Product Competition Driving File Format Debate

In the past few days, the rhetoric on both sides of the file format debate seems to have heated up. The ODF Alliance has been launched and it is a good thing on the part of the developers of the software that will use that format. What seems to have really gotten people heated up is the fact that we had the temerity to point out that the software they are seeking to advance is predicated on commercial interest. What seems even stranger to me, is that the blogs of those most vocal in this discussion (Simon, Bob) are incensed that we are driven by commercial intent as well.

….sssshhh…we’re huntin’ wabbits…I’m going to let out a really big secret in my blog tonight. Microsoft wants to continue to deliver commercial value in its Office products, and then we are going to…wait for it…sell that product on the market. If we are really sneaky, I think we might try to do the same thing with Vista. No one will see that coming.

On to other painfully obvious points that seem to be getting glossed over. SUN has economic interests in seeing StarOffice and OpenOffice become successful. IBM has similar interests in seeing Workplace become successful and, in the interim, may even continue to push SmartSuite for a while. How they choose to monetize the success of the software (consulting engagements vs. software sales, which both mean booking revenue) is based on how they are structuring their businesses.

The fact is, Microsoft continues to invest enormous R&D into the improvement of our existing products while adding to them innovative new technologies or full products to augment value further. Word, Excel, and PowerPoint continue to be improved on a feature set level while the addition of product such as OneNote increases the value of Office as a whole. In fact, the investment we made in the TabletPC – hard core R&D behind the screens, pens, ink as a data type…go even further to augment the value in the applications above. I can redline documents in digital form with a stylus in one hand and a cup of hot tea in the other. Office is enabling me to work in a way that is more suitable to my style and, as a user, I care more about the this kind of feature than about the format in which that file will be stored. Moreover, is it a reasonable assumption that I would want to have the Office ink features supported in future doc formats so that my work today is accessible tomorrow?

I’m not an expert in the OpenOffice world, (this is a genuine question and not meant to be snarky) are ink annotations supported in the product? And if so, are they supported in ODF?

As for the ODF Alliance, let me say it again – I think it is a good thing they are building an ecosystem around the format and have people building compelling technologies. That is what competition is supposed to be about. For both SUN and IBM, their products are still maturing – and I have no doubt that they will continue to do so. It is in their economic best interest to see ODF be successful as it presents an alternative to the Microsoft formats. For the market that is a good thing – markets are improved with a broader range of choice. The reality is that the conflict over the formats increases awarness of their products and creates an opportunity against the incumbant.

The fact that we are listening to our customers (basing our formats on XML, moving the format into an open standard, working with industry partners and customers on the spec, supporting .pdf in the product as well as numerous other standard formats) and putting forward arguments that favor our products seems obvious to me and not at all disingenuous. Wrapping yourself in a the flag of openness and then being upset when it is pointed out that you too have economic interests that drive your behavior seems a bit less direct.

Simon called me a sophist <sigh>, I’m just glad it wasn’t worse than that.

Comments (16)

  1. Nektar says:

    I think that you are a bit unfair. Firstly, ODF supporters did not accuse you of saying that they have economic interests behind their format only but also of other more important things that you do not address. These things include the fact that some MS employees have accused the alliance that they want to push one single format in place of everything else including HTML and PDF, which is clearly not the case, given the fact that ODF is for editable documents for desktop usage. In addition, calling the ODF Alliance as Sun, IBM and its friends, is not the best way of presenting Microsoft’s arguments. After all, the ODF supporters have simply responded to the Microsoft arguments voiced before, so what did you expect them? To stay silent? In their response the economic interests argument is not even the main one and you seemed to have left out the other ones.

    However, the retoric here is not what customers need. Customers need answers and a person at your position cannot say that he does not know. You should know and if not please find out:

    1. What was so special about Office files that a new and incompatible Open XML format had to be proposed instead of working with the already standardized Open Document Format. Please be technical here, we are not persuaded by simply saying that Microsoft had a legacy support to accommodate in the binary format. Why couldn’t you extend the already standardized ODF should be address? And addressed fully and clearly, not by empty arguments about economic interests, etc. Be technical.

    2. Why didn’t Microsoft get involved in the Open Document Format standardization efford and had stayed out of that? Why have you chosen to provide XML formats but at the same time choosing not to support and help make compatible with your Office software ODF? Do you now admit that not taking part in ODF standardization was a mistake? A mistake that your Open XML submition to be standardized in 2005 tries to aleviate?

