Open XML – SC34 Maintenance Agreement

Throughout the months leading up to the BRM and then following it, there was discussion over the future maintenance of the specification. Rather than recapitulate the results of the last SC 34 meeting, let me just point you to Alex Brown’s blog. For those of you who were not following all of the players so closely – Alex was the convener of the ballot resolution meeting.

I still hear patently untrue claims that MS controls Open XML – this wasn’t true following the adoption of Ecma 376, and is now permanently a moot argument. Alex’s blog title says it all:


Does anyone know if the commercial interests in control of ODF would be willing to take the same step?

Comments (11)

  1. cybervegan says:

    Back in 2006 this very question was answered by David A Wheeler on Groklaw:

    There’s no single entity that has control of ODF – it’s run by consensus all the way down. This is a moot point.

    There are a large number of fully compliant, full applications, both commercial and open source that implement ODF, wheras MS Office doesn’t even implement OOXML yet, and there are no other full implementations by other vendors either – OOXML is as yet, vapour-ware. And MS don’t even promise to fully implement it in Office *ever* the last I heard.

    At least that’s what they say here:

    … and here:

    … and here:,1000001161,39348275-39001068c-20091749o,00.htm

    So until there’s a real life full implementation of OOXML it doesn’t matter who controls it really. It’s a straw-man argument – MS only want the standard so they can win business that mandates an international standard file format, without really implementing it properly themselves. They know that most bureaucrats won’t look any further than "OOXML is Microsoft’s standard, so naturally Microsoft Office must support it, right?"



  2. RichL says:

    The real question is when will Microsoft implement OOXML?  Which release of Microsoft Office is slated to have the spec fully implemented?

  3. Nadi says:

    > Does anyone know if the commercial interests in control of ODF

    There are no commercial interest in ODF. It was not developed in secrecy for the bennifit of one, several companies participated and many individuals and MS was invited many times by Bob Sutor to help.  

  4. Swashbuckler says:

    Gotta agree with RichL, if Microsoft goes to such great effort to get OOXML standardized and then won’t even commit to supporting the standardized version it really doesn’t matter who controls the standard.  If Microsoft doesn’t support standardized OOXML, who will?

  5. Swashbuckler says:

    "There are no commercial interest in ODF."

    Oh puhleez!  You can’t be serious!

    There’s as much commercial interest in ODF as there is in OOXML.

  6. jasonmatusow says:

    Microsoft did commit to support the standard guys – where have you been? Also, the standard just became a standard about oh…a few weeks ago…so I love your eagerness to see it implemented, but it may take some dev time for folks to start using it (that’s just a guess on my part).

    Seriously – I think the griping about the implementation is really something on your collective part. The spec carries IP terms and technical description that enables partial implementation or full implementation. If the specification is being implemented…at all…then it is doing its job. There will be the FULL spectrum of implementations out there – the success of Ecma 376 already shows that. I realize that for some – nothing that MS does is good, and that you look at the world through polarized lenses…but come on. Have you ever used an X.500-like directory service???? Oh yea, that’s right, basically none of the big commercial implementations supported the full spec because it was flawed (if my memory serves). But I’m pretty sure that all the vendors touted the standard in their marketing of their services.

    Seriously…the "full implementation" argument simply doesn’t hold water when you are talking about the value of a specification. A standard is meant to be useful…and to have success in the marketplace. Pretty sure IS 29500 is there.

    Cybervegan – I have to disagree with you. There is a massive difference between consortia and international standard bodies. To the national standards bodies – control of maintenance of the spec does matter. Just read Alex Brown’s blog that I referenced when he speaks about what happened with ODF.

    I have always found the double standard (excuse the pun) interesting between the treatment of ODF and Open XML. Open XML can’t be in the control of a consortia…but it is ok for ODF. Open XML should have a BRM…but it was ok to ignore comments for ODF. Open XML has IP problems…but ODF doesn’t (when both are under the exact same terms for ISO). At some point it becomes a bit silly.

    Nadi – I’m with Swashbuckler (see, we can agree with each other) on this one. Nadi, wake up and smell the roses my friend.

    thanks all…


  7. Andre says:

    Jason, you are free to take part in the beauty contest. It ressembles snow white: mirror, mirror on the wall…

    SUN played fair and ODF became the fairest of them all. Of course you can make OOXML more beautiful. SC34 is no argument in favour but a fulfillment of a formal criteria. The question is: What difference does it really make to you?

  8. Brett says:

    "Does anyone know if the commercial interests in control of ODF would be willing to take the same step?"

