Open XML and ODF – What Do You Want To Encourage

I have been traveling through Europe this week talking to people about interop generally, and with many people about Open XML. I am really encouraged by what I’m hearing from people as they think about the role of formats in the broader discussions of innovation and openness. I have been asked repeatedly to bring the debate on open formats to core issues.

As anyone who is close to the Open XML / ODF industry fracas is aware – these two are proxies for much larger issues. Product competition is clearly the most obvious large issue as the debate regarding the formats so clearly highlights the compeittion between MS, SUN, and IBM. Ok – fine, I think we all get that. But how about innovation? How about the benefit of ALL XML-based formats on openness?  Where does this leave us?

The first top-level issue is competition. Many people have commented repeatedly on the fact that Microsoft Office, OpenOffice, IBM Workplace, Adobe Acrobat, and other office productivity applications are in competition. The history of the creation of their document formats, and the representation of the various products’ features in those formats is the most obvious motivator for the contentious nature of the ISO balloting period under way right now for Ecma Open XML. 

 The second top-level issue is innovation. The encouragement of greater innovation is the goal of all governments and private organizations. I have not met anyone who is satisfied with the idea that software development has reached its zenith. Are the developers of OpenOffice going to stop at its current version, seek to add no new functionality to the product, seek to limit its growth based on the representations available in the current ISO-ODF? I think not. In fact, OpenOffice continues to be improved as a product and I’m pretty sure they are not going to be satisfied with just copying what other office suites have done in the past. in fact, it would seem essential that they must differentiate or fail. Also, just look to the fact that the OASIS working group has already put out multiple revisions to ODF as that piece of the puzzle continues to improve as well. The fact is, office applications, and their formats, are part of a dynamic innovation environment rather than a static one. This one argument alone is the single most important reason to me that the argument for a single document format is flawed. I want to see Adobe push PDF further. I want to see the Ecma TC-45 folks push Open XML further. I want to see OASIS ODF working group to push ODF further. I want to see China push UOF further. In short – I don’t want to see an innovation dead zone for application or format development because of the desire to drive to a least common denominator for formats.

The third top-level issue is around the encouragement of openness. I heard a few people this week be very eloquant around the idea that it is better for society as a whole to continue the progression towards open formats. The points made were not the deep details discussion as to the process of one standards body vs. another – it was about the top-level point that all of the activity from the major players around document creation are leading to greater openess for document formats. This is an increadibly positive trend and should be encouraged overall. This point has been echoed to me by universities, national libraries, national archivists, global multi-national corporations, small services providers, and many more. National bodies are being bombarded with endorsements, comments, criticisms all realating to Open XML and the upcoming close of the JTC-1 balloting process. I think that in all of the politics – a very real issue stands to be lost. We should all be in favor of increased openness for file formats. There is a good reason that Microsoft did not vote against ODF at any point, and in fact voted in favor of its adoption by ANSI recently – we agree that this trend is a good thing. We started working on XML-based file formats for MS Office back in the late ’90s and continue with this work today.

I recognize that the details of a discussion are often the critical factors in understanding its implications, but in this case the macro issues are so very important and are being washed out in the endless back-and-forth on specific points. I’m often find myself in the details conversation, just look at the leading technical community blogs you’ll see that there is no shortage of commentary out there. But, this week a number of people asked me to bubble the issue up to the real issues that matter. I think it doesn’t get more basic, or more important that this – competition, innovation, and openness.

Open XML, ODF, PDF, UOF…all good things. Advocating against any of them becoming an ISO standard is inconsistant with the idea of promoting competition, innovation, and openness.  


Comments (16)

  1. Luc Bolen says:


    If you need an example about how 2 standards for the same task is "inconsistant with the idea of promoting competition, innovation, and openness", have a look at the following story by David Berlind : "Tragedy: Commercial developer reports 35% of coding time spent on IE/FireFox incompatibilities"

  2. Daniel says:

    One of the biggest issues regarding OpenXML is patent encumbrance.  Could you please go into detail regarding people’s concerns about that?  Thank you!

  3. jasonmatusow says:

    Daniel – there is absolutely NO issue around patents for Open XML.

