretty sure that Mr. Cerf is making a hypocritical argument when he says that there should be only one standard…his baby (TCP/IP) would not have had a place in the world if it was really a question of first to ISO, first to be right.

    The market has a very strong voice in all of this, and that is particularly true in the physical world where the cost of entry of production is so high. The adoption of specific standards by leading companies in the physical world plays a huge role in the outcome of the standards. I’m thinking 100bT vs. 100VG here. How about HD/Blue-Ray? The software world is a bit different. Translation makes for a very different discussion. How many image formats are out there today? Is that damaging to consumers in the same way that having bought tons of HD-DVD or Betamax tapes would be?

    Anyway – I think there is much to be discussed on the multiple standards front. This is not the last time you will see competiting standards…and I think that is a very good thing. I’m have something of a biased oppionino in favor of innovation myself.

    thx –

    Jason

  5. lori says:

    To comment on what Andre has said earlier, it is really sad to see how much energy Microsoft have to spend to convince people of the quality of standard. Normally a good standard developed commonly with other contributors should have entered the ISO process with a much higher quality, and should have not reinveted the wheel.

    And if the community against OpenXML is gonna be furious if the standard is adopted, it is certainly thanks to Microsoft and its attitude. People are not stupid, even if taking Business Partners in order to have enough support might be ‘within the rules’.

    Normally, for a perfect specification, Microsoft would not even need to appear in the Technical Committees of all the countries all over the world.

    A perfect spec should roll through the ISO Fast Track without even a technical comment.

  6. Andre says:

    I partly agree and IBM found even more issues with the specification. Let’s think about a disapproval scenario: You could start an review process in which all parties could openly discuss the remaining issues and review the current text. Release early, release often. Ideally this would not involve the ISO secrecy and the opposition to the format as such, beeing carried out by an electronic process. The BRM was too short but it did improve the ECMA dispositions it discussed. I was impressed about the Greece contribution to resolve the date issue and think its much better than what was agreed upon.

    Also the legacy specification of the Office formats probably needs an open process for review. I get the impression that Microsoft doesn’t fully understand its own formats. Microsoft lacks experience how to crowdsource documentation. So you would need at least a bugtracking system and the willingness to work together with external parties to improve the documentation of the formats. The legacy formats are still of great importance and good documentation would benefit the market as a whole. It could be a promising research project to actually fully document the formats.

    Then we have the missing mapping between the formats that ECMA was unable to deliver.

    Finally the DIN work looks promising to identify conversion issues tied to semantic incompatibilities of the formats. These issues should get resolved in order to enable easy and faithful conversion. In this field also ODF needs to be be improved. None of the formats is perfect. An open review process could also help to get rid off legacy.

    As of the OSP I would suggest a similar open review process by a worldwide lawyer community which would lead to an OSP 2.0, ideally managed and developed by an independent institution with the aim to get it applicable on a worldwide scale. The EU IDABC did a great job with its EUPL to get lawyers from different nations involved. Private law is still quite different on an international scale. Jurisdictions are not the same. This is e.g. why the IDABC neither did chose the French license nor American licenses nor the GPL (with its embedded ideology) but developed their own copyleft license for EU public sector software. The OSP is cast in legal no man’s land and patent indemnification models need thoughtful review.

    "If DIS 29500 passes, I predict you will see a massive process attack by the anti-Open XML community. They will claim foul in every way possible and effectively attack the process with a mind to tear it down. I think you should look carefully at who you are accusing of damaging the ISO institution."

    If Dis 29500 passes the situation will be difficult to get it adopted by the public sector as the whole process was pushed to its limits and many public decision makers got upset.

    And finally this is the whole rationale behind the ISO standardization, to get acceptance for OOXML by the public sector. You cannot sell it when you have an angry community against you and upset the standards community.

    I do believe in a sound technical review process without a commercial interference. The ISO system would of course need reform. If the specification was ready Microsoft could unplug from the involvement in the whole adoption process and leave it to the ISO members to decide positively. I don’t believe that a party as IBM would oppose a format just because it is from Microsoft.

    The hard negotiations led to many positive developments including the opportunity to resolve accessibility issues: the Daisy cooperation as a direct response to the Canadian criticism for instance.

    The ISO system was tested. Parties involved know how it can and did break. Here the procedures need to get fixed. For instance the P membership could be determined according to actual participation and experience in standard review. If a member body contributes it gets automatically upgraded and more say in the process, a meriocratic scheme.

    In short more "requests for comments" than "killing the messenger and the message".

    "Pretty sure that Mr. Cerf is making a hypocritical argument when he says that there should be only one standard…his baby (TCP/IP) would not have had a place in the world if it was really a question of first to ISO, first to be right."

    You know very well that the agenda of OOXML ISO standardization is to undermine the competive advantage of ODF. Now, nobody hinders Microsoft to develop an ECMA standard but on the ISO level it would be preferred to have a single document standard. Having that said you need to take into account that both formats were not on the same level. OOXML is still a single-vendor format and contractors who implement interfaces to OOXML as a showcase don’t create a true community around the format. Having that said there is indeed potential in office automation. Today we want open standards, that is standards free from vendor capture.

  7. Marbux says:

    Jason said: "I could make a very compelling argument that Open XML has done more to revive the reputation of international standards than any other standard before it."