    I hope not, ODF’s been getting great traction under the direction of a consortium of industry interests with quite a number of applications able to both read AND write ODF docs, most ODF capable apps are now tracing the proposed ODF 1.2 standard with near full implementation.  There are still "0" full implementations of OOXML, even now it’s gone as far as it can in ISO ratification.  

    Naturally the biggest concern about handing the spec over to ISO now, especially after witnessing the DIS 29500 debacle would be the uncertainty of what would happen to it… I have a seemingly unlimited number of links to demonstrate NB and expert observer concerns if you’d like me to point them out complete with summaries, I’m not sure I have to point it out but it seems that ISO has little credibility left in that regard.


  9. jasonmatusow says:

    Brett – the reason the big 2 vendors behind ODF like OASIS is that they maximize their influence on a spec that has great impact on their products. Moving to SC 34 presents them with a challenge. It was convienent to demand that SC 34 take ownership of the spec when they were seeking to create arguments that would stop the adoption of the standard, but when it comes to their own technology they want different criteria to apply. Okay – I get it, but people should see that for what it is. It just proves my overall point on document formats that it is really about the applications and commercial competition there.

    Brett – Open XML is being implemented on Linux, Windows, OS/X, even zOS. It is on mobile devices, in Free Software, and yes, part of Microsoft Office. The point I made in my comment just 2 before yours in this thread still stands. You are saying no bird is flying into outerspace while ignoring the entire population of birds who are flying in the air. Commercial, academic, government, non-office automation apps, etc. etc. etc. are using the Open XML specification to deliver real value. That’s the point of a spec – to enable people to go build real code that works.

    Andre – ODF became fairest from your perspective and I think that is grand. That is what choice is all about. The thousands of ISVs currently working with Open XML and not at all with ODF are making a different choice. I put up a blog post a few days ago with a long list of case studies – it might be educational to read through a few to see what folks are doing with it.

    As for the SC 34 question – take a moment to consider how a national body feels about the influence they can exert on the future of ODF through SC 34 vs. through OASIS. There is a reason they prefer to see maitenance of international standards happen in their SCs.


  10. Brett says:


    The Big two vendors might like to steer the formats for their respective products but by the same token they don’t introduce incompatibilities between each others’ products nor impede other vendors from using their standard formats via vague descriptions of tags nor proprietary and undefined addons and uses nor licensing and patent encumbrances.  If you weren’t around when the whole patent thing started to become an issue a couple of years ago, several big players (including these very same big 2) were subject to the same scrutiny Microsoft now gets but they all stood up to pledge protection and covenants not to sue for GPL’d software usage.  Microsoft is expressly the opposite.  That’s not to say they’ve announced they’ll pursue violations against GPL software and developers vigorously but they’ve especially avoided saying they wouldn’t. Think fear, uncertainty, doubt… a well known hostile strategy indicator.

    I can see why these other players don’t want to let go of ODF now, especially since Microsoft has amply demonstrated it’s sphere of control is incredibly pervasive.  It’d be an easy step to conceive a working standard like ODF could be subverted with minor nuances that cause creaks in the same way an oversized, confusing, contradictory and non-working OOXML standard was crashed through ISO by Microsoft and it’s rent-a-crowd. Until my faith was restored in the system that ISO employs, I’d be reluctant too.  It doesn’t matter which body manages a standard just as long as the standard is maintained in a fair and openly transparent way.  So far, OASIS has done a better job with ODF than MS, ECMA and ISO has done with OOXML.  

    I’m aware everyone is trying to get OOXML working but in stark contrast to ODF it doesn’t seem like OOXML has any near compliant read and write application in the wild yet, there’s been versions of OOXML kicking around for a few years now so saying it’s only just been approved doesn’t really hold water.  

    As for the thousands working with OOXML out there, many of them are in the same boat as ThinkFree, the Director even commented on their blog that he really wants to implement ODF but they had to work with Microsoft’s OOXML and didn’t want to cause problems for themselves because they needed MS support. It comes back to that sphere of control thing, Microsoft couldn’t use discounts to maintain their court verified monopoly so when companies like Dell did start openly selling Linux, they paid the same price for their windows licensing however they conveniently no longer qualified for their MS marketing subsidies. there’s a way around everything when you comply to the letter of the law and not the spirit.

    Anyway, some questions if I may,

    1 – Is there or will there ever be anything like full-featured control documents in the works to ensure proper compliance and testing?

    2 – Does MS have a timeline for the introduction of said compliance to ISO-OOXML in MS Office?