    Microsoft originally posted Open XML under a Covenant Not To Sue – and then amended that with the better Open Specification Promise. (

    What it all boils down to is an irrevocable, legally-binding promise that we will not sue anyone who builds an implementation of Open XML based upon the Microsoft patents necessary for that implementation.

    Let me know if you want me to comment further.



  4. jasonmatusow says:

    Luc – I’m afraid I don’t see the connection in your comment to what is happening in the world of ISO/IEC JTC-1 standardization. There are endless examples of incompatibilities that have arisen between commercial implementations of standardized technologies. That is the very nature of competition, and it is part of the idea that you want separation between the process of standardization (the agreement of multiple parties on technical specifications) vs. implementation (the actual building of the technology based on the spec – by anyone no matter if they were part of the standardization process or not).

    Having multiple standards in a given technology space is not uncommon, nor has it been shown to dampen competition or to slow development. That is not to say that it has not been the cause of frustration or inneficiencies, but it is hard to argue that having similar standards has acted as a brake on innovation. The marketplace has a funny way of being far more important than the standards themselves. OSI vs. TCP/IP comes to mind.

    In the document format arena there have been many standards (some market, some industry consortia, some international), yet that has not slowed innovation nor created a situation where if I have docs of one format on my machine, others won’t open. Coexistence is possible, and competition/innovation continues to thrive within the applications that use the formats.

    I think that openness of doc formats is a good trend – not just one format – but any doc format. Translation is the answer to the differences that exist between them.

  5. Massachusetts is now in a public review period of their updated ETRM policy that moves Ecma Open XML

  6. Hey folks, sorry for not posting in awhile. I was initially going to be on a two week vacation (starting

  7. Hey folks, sorry for not posting in awhile. I was initially going to be on a two week vacation (starting

  8. Massachusetts is now in a public review period of their updated ETRM policy that moves Ecma Open XML

  9. Daniel says:

    Hi Jason, I think a concern is that the more that MS’s OOXMLtakes hold, the more likely MS is to claim that there are patent violations for various programs that process it and depend on it.  I don’t actually understand what the statement "builds an implementation of Open XML" means– couldn’t that mean anything you or I wanted it to mean?  Doesn’t Novell’s version of now implement MSOOXML?  Does that mean that everyone who contributes to is free from being sued, or being told that they are at risk of same, for their work?

  10. jasonmatusow says:

    Thx for the comments Daniel –

    I am not a lawyer, and if you have legal questions about the OSP, I’d recommend you ask one. (ah, disclaimers….)

    As to what the scope of the promise is, the OSP states that it covers:

    "making, using, selling, offering for sale, importing or distributing any implementation to the extent it conforms to a Covered Specification"

    Keep in mind that legal text like the OSP, a license, or any such document will be limited to the spec itself. It is not good if the terms are overly broad in scope as things get very messy. I talk to many lawyers about this type of thing and it is always better to have things narrowly scoped so they are clear. In the case of the OSP – it covers:

    "patents that are necessary to implement only the required portions of the Covered Specification that are described in detail and not merely referenced "

    I understand what you are saying about the programs that process and depend on it. Yes, we do have patents on the features/functions/UI etc. of MS Office – but that is separate from the file format. The folks at Novell, or anywhere else for that matter may build a partial or complete implementation of Ecma Open XML and be covered by the OSP.

    I am not being purposely thick – I konw that still leave open the question about the tools etc. you raise – but it is really important to keep things separate when you are talking about legal issues.

    Our intent is to see broad adoption of OSP-covered specification. I think all of our actions and released over the past 12 months for interop should speak volumes about what we are doing.



  11. Daniel says:

    "It is really important to keep things separate when you are talking about legal issues."