    That would be a very interesting blog article; I encourage you to write it. It is difficult for me to comprehend how a standard that does not "specify clearly and unambiguously the conformity requirements essential to achieve the interoperability," as required by JTC 1 Directives, pg. 145, could conceivably "revive the reputation of international standards." Page. 13 states that it requires the consent of the Secretaries-General of ISO and IEC to depart from the "strategic characteristic" of interoperability and so far as I am aware no such consent was granted. http://www.jtc1sc34.org/repository/0856rev.pdf

    I think it entirely regrettable that ISO/IEC:26300 OpenDocument suffers from the same weakness, but a previous violation of a core exit criteria is scarcely a valid defense to violating the Directives again. From here, the view is that Microsoft, Sun, and IBM spend far more effort talking about interoperability than they do implementing it.

  8. OOXML sucks ass crack!!! says:

    I have M$ Office 2007, it hast advantages and disadvantages compared to OpenOffice. One of the major disadvantages is the format! – I don’t want to have problems to convert one format to another… I even don’t want to be forced to convert my documents at all!!

    ODF is the format for the future! OOXML is totally grap!!

    If M$’s next office suite does not support ODF, I will switch to OpenOffice for good!

  9. pirast says:

    great, now we have another document format standard for something we already have which is > 5000 pages longer than the other one.

    why can’t microsoft sit together with OASIS and improve ODT and has to build up its own standard which is based on the proprietary office format?

  10. jasonmatusow says:

    Marbux – good to hear from you. It has been some time since we sat on that panel in Portland. My point is broader than any one standard. The discussion around Open XML has raised many important questions and absolutely garnred elevated attention to international standardization. My opinion is that governments are likely to place more weight on the ISO impramatur now than they had in the past. For software standards, consortia standards have been both more numerous and effectively more influential. The effects of the Open XML discussions will be felt for some time to come.

    Mr. ass crack!!! – I’m not sure that is the name I would choose for myself when commenting on someone elses blog (seriously….leave profanity alone. I’ll moderate it out next time). In case you missed it, we have announced that Office will support an API that will allow ANY file format to be hooked in for use and ultimately to be the default format. This was part of our interop principles announcement.

    Pirast – you are asking a question that has been answered so many times at this point….you should read past blog postings.

    Jason

  11. Till Tor says:

    How many Microsoft employees are necessary to install a light bulb?

    None. Microsofts defines darkness as the standard. ;P

    OK .. seriously i am really not sure why we need another 6000 page standard. It is inconsistend (different color tag for word, excel and powerpoint) and vage (OLE handling), dont you think so?

    Cheers

  12. Praetorian says:

    I still feel Microsoft did this just to appease growing demand for official standards by the public sector. The spec doesn’t really allow anyone else to write completely compatible software – Joel Spolsky’s explanation at http://www.joelonsoftware.com/items/2008/02/19.html is quite insightful in that regard. So it’s a "standard" that only one manufacturer can implement completely…sorta misses the point of standards.

  13. Oliver says:

    "For example the fear-mongering misdirection on the IP issues."

    Whether this was "fear-mongering" or a valid point, will be shown in the future. For now you managed to save your job – congrats.

  14. >"Imagine if the time/money put into the pro-Open XML

    > campaign had been put into improvement of ODF

    > and/or Microsoft Office? It would seem a better investment."

    Imagine if all the time/money put into the *Office "Open" XML* campaign had been spent on integrating with/improving the existing standards such as ODF? It would definitely be a better investment.

    Imagine that – if Microsoft had helped improve ODF we would now have a single, stable, mature standard which had the added advantage of being universal *and* open. Instead, because Microsoft has decided to preserve their monopoly, we are left with and argument and fragmented standards – regardless of who’s right.

    Something doesn’t add up here.

  15. I prefer ODF says:

    you company is using pressure techiques to persuade the 3rd wolrd countries …. your company is evil as allways you guys haven’t changed i hope that googles destroy your market 😀

  16. jasonmatusow says:

    Till Tor – there are multiple ISO document formats. There are multiple ISO image formats. This was true before Open XML and will be true after Open XML. The real value in this whole equation are the capabilities and quality of the apps that use the formats. I’ve been saying that for a long time and still think it to be true. This whole competitive contest has been a proxy for the real issue of app marketplace competition. (which is fine – but that is the reality)

    Praetorian – there are already hundreds of orgs doing independent implementation work – including IBM. The work is being done across platforms and for both office productivity apps and completely unrelated apps who want to use the format for other purposes.

    Oliver – my job was never in question over this issue. But thank you for your concern. I’m glad it has been approved, but truthfully, there are more important issues to be worked on than this.

    Long-time ISO Analyst – we have put serious resources against translation and interop. We don’t believe, nor will believe, in the single format argument. Check out the very next post in my blog following this one. How about DAISY? How about Adobe’s work? How about the National Library of Medicine format? How about China’s UOF? Come on – the single format argument simply doesn’t hold water. When ODF was started in OASIS Micrrosoft already had years of development going in our XML-based formats. There are fundamental design differences between the two – and that is completely okay. I favor innovation in the apps and having formats represent those innovations. ODF represents the innovation in OpenOffice – groovy. That does not mean that everyone else who is using a document format should bend their knee to ODF. They should have a choice as to what document format best meets their need.

    suribe – I think you should check out the statement from Standards Norway. They don’t agree with the assertions being made.

    I prefer ODF – no, you are seeing what you want to see. The entire process has been about listening and making changes (to the extent possible) to accomodate those requests made by ALL parties. If you like ODF – then I would suggest you read what the editor of ODF (Patrick Durusau) has to say about the process and the work that has been done on Open XML. Maybe that will help you see what is really happening.

    Thanks all –

    Jason