    3. Why although on the one hand seemingly providing open XML formats in Office 2003 have you created and not tried to correct even after 3 years the issues surround the licensing of those XML formats? Why hadn’t you addressed these licensing issues from 2003? Why even today don’t you provide clear guidance if your Open XML license is compatible with other open source licenses eg. the GPL? You might not be a law expert but Microsoft has plenty. Don’t you realize that licensing confusion and lack of clarity in the open source world surrounding your XML licenses in Office 2003 and 2007 creates more uncertainty than the benefits of those formats. It would have been much better to come outright and cleared things up.

    Thanks for your time.

  2. Simon Phipps says:

    Hey Jason – I’m not incensed about that; I was merely pointing out that it’s a bit rich for Microsoft to criticise others for having "commercial intent", which is how you were quoted in The Register. Could it be you are projecting your frustration onto me… 🙂

  3. StefanW says:

    this is so tiresome – the altruistic vs. the evil empire…

    i’m more confused by simon’s attempt to deny the existence of compatibility issues between formats in the same article. this is so obviously wrong when one considers the semantics of the format, not just the syntax, i wonder how he justifies making such remarks. (oh yes, OO is .DOC compatible. i bet there are no issues with converted documents. i also bet that word macros, were they supported in OO, would have no problems working with documents that have been converted from a semantically different format before. oh my)

    also funny how he seems to think about XAML being a HTML replacement. so much disinformation in the "honest camp"… do they consider swing a HTML replacement? hardly.

  4. jasonmatusow says:

    Hi Simon –

    Never projecting onto you. You know I don’t get angry about this stuff. Please don’t misunderstand about my pointing out the commercial intent piece as frustration.

    The "open" standards discussion is following the same flight path as the open source discussion of 5 years ago. Commercial motivations are seen as somehow dirty or underhanded. I’m not sure where along the way it became unsavory to have a business model, but it has. Maybe it is the endless run of movies where the villian is the big corporate exec who has completely lost his moral compass.

    SUN and IBM are doing a great job of driving a marketing strategy with ODF at the lead to provide a fertile environment for their business strategies to grow. I just have a hard time swalling the holier-than-thou rhetoric that is used to justify the effort. Particularly from IBM – this is the company that invented vendor lock-in. In fact, for those of you out there with IGS employees on-site, when was the last time they said "we’re done, we’ll be heading out now, our desks will be avaible for someone else next week."

    There are legitimate issues to be addressed (such as doc archival and facilitating new entrants to markets) and we are absolutely interested in understanding how best to address those.

    This is why I love the blogsphere – Nektar’s comments above are great and I absolutely own a response to those questions.


  5. StefanW says:

    Nektar, i think Jason has pretty much answered most of your questions in the past.

    1) there are hundreds if not thousands of features in office. a feature that is not supported, or supported with different semantics, will bring up issues. documents will look different and be harder to edit further. i’m not an OO expert, but just look, for instance, at the various property dialogs for text frames, images etc. the positioning, the text flow. it’s absolutely beyond by imagination that any format that did not have MS office in mind could seamlessly support all of these. i don’t understand why people ask for examples all the time. just look at MS office and ODF, and i’m sure examples will turn up in the hundreds. (and then imagine a MS Word macro acessing any of the old properties from within a converted, ODF-based document.)

    2) why did MS not participate? we’ve had answers. the truth might be more politically loaded, we’ll never know, but answers were given.

    my opinion is this. MS had the choice between:

    a) struggling to get all their features supported in OASIS within the groups time frame. fighting about features that are already supported in ODF, but with slightly different semantics. fighting about features that other group members think are so specific (or badly designed) that they should not become standard. finally ending up with a compromise format that supports most of MS’s features, but still has little problems. a format that is based on the format of a smaller, but agressive competitor.

    b) just proceede with what was already started with office 2003. make sure that everything in .DOC (and family) is supported seamlessly. do it within their own time frame.

    basically, if MS chose to switch to ODF, they’d be in a bad position because they would face some of the same compatibility issues OO is facing today. (yes, i think even with OASIS participation to some degree)

    in contrast, OO support for ODF will be perfect, because it’s based on their own format.

    that would be the tail wagging the dog. i don’t see how anybode can really expect MS to do this. it’s not in their own interest, and not so much in their existing customers’ either.

    3) GPL compatibility is not MS’s call. the question is not, will MS allow GPL programs to use the format, the question is: will the GPL allow licensees (including contributors) to use the format under MS’s terms?

