Snarky Comments

My last blog posting was in a tone that was a bit too snarky for my tastes. I was writing late, and tired – and that is never a good combination. Readers of the blog have, in the past, asked me to maintain a more level tone given the issues I’m addressing. So, my apologies for that.

There are some very serious issues on the table for industry right now with the document format discussion as a whole. The role that standardization will play for technologies is clearly a huge issue. There are many who are saying that standardization is THE answer to interoperability, and that “open” standards are THE answer to standards.

I have blogged in the past on the idea that interop is acheived many ways and that all companies use some combination of four ideas (products, community, access to technology, and standards) to bring about interop. Also, there is a natural, and healthy, tension that exists between unique value of solutions and interoperability. Perfect interop may make for a crummy commercial opportunity for a product, whereas perfect uniqueness makes for crummy interoperability. 

In the Open XML discussions, those that oppose the Ecma format are using some very heavy-handed tactics and rhetoric to advance their position. The work done by individuals on behalf of ODF – I’m completely fine with their approach. IBM, I don’t have as much sympathy for. Their own business model for many of their current software products contradict the positions they have taken in the document format discussions. Sun has been more consistent in terms of their actions on this front – but they too are motivated by commercial gain. And, so as not to put myself in an awkward position – Microsoft’s position on Open XML has been one of commercial interest as well. The difference there has been that our motivations have always been rather transparent, we’d like to see broad adoption of Office.

The creation of standards is not an pretty process, and it is important to remember that no one participates in a standard-setting activity by accident. Standards are not created by accident – there has yet to be a situation where someone stumbled over an obstacle while carrying a stack of technical specifications and serendipitously ended up with a standard. Standards participation is generally focused on advocacy of a specific technology (usually in order to create a market opportunity) or on unseating an encumbant technology. These may seem to some as unsavory goals, but they are usually the drivers for companies to participate. Governments have a different set of goals in standards, for example health and saftey or the fostering of market competition. Yet governments too have some corners of their standards closets that they don’t flashlights peering into too deeply. There are constant claims and counter-claims of block voting by some countries, or trade barriers being erected through standards, or protectionism of a specific industry from disruption by new competitors. 

The next few days are going to be very interesting. I’ll do better at keeping the snarky comments to a minimum.


Comments (12)

  1. orcmid says:

    Good, because I recommend you as an exemplar of level-headedness to my friends who want to know what the fuss is all about.

    Of course, I don’t have to be quite so polite:

  2. Stephane Rodriguez says:

    Jason said "The Open XML Translator is now in production, and delivers interoperability."

    What do you mean to imply here? Do you mean that because this component is running on say ONE physical customer site, it is ready for mass consumption?

    That is a gross mischaracterization of not just that case, but of software deployment principles in general.

    It is quite possible that the component meets the needs of some customer. After all, it does not process air, right? But you cannot extrapolate to all customer needs out there, especially when on the one hand Microsoft themselves say that customers out there use 10% of MS Office products, all different 10%, on the other hand the list of supported features by this component is arbitrarily limited.

    Who made that list?

  3. Stephane Rodriguez says:

    Jason said "The difference there has been that our motivations have always been rather transparent, we’d like to see broad adoption of Office."


    Back in 2000, in the WYSIWYG HTML era, Office 2000 shipped along with proprietary IE extensions (markup language, and a proprietary rendering run-time) known as V.M.L.

    (that’s perhaps the only time the Office product group worked with the IE product group in the whole history of Office.)

    There is a BillG exhibit in the IOWA consumer case that mentions that MS Office "must be the only platform able to render Office documents, on top of Windows".

    Hence VML. When creating HTML documents from Office 2000, the MSO semantics (proprietary) are translated into VML markup that only IE can render.

    What you did, in the name of HTML standards, was probably because you cared about interoperability, right?

    No wonder Office 2007 documents contain not just those VML parts (instead of DrawingML parts), but there are new VML parts introduced. Meaning that Microsoft does not intend to kill their old strategy by a long shot.

  4. Hey, at least you’re in the game and contributing. Beyond control, it seems evident that many governments see ODF as an opportunity to save money. When you look at their terribly slow upgrade cycles (how many, e.g., are still stuck on Office 97?), you soon see how daunting the cost of converting their systems and data are. Add in a new file format, and they feel stunned in the face of sticker shock at the task.

  5. JasonG says:

    Hey, maybe the ODF supporters will stop posting snarky comments as well?

  6. I guess my one objection to the whole process by which the Office Open XML standard has been developed is that there was such resistance to suggestions and over-reliance on what seemed easiest for the current code to work.  Even in places where it seems almost certain to come back to haunt Microsoft, the easy way was chosen.

    As a small software developer, I don’t "favor" ODF and "not favor" Office Open XML as much as I favor the approach taken in developing and modifying ODF.  I wish that Microsoft could get together with OASIS and develop a comprehensive standard that would allow true interoperability and support for legacy documents.  It would certainly be a lot easier than the current chore of developing for both ODF and OOXML, without a lot of rational explanation about why they shouldn’t be made one.

    So, aside from IBM’s stake in the game, we independent software developers would be better off if you and IBM were to get off your high horses and work together.  Then we could all serve our mutual customers better.  At the moment, the ODF standard is a heck of a lot easier to support, so it will tend to "win" resources, but the Microsoft Office dominance means it forces us to develop for that too.  It is just a pain in the neck, and doesn’t lead me to feel all too happy with either side.  

    I really don’t understand why Microsoft couldn’t have just supported and extended ODF when it had a chance, but I guess that horse is out of the stable now.

  7. One quick suggestion – if you can, at least edit the previous entry to put a link to this post at the top of the page. Otherwise, the two entries stand pretty much alone to the casual passerby, and may not give the impression you wanted.

  8. jasonmatusow says:

    Got it Philip – I will make the change.

    Ben, thank you for your comment. I can imagine that the clash of the titans, while good fodder for the media and the blogosphere, is painful for that smaller ISVs.  I wish I could see a silver bullet in this, but I don’t. Commercial interests in this one are very strong with literally billions of dollars in the mix. IBM does not want to go back and recode all the work they are doing on Notes to support OpenXML. Microsoft had already been going down the Open XML path (back in the pre-Office 2003 days). Also, the architectural decisions behind the two formats are very different (verbose vs. abbreviated XML etc.).

    There is never an easy answer – and clearly self-interest by many parties is going to continue to push hard on the whole discussion. As Stephanie points out in her comment, factors of marketplace competition drive all of the participants.

    The good news is that the underlying move to XML by most players is very strong. Interoperability is more of a design goal for everyone now than it has ever been. The trending is positive.


  9. heavy-handed tactics and rhetoric? i think its called corporate strategy isn’t it? I am pretty sure I have seen some rhetoric from Microsoft in my time.

    fwiw Jason I think you would gain even more credibility by linking to some non-Microsoft bloggers.  Your blogroll is very whitebread. perhaps a link to Rick Jelliffe? 😉

    Just to be self-serving for a moment I do tend to think you could call out Stephen’s blog. I know you read it and take it seriously.  

  10. jasonmatusow says:

    James – thanks. Yes, you are right about the fact that I should be linked to many voices. I am bad about putting time into my blog and this is one of the things that is important for me to update. I take many people’s blogs seriously – those from Redmonk more than most.

    Yes – I agree completely about the rhetoric being about corporate strategy. That has been my point all along.


  11. jasonmatusow says:

    James – I added some folks that I like to read to my site. And yes, my site is very whitebread. <sigh>


  12. Jason – I would agree completely that the general trend towards XML formats is strongly positive, no matter how this particular skirmish turns out. – Ben