    I don’t know if it is, though.  I think Microsoft’s intentions and likely future behavior in regard to the format should play a big role in the decisions governments and businesses make. MS might like to make the argument that OpenXML is just as open as ODF, but it would be much harder for any of  Sun/IBM/KDE/Gnome/ to pull a fast one with ODF (i.e., extending the format in ways that are then legally/technically incompatible) than it would be for Microsoft.  This is because the main justification for use of OOXML is compatibility with MS products.  (I venture that very few people would choose OOXML based on the merits of the format itself, except perhaps for some limited areas such as parts of the spreadsheet specs).  Nobody will be coding their apps with the primary goal of being compliant with OOXML– rather, they’ll be coding to be compliant with MS’s own implementations, which may or may not conform to OOXML, or may extend OOXML in legally or technically incompatible ways, as MS sees fit.  While it might be possible for other companies to do something similar with ODF, I think it would be much, much harder and less likely that they would succeed.  Given the histories of the players, I also think it would be much more likely for MS to try something anti-competitive through legal or technical maneuvering than it would be for Sun, Google, KDE, etc.  In this vein, do you have any comments on Andy Updegrove’s comments at


    "I think all of our actions and released over the past 12 months for interop should speak volumes about what we are doing."

    That’s exactly what I’m worried about.  "Interop" for Microsoft always seems to need some sort of legal catch– it never seems like it’s just interop for the sake of "rising-tide-raises-all-boats".  In particular, I’m thinking of the recent issues with Microsoft’s patent issues against Linux, as discussed in this article:

    Do you have any thoughts on either my concerns or those raised in the Updegrove article? Particularly in regard to the potential and/or likelihood that Microsoft will eventually try to use proprietary extensions to limit patent-free interoperability?

  12. jasonmatusow says:

    Daniel –

    I can’t speak to the future for any of these formats as, like everyone, I just don’t know what direction things will take. I do know that we have committed to implementing Ecma Open XML in this and the next version of Office – so given our history of product releases, and our support policies, that means that we will have a supported version of the Ecma standard in a widely-used product until at least 2017. Ecma is the maintainer of the specification, and we are one vote of more than 20 in TC-45 which will do the maintenance work. If I am not mistaken, there are approximately 9 orgs voting on OASIS ODF working group at this point. They are both standards, they are both maintained by working groups, and both are subject to the realities of implementation when it comes to how customers will experience using them. If I am not mistaken, there has been much discussion over the fact that IBM’s implementation of ODF for Workplace is not the same as what Sun and the OpenOffice community are doing for OpenOffice.

    As for people coding to the Ecma Open XML spec – we are already seeing this happen. There is significant economic incentive for organizations looking to offer functionality that works with MS Office, or provides migration paths, or sucks data from Office apps into non-office productivity apps etc. I think this space is going to be very interesting, and all parties have the same motiviation for working with the spec with interop in mind.

    I will comment with a top-level posting on Andy’s comments. He is a really bright guy, and I need to make sure I am equally thoughtful in my response.

    I also think your comments on our interop approach should get answered, but it is July 4, and I’m going to go play cribbage with my wife. 🙂

    Thx Daniel.


  13. JR says:

    Why can’t you just join to an existing standard? Why do you always have to mess around re-inventing the wheel? You already broke the internet and it’s taken around 10 years to get to a point of stabilisation (imagine how the internet had been if there had only been one HTML standard to follow 10 years ago..), now you wanna break the opportunity of a perdurable, stable, archivable, file format?

  14. jasonmatusow says:

    Hi JR – sorry it took so long to respond, I was on a camping trip with my family.

    There are times and places where MS has joined existing standards efforts, and more times where we are the contributing party for a standard. There are of course many standards supported in our products where we were neither.

    I have a hard time agreeing with you that anyone "broke" the Internet. My understanding of history (and goodness knows, I could be wrong on this one), that MS and other players were all building custom extensions to the standard in the innovation race that was under way at the time. I think you call out a really important dynamic about standards/standardization – the implementation is hugely important. It is one thing to make the statement that MS should have just adopted ODF and avoided this whole mess – and an entirely different one when you consider development costs, product release cycles, market competition of products, customer investment in features not represented in the standard, etc., etc., etc.

    We have an open standard format. It is archivable (better technology for it too considering the native support of custom schema), it is stable, it is maintained by an independent group of interested parties, it is performant, it is full featured, it is backwards compatible, it is translateable…

    We have zero interest in breaking ODF – just look at the record on the ANSI vote, our OASIS votes…

    To the broader question of MS involvement in standards, and how we deal with this macro issue overall – that is one that takes more thinking and a long-term approach. I will blog more about this.

    Thx again for the comment.


  15. After more than a year of writing about standardization of document formats, we sit on the verge of the