    MS seems to think that it’s possible, but it’s not up to them to decide at all.

    just do what you’ve asked jason to do: stop the rethoric and bring up real issues with GPL and MS’s Open XML licensing.

  6. Simon Phipps says:

    Hey Stefan:  Yes, the semantic concepts from .DOC are present in ODF as far as reverse engineering will allow – most of us estimate it in the high 90 percentile. Macros are a different issue being addressed a different way, but all you’re doing here is providing ample examples of how bad it is to be forced to reverse engineer and how valuable it would have been if Microsoft had joined the Office TC at OASIS back in 2002 instead of snubbing it. Of course there will be issues in this circumstance, that’s the whole point of wanting an open standard. But the Office TC at OASIS did the best job possible at semantic transferability.

    And XAML:  Well, time will tell, that’s just the independent advice I am receiving. Just as Jason is not an ODF expert, I am no longer a Windows expert having finally switched away from it a few years ago. If I’m wrong I’ll admit it (eventually, anyway!).

    And altruism:  Someone once said that "altruism does not exist". Everyone has an agenda. I am not being (solely) altruistic when I support open source and open standards, I seem them as the dynamic of the era society is moving to and I want to succeed with both.  There are real issues here, and each party is addressing them. But if the only issue you can find with my blog entry surrounds my view of XAML then perhaps I have a point about Microsoft avoiding these real social issues that frame the marketplace for both customers and competitors? Maybe it’s time to address those together rather than allowing the incumbent monopoly position to prevent it? There’s no reason why market success has to be a dichotomy between altruism and evil.

  7. Simon Phipps says:

    So, Stefan, you’re saying it’s OK for Microsoft to use a format that gives it a 3-year market lead even if competitors choose to adopt it, whereas it’s wrong for others to propose a standard that has an implementation all can use in their products because they might face less of a lag in time to market? This would be the "monopolies are good" argument?

    If Microsoft had joined the OASIS group the format would have changed and we would all have faced the same time pressures. As it is, by leaving the matter until the release date for Office 12, Microsoft has guaranteed there is no all-round-equal solution.

  8. Nektar says:


    1. Were the technical obstacles so difficult to overcome or were there political reasons behind the decision not to try and adapt Open Document Format? Or was it simply that Microsoft had miscalculated the success or the demand for an open XML document format?

    2. If IBM, Sun and another 50 companies can colaborate why not Microsoft? These other companies too have their differences both political and technical. And yet it seems that they worked them out. This situation looks even worse for Microsoft which historically has a bad record of working with and most importantly contributing to standards which do not further directly its economic interests. Other companies, given both their open standards and open source contribution have a better image. These positive and negative images might not be accurate but the fact that they even exists shows that work needs to be done by Microsoft. For example the latest fiasco if I remember rightly was with SenderId. And remember, it is always easier to blame the other parties of mistrusting Microsoft. How did Microsoft try to aleviate those fears? Which brings me to the third point.

    3. If other companies can seemingly work out their issues with the GPL and have customers and other users trust their licenses when they are used in conjunction with GPL code then why not Microsoft. By not answering the question around license compatibility then Microsoft creates fear and uncertainty. I disagree that it is for the GPL to say whether Microsoft licensing is acceptable. If a vendor wants to play nice with the community questions like the above should be addressed and addressed promptly. Look at how other groups eg. Apatche, Mozilla, etc have resolved GPL issues without leaving the question open.

    The above are not retorical questions at all.

    I am not saying that Microsoft is right or wrong. I am simply saying that more transparency and more collaboration with the rest of the world is required by Microsoft and that real answers are saught by its customers and no uncertainty should remain, let it be in GPL licensing or in why Microsoft didn’t take part in the ODF standardization. This effords should underline the involvement of Microsoft in all standards’ issues and not only this one.

  9. StefanW says:


    as for the 90 percentiles: I was not aware that the TC was actively pursuing MS Office compliance. Sorry if I got that wrong. I also don’t know how much reverse engineering would have been necessary, but if they really started at in 2002 – when was the exact time MS published its WordML specs? As soon as you have them, there is absolutely no need for reverse engineering of .DOC for semantic compatibility. Since WordML is pretty much guaranteed to be 100% semantically compatible with the undocumented .DOC format, you could reach the same level of compatibility by aligning ODF with WordML. (And there has always been RTF, which should do the same trick.)

