Formats, Formats, and more Formats….some say there should be only…except the other one…and that one…and the new one…and…<sigh>

In the comments section of my last post, there is an absolutely solid discussion going on the idea of a single standard. I have promised to put some more sophisticated thoughts into a top-level blog, and I will. But, first I simply have to comment on some of the stuff I'm reading out there.

I encourage you to check out this blog posting from Sam Hiser. Sam is not a fan of my points of view. So much so that I keep him linked in my critics corner because I strongly believe in keeping an open mind about issues before coming to a conclusion.

Sam, and the organization he is closely associated with - The OpenDocument Foundation - have come to the conclusion that ODF is not heading in a direction they support, and are now advocating that attention be moved to the Compound Document Format (CDF). The CDF work is currently under way over at the W3C if you are curious about it.

In Sam's words...

It may be news to some -- not to the ODF Community, certainly -- that we at the OpenDocument Foundation have been displeased with the direction of ODF development this year. We find that ODF is not the open format with the open process we thought it was or originally intended it to be.

This leads then to...

Among ODF's weaknesses is its provenance from a specific application and the unwillingness of its originators to release it into the Bazaar. Merchants of irony will note this is the identical problem that paralyzes the incumbent gorilla's format.

Before I comment on this (and I'll note here that I may have made points similar to the comments about commercial intent behind ODF in the past), let me finish by noting that Sam supports CDF because it meets the following criteria (again, from his blog)...

  • openness & objective oversight
  • full compatibility with legacy MS formats
  • convergence of desktops, servers & devices
  • cross-platform portability
  • vendor independence
  • an explicit interoperability framework
  • freedom from patent & other encumbrances

OK, now for my comments. I will keep them somewhat brief.

  1. There are many document formats out there. Innovation will continue to push technology forward (especially in the applications) and thus the need for evolution and flexibility with document formats will continue to move forward at a rapid pace as well. Now, with the push towards standardization of these formats the argument is one of consolidation. Yet this does not jive with the actual situation in the marketplace. The OpenDocument Foundation could not be making this point any clearer for me. They had hoped one technology was going to get to a certain result, but that result did not materialize. So, they are now hoping the next one will do it for them. Let's say CDF was actually adopted in mainstream apps (ODF is certainly ahead of CDF on this front) and the format moves out of the theoretical specification phase and into broad commercial implementations phase (meaning that things like app compat, competitive differentiation even among "friends," and a range of other potential points start coming into play)...they may find that there is a need to move to a new format for the cause once again. Oops, more formats.

  2. I will point out again (and I'm sure I will do so more than this time), Adobe is not only standardizing (through Fast Track none-the-less) their doc format, their lead engineer eagerly pointed out the fact that Adobe also has their own "XML-friendly" format in the "Mars Project" aside from PDF. Anyone else interested in pursing their own strategy with this stuff? Oh yes...that would the Chinese Government with UOF. Ok - so other than those efforts...there should be just one.

  3. It would seem that some of the rhetoric around ODF might also need some re-vamping considering the points being made by their own community. I'll leave it for that community to discuss their own differences for now.

  4. All of this seems to make the point stronger than ever that when you are speaking about document formats, you are really speaking about an adjunct technology to the applications which are the real "solutions" in this discussion. If consumers want those apps to be pushing the innovation envelope year after year (I'll point to the MS Office team adding OneNote to the mix over the past few years and how different that experience is for doing the same thing as Word on the surface...typing text...but yet oh so much better for other uses.), then the formats will be a representation of those features. This seems much better for consumers than constraining the apps to the limitations of the format in the name of consolidation.

  5. Again, just to put the point in there - translation and metadata will figure in very large to the long-term consumer concerns of data control.

Ok, so my comments were a bit longer than I thought they would be. I'm still going to ruminate more on the doc format vs. network protocol discussion as it applies directly to this same discussion.

Comments (11)

  1. Sam Hiser says:


    Yeah. We believe it is worth trying to establish a single format around which all applications can function with equal status.

    Our position is controversial among makers and sellers of software because it tends to de-emphasize the importance of applications.

    We also disagree with our respected friends & colleagues over at ODF about the strategic importance of working with your / Microsoft’s installed base of applications | documents | formats. We feel that if one ignores the 1/2-odd billion desktops out there, then one is not solving anyone’s particular pain-points. We kind of like your company’s old Embrace & Extend concept.

    Thanks for the thoughtful comments. Multiple formats is a reality; we’ll never get down to one, but a universal open standard format is worth a try. The main reason in my mind is to neutralize the forced upgrade path to Microsoft’s next set of products. Don’t you imagine people would like a <em>choice</em>? At least a choice.

  2. Ian Easson says:

    What is extremely ironic about Hiser’s list of criteria for an ideal format is that they are all true of OOXML.  ’nuff said.

  3. Len Bullard says:

    Big sigh…

    We’re having the same discussions in the virtual worlds market.

    1.  A standard exists, is open, is supported, works, has ancillary support, is actually older.

    2.  IBM steps up to the plate to ‘lead a virtual worlds interoperability forum’ yet has no products, no credentials and no experience.  They brought a checkbook.

    3.  Because of the checkbook and the recruits, the media jumps in and says this The Future Here Now.  Most of those articles are also written by people who have built no worls, have no credentials and no experience.  The brought a blog.

    4.  Some companies show up with solid proposals and running code.   These are the companies to watch but also to buy from.  Ultimately, the vote of confidence is the sales invoice.