    3 – Will MS be marketing any form of ISO compliance in it’s products before it actually produces fully compliant files?

    4 – Any reason why (after the big hurry to fasttrack DIS-29500 to an ISO certified standard) there is no real rush to get a working committee together to field and address technical concerns over IS29500?

    5 – Why is Microsoft so unwilling to see broader adoption of OOXML in GPL’d software?

    6 – When will the NB’s that recently voted to approve OOXML in the BRM actually see the full standard that they voted in?

    I ask this last question because Microsoft Employees seem to be referring to the completed text in a number of comment posts and as far as I can tell, nobody has the final text yet, I’d love to see it.

    I’m sure I’ll have more questions but too much in a straight line would probably cause me to meltdown.  On another note, I am interested to see what binary blobs are created in MS-OOXML and MS-ISO-OOXML and how accessible that data would be.  I recently found to my surprise I was able to recover an image in perfect fidelity I thought had been lost for some time from a pre ODF doc.  I’d created an ODF version of the doc and unpacked it to see what the doc looked like against the spec.  I’m sure that ODF and OOXML wouldn’t be the only saving grace capable of such a deed but it sure is easy to stripe out particular elements in a finely granular way if I wanted.  Binary blobs would be a hinderance and I’m interested to find out how Microsoft intends to work on clarifying and opening their OOXML formats in MS Office.


  11. Brett says:

    # jasonmatusow said:

    "Brett – Open XML is being implemented on Linux, Windows, OS/X, even zOS. It is on mobile devices, in Free Software, and yes, part of Microsoft Office. The point I made in my comment just 2 before yours in this thread still stands. You are saying no bird is flying into outerspace while ignoring the entire population of birds who are flying in the air. Commercial, academic, government, non-office automation apps, etc. etc. etc. are using the Open XML specification to deliver real value. That’s the point of a spec – to enable people to go build real code that works."

    Help me out here, I’m struggling to find any OOXML apps that read and write OOXML of any type, be it ECMA-376 or ISO Pre/Post BRM.  There are some that read OOXML okay but none are 100% and more apps are actually just chasing MS Office Output conversion, not OOXML.  the likes of IBM for example, they can’t care about the OOXML Standard however MS and pro-OOXML exponents were quick to point out IBM’s support of OOXML despite IBM denying this publicly.  IBM are touting support of MS Office Docs for obvious reasons in this competitive environment. It sounds like you’re insinuating all applications that are following MS Office Doc capability are by association OOXML apps? This has been the slippery Holy Grail of all Office Suites since the dawn of MS Office Dominance and not in any way an endorsement of support for OOXML.  OOXML in it’s current form isn’t altogether used by Microsoft anyway and the fact that other larger players are trumpeting compatibility with MS Office Docs as opposed to OOXML is if anything, actually telling of a lack of confidence in OOXML ever becoming a mainstream format for MS Office.

    Anyway, the ODF Birds seem to be flying in space just fine and the OASIS format is delivering credible competitiveness to the market that’s been sorely missing for some time, there’s plenty more real value in that too.

    OOXML seems like a "nearly there" fork of MS Office Output that isn’t really supported. Some applications are trying to get compliant filters but just can’t work out how to get it going quite right because there isn’t an existing app that actually does produce this ISO (and earlier ECMA) standard version of OOXML.  It’s something like having a special type of PDF standard that doesn’t work quite right in practice… It’s far easier to chase MS Office compatibility because it is possible to see how it works when you type something in and save it out.

    A note on the power of NB’s  Quite a number wanted to submit additional concerns after having found more problems during the BRM and even existing comments of decent substance that would require major work were almost uniformly pulled aside by Microsoft or ECMA bodies and strongly urged not to raise them and that it’d be assured a priority in maintenance, especially where these concerns were large technical showstoppers. Not a lot of NB’s were happy about it and certainly many didn’t feel they were heard let alone had influence.  The way in which thousands of comments were arbitrarily accepted by mass vote without discussion let alone consensus is also a sore point. By contrast, no NB has indicated any problem with the way OASIS has conducted management of the ODF standard.  You’ll probably find the reason that they wanted SC 34 to manage it is to cede some control away from ECMA and Microsoft so it might have a chance at becoming workable and not so unwieldy.  I see though that ECMA and Microsoft influence is still prevalent in this working group so it’ll be interesting to see how it goes.  

    I’ll have to go back through the case studies you mentioned in the previous article but when I looked through a quick handful, it seemed to me like it was more a technical showcase of Microsoft products and not a demonstration about openness and transparency/control of one’s own data and application choice capability.