    Besides that, I agree that MS should have documented the .DOC format long ago, but now that they are switching to an open (to whatever degree – at least that means "documented") XML format, I pretty much lost interest in this discussion. Of course, competitors might have a different angle on this too.

    XAML: let me try a quick explanation: We’re talking about Windows Presentation Foundation (WPF) here. This is a technology for building rich clients. Its primary target is rich applications, but it can also be used for browser-based apps, but here it competes with the likes of ActiveX, Swing, Flash. The fact that it’s based on a markup language is just a technical detail, it’s not markup that’s intended to be interpreted by a browser. The XAML markup will be compiled, you can in fact create the same application without a single line of markup, it just helps during development. I don’t see this targeted as a competing standard for HTML at all. Basically every statement to the contrary I’ve seen so far was based on poor understanding of what XAML actually is.

    More technical, XAML is just a way to create arbitrary .NET objects, including hierarchical relationships and properties, from XML files. In case of WPF, it’s user interface controls, but consider Windows Workflow Foundation: XAML is used here to create process objects containing activity objects.

    For the altruism debate: I agree that the general direction of open standards AND open source have advantages from a society perspective. But as soon as companies like Sun and IBM are involved, I can’t help thinking that they are really deciding their direction based on their business models and then make public statements that give the impression that they’re doing what they’re doing for society’s sake. Don’t get me wrong: This is still a good thing, but I understand how MS questions the high moral ground you’re claiming.

    What if Sun had > 90% of the office market? Would they still try to be as open as possible and support their competition? Would all the ODF Alliance members work together if not because the market confronts them with one single competitor (MS) that’s a greater obstacle for their sales than all others combined?

    My point is this: MS owns the office market. They were playing it unfair in the past, considering undocumented formats and all. But if they now open up enough to create a _somewhat_ level playing field, their dominance will hardly be ended based on philosophical questions, or on technical details like mixed content XML schemas. I think Open XML is neither too little nor too late, so why would customers want to switch? Compliance with Open XML is probably easy for ODF-based apps to achieve. If you’re right and the information model is compatible with Office, probably a simple XSLT transformation would be sufficient. Still, one format will be dominant. What’s your offer for customers, that’s worth more than 100% compatibility with the existing millions (or billions?) of existing documents, templates, macros and other solutions? What’s their benefit for standardizing on ODF, especially from the pragmatic viewpoint IT decision makers will usually take?

    The credits go to ODF for forcing MS to make this step, though.

  10. StefanW says:


    regarding your second answer:

    I really don’t know the exact process of the ODF standard, you may be right about MS having had a good chance to support ODF and dropping it’s own format completely. But after all the years of .DOC secrecy, I think Open XML is a giant step forward. OpenOffice has been able to keep up with MS Office even during the binary, reverse engineering days. So how bad can it be in the future, with the new, ECMA-standardized format?

    Given the market share, it’s just natural that competitors follow MS. It’s in ECMA now, ISO has been promised for the future. I think that’s good enough for anyone to compete with MS. They still have an advantage, granted.

    The ODF alliance is free to compete with Open XML. I don’t consider it a bad thing, never said so. But as a MS customer, I’d rather go with the MS format and see Sun and IBM follow, than risking problems with existing documents. As I said, I think it’s open enough. The political discussion surrounding it is intellectually interesting, but that’s it.

    Another question: Why does Sun not try to participate in ECMA? What are your real concerns with Open XML? Can you name them? Would you be willing to challenge them in the TC? Of course I do understand your reasons: You cannot oppose MS Open XML 100% and take part in ECMA at the same time. You’d even risk failure with ODF while still not gaining enough influence with Open XML. But that’s a tactical decision. Unless you try, it’s hard to blame MS for the outcome.

  11. StefanW says:


    1) I don’t know the answer. I only said that the answers you asked for were there, but of course you don’t have to believe them. Also, as Simon pointed out, I might have underestimated the effort that went into ODF for MS Office compatibility. But I also heard Jason talk about the hundreds of details they had to consider when creating WordML (was that you on Channel 9, Jason?). I can really imagine the pain of doing this work embedded in a committee.

    But yes, the success of OpenOffice and ODF might have changed MS’s strategy. It was important that they tried to establish an alternative standard. But now the pressure is there, and I hope that ODF lives long enough to prevent MS from reducing its openness. I really do.