    I don’t need to go on.  The pattern of waiting for solid emergence then jumping into a market with branding efforts to sell iron has been repeated many times now.  Most veterans get it.  If you are a startup with real IP prospects, keep your mouths tightly shut until the patent process is complete.  Then take that valuable property and go work with the consortium that works with your business model.  Sometimes you really do want to trade the IP into an open IP consortium where you get the trading benefits, and other times, you really want to hang on to it until you are established.  IP rots so invest wisely.   If you have old IP and the market for your products is aging, open the source as much as you can afford to and push the IP toward a consortium with participation agreements that ensure you will get fresh IP back.

    Here’s my advice to Microsoft:  with the exceptions of mandates from customers such as the Massachusetts incidents, you have to stick to the old rubric "running code and rough consensus".  There will always be Spy vs Spy battles heavily financed with layers of indirection.   These are 1980s tactics and they still work to delay market coalescence, but ultimately, sales volume prevails.    This doesn’t mean the best technologies won, of course, just that the customer muddle through somehow.

  4. Sam Hiser says:


    True if your world is all-Microsoft. Outside the MS catalog, OOXML is an orphan.

    An organization working in Zimbra (rather than Exchange | Outlook | Sharepoint) or Enterprise Gmail or most Office 2 apps looks at OOXML and their head turns sideways with their ears cocked, like my dog.

  5. jasonmatusow says:

    Len –

    I have been watching the virtual worlds discussion with interest over the years. In fact, one of my favorite examples to give when thinking about the role of interop on a given market segment is "How many dwarves want to do a land speculation deal on an asteroid?" Everquest gold can’t be spent in any other massive multiplayer game…I can’t take my e-bay reputation points and use them in World of Warcraft (to my knowledge). Users have not been asking for this yet, the technologies are relatively immature, etc. etc. But, when that needs to happen it won’t just be a standard that solves it. Product dev cycles need to be put towards implementations of protocols, formats, APIs, etc.. Business partnerships, collaborations, joint dev projects will have to be spun up. IP will have to be licensed for collaboration…and standards will need to be agreed upon.  I think the whole virtual world discussion will be significant for the next decade of computing.

    Sam –

    I still think the point of the opening of the formats is to see them used broadly, and for more than current implementations in office automation suites. We are already seeing this happen with both Open XML and ODF – and I think that is a good thing. I disagree with your comment that outside of Office, Open XML is an orphan. Impirical evidence of independent implementations (in fact, comments in this blog) prove that to be true.

    Thanks guys.


  6. Len Bullard says:

    I agree.  OTOH, there are good working examples of how to do this, settle the IP issues, and continue an open collaboration.  A-priori participation agreements transparent to the public are necessary from the keiretsu bodies so that the work with organizations such as ISO is "clean as a hound’s tooth" to quote a former President facing the possible mistakes of a soon to be vice president.

    The roadmaps here are pretty well made and easy to follow.   The problem will be very large dysfunctional companies with one division not being aware of another’s strategies and the easily misdirected ambitions of startups and their VC financiers.   Otherwise, the technical issues aren’t so hard as to involve lots of research and innovation worthy of acquiring IP.  My sense is the expense is the hardware/infrastructure, not the software or licensing.   Regardless of how far the existing systems have come, up ending them with open source open collaboration will be easy to do, so investors beware.

    Also note well:  the European companies are well ahead of the US in their rendering technology.  They aren’t going to be cowed by attempts at faux standards.  They implemented the ISO standards, have excellent product and are not so dumb as to overlook their current advantages in these markets particularly as those standards already have US DoD buy-in.

    So it won’t be *without* the standards either.    This will be the case that makes or breaks the public ethics of the big companies with regards to fair competition.

  7. Dave S. says:


    You are probably right – the joy of standards is there are so many to choose from.

    However, document problems seem to mostly emanate in the marketplace from the inability of Microsoft to get its own act together.

    Word has crippled table functionality, and, as far as I know, a faux database concept for mailing lists and the like. Excel has its own database look-alike. Isn’t  Access the database portion of the Office Suite? You shouldn’t be disparaging  of anyone else’s community when Microsoft has remained so dysfunctional.

    Now, with the XML concept put into play, and if the MS advertising is to be believed, customers can have custom XML embedded in documents. But those documents are carved into pieces and joined through the magic of the zip format. This could have been done already by anyone – zip the custom XML with the old formats and you get pretty much what you have today.

    An old adage for software is – show me the application and I might guess the formats, show me the formats and I can create an application.

    Innovation in applications? In the software industry? It’s like innovation in Detroit used to be – this year more chrome, next year the subdued look. While the hardware has gotten faster and storage more extensive, I can’t recall any innovation as useful as the work of Ivan Sutherland and Dan Bricklin.

    Finally – would the MS reps get their act together on what constitutes implementation? There are -no- confirmed implementations applications compliant with MSO-XML. Zip. Nada. Nein.

    This is true for two reasons. First, most applications obviously deal with a very limited feature set. Second, there is no public cross-reference between the source code of the applications and the requirements as laid out in the MSO-XML standard showing that for every requirement there is a corresponding, public-source-code, documented function and that there are no functions for creating files that create non-compliant entries.  

    In set terms – no demonstrated one-to-one correspondence between those elements of the MSO-XML set and the elements of the applicaiton source-code. No more, no less. Otherwise it’s either a partial implementation, or the application includes undefined elements and is a non-conforming generator.

    Wanna do a favor? Fast track the Office Suite Source code through ECMA as a reference implementation under Creative Commons or similar license. Then the users will know for sure how the data and the application work together to have "applications which are the real "solutions", " but won’t be held hostage if MS decides to mess with the format.

  8. It is worth recapping some interesting conversations related to a blog post I made last week about the

  9. Open XML says:

    Deux membres du comité technique OpenDocument à l’OASIS, Gary Edwards (fondateur et président de la fondation

  10. I have repeatedly made the argument that it is bad logic that leads you to the conclusion that there

Skip to main content