    2) I could argue that IBM and Sun open up their products as soon that they realize that they cannot win against MS using the same proprietary approach and have to look for new business models. But I do realize that a company’s willingness to support open standards is crucial, even if they just do it for their own sake. The difference between our perceptions is that I think MS is moving rapidly towards standard compliance, especially compared to their past behaviour. The whole XML and Web Services field, LDAP and Kerberos, finally they are even trying to support HTML and CSS correctly in IE7. They need strong competition to stay on track, but which company does not start to behave badly in a monopolist position? Granted, there is some rearguard action with SMB and other technologies. That’s where they should be attacked. The Open XML thing is not one of them, as far as I can see.

    3) I still don’t know what’s wrong with Open XML and GPL. There were some issues in the first proposed license. I think it was about some crediting requirements that were not allowed by GPL, but I’m not sure. They proposed a new, better license, and I have not heard any serious claims that this license is not compatible with the GPL. IANAL, but I don’t see how a covenant not to sue without any additional requirements would leave room for either MS or the copyright holder of a piece of GPL’d software to sue somebody for implementing Open XML in that piece of software.

    Again, MS is not in charge of the GPL. It’s their job to specify under which conditions the Open XML format can be used. It’s up to the FSF to interpret the GPL and point out any problems a GPL licensee would have by building Open XML support into a derived work. BTW, interpretation of the GPL is not always unambigous, and giving legal advice could bring legal problems for MS too. (How about liability?) Take into account that GPL software is not only published in the U.S., so other legal systems may apply to it too.

    To sum it up: I disagree with how "open" MS should be about the GPL, because they cannot provide binding interpretations of the GPL.

    And I also disagree about how open MS should be about the motivation of past decisions, because if you ask any company such questions, you won’t be expecting the truth anyway. (For reference, go back to the original discussion about the decision of Sun and IBM to be open – was it for society’s sake, for the broader economic perspective? Or was it because that’s the only way they could find to beat MS? Who knows? Who cares?)

    You can ask for answers, but you’ll always get rationalizations. I think we’ve taken this discussion far enough to make clear that there can be valid alternative viewpoints. The only thing that can be said for sure is this: Had MS chosen the OASIS route, there would be no discussion about their willingness to be open. It’s much harder to prove the opposite, though.

  12. Wesley Parish says:

    Some interesting points.  I don’t doubt that economic interests are driving this whole shebang, at least as far as the companies are concerned.  I don’t doubt in the slightest that IBM has not the cleanest of hands with respect to prior vendor lock-in – their greatest successes (S360/370 and the PC) were simultaneously their greatest failures, as far as vendor lock-in.  Which as far as I am concerned, is a bit of smoke-and-mirrors.

    I’m going to concentrate on writing the most compact ODF I can, for my project – that’s because my target market is not only the embedded platform, it’s also the low-spec PC.

    As far as "ink annotations" in ODF, I haven’t seen anything yet – but then, ODF successfully incorporates SVG, and I don’t see why not.  OCR should be a breeze, since the underpinnings are already there; you treat anything vaguely circular/elliptical as an ellipse, and anything that can be read as an alphabetical letter – remember ODF incorporates bidi from the ground up – as an alphabetical letter.

    Oooops, I’ve given away the top-secret IPR ingredients!  Woe is me!  ;^)

    And of course, the nice thing about getting a GPLed project underway is that if I like, I can also write ODF for SIAG (for example) without worrying about NDAs, or any such petty bureaucratic nonsense designed to stop people doing things.

  13. StefanW says:

    Nektar – I think I was wrong. It was probably not Jason who has tried to explain the decision of MS not to join ODF, but Brian Jones.

  14. StefanW says:


    i have not read the licenses, but I’ve heard that MS’s licenses are not so much different from Sun’s licenses on ODF (or rather, the patents and other IP that Sun holds which could prevent third parties from impelementing ODF). Never heard about NDAs in this context.

  15. David Lane says:

    Given Microsoft’s superior "Open" XML format, why don’t you throw the world a bone and also implement ODF, on the same basis that you implemented RTF compatibility – although it wasn’t a patch on the .DOC format in terms of flexibility and feature sets, it’s is (was?) a useful format for many people and, at the time it was implemented, helped you compete back when competition was an issue.  I think all the arguments about ODF being too limited to cater for all of MS Office 2007’s innovative features are a total whitewash and are irrelevant.  Support *both* and let the market decide, eh.  I think Microsoft is just in denial that it’s empire is crumbling and its emperor is flappin’ in the breeze.  

    Kind regards,

    Dave Lane