My testimony to the Texas House and Senate regarding the open document format legislation

Print Friendly

This is the text for the testimony I delivered to both the Texas House and Senate this last Monday, March 24. The words I said varied from this because of time constraints and also some additional comments supporting or questioning previous testimony.

Good afternoon/evening, Mr. Chairman and members of the Committee. IBM supports this bill. This bill is about the future, increased competition and innovation, and about more choice for Texas. It is completely consistent with the technological and intellectual property directions of the software industry.

The current file formats for how you save office document used by most of you and your citizens are based on technology and practices from the 70s, 80s, and 90s when some companies locked customers into their products and upgrades. This is not acceptable today.

When you and your citizens are effectively restricted to a single software supplier to access government information, you and they pay what I would consider taxes. Open standards avoid this.

The first tax is the difference between what you must pay to that supplier vs. the lower cost if multiple suppliers existed and prices had to be competitive. You would also pay an innovation tax. The sole vendor has limited reasons to improve the product. Fresh ideas from new players such as Texan entrepreneurs are kept out of the product category. This is bad.

Those who are against this bill are, in essence:

  • comfortable with having a single supplier,
  • satisfied with procurement policies that allow you to only have a single supplier for government document software, and
  • just fine with paying the financial and innovation taxes above.

IBM is not fine with the status quo. Neither were the drafters of this bill. Nor are most industries; those in life sciences, education, healthcare, and so on, that are trending toward “openness.” With the creation of the Internet and the Web, based on open standards such as HTML, the value of real open standards has been seen.

Think how much easier, more affordable, more transparent it is for you to collaborate within government and connect with your citizens because of the Internet and email, blogging, and all that has come from open standards. Now it is the time to take this collaborative power to documents and open them up giving control to governments and choice to citizens.

IBM joined our industry colleagues to work on an open standard for file formats, namely, the OASIS/ISO OpenDocument Format (ODF). File formats are merely blueprints for how a document is structured ” headers, footer, paragraphs” and how it should be saved and exchanged. OpenDocument Format is being openly and actively developed by a community of global experts from many organizations and is seeing broad implementation in independent ways from both open and proprietary sources.

Its adoption rate is growing. Teenagers are using it. Politicians are using it. Some CIOs in organizations that officially use only proprietary formats are using ODF at home when it comes time to spend their own money and technical expertise to pick products for their personal use. The huge and growing base of Open Office users are saving and distributing files in ODF format. The next generation of IBM’s Lotus Notes will support it later this year.

So why do you need legislation on this?

First, Texas as a sovereign state and a major force in the IT market and must, in my opinion, be able to do anything whatsoever with the office documents you create. This means today, but it also means the documents you create tomorrow and will be the historical records fifty years from now. You have the opportunity to clearly make a statement that Texas will not be beholden to any vendor for access to your state’s information.

Second, change is happening now and users will, over time, get new applications that use new document based file formats. I have never met a CIO or financial person who has told me that they will never get new software. So, there’s a fork in the road approaching rapidly and you must choose: go with a single supplier and pay those taxes I mentioned, or go with truly open document formats that are not dictated by a single vendor and get increased competition, innovation from many parties, and real choice of software and its providers.

I can assure you that the software we have in fifty years will work in radically different ways and will be supplied by completely different providers than we know today. We must leave our options open and, luckily, with ODF, we have an excellent choice compared with any alternative. Further, personally, I would rather bet on the “intelligence of the crowd,” the collective smarts of the IT industry who truly manage open standards to set us up for success in the next few years. Relying on one vendor to optimize things for his or her success is yesterday’s solution.

Third, to be clear, EVERYONE can implement a true open standard. This bill is about choice. ODF and open standards for file formats will drive choice of applications, innovative use of information, increased competition, and lower prices. Personally, I think these are good things.

In closing, the world is shifting to non-proprietary open standards based on the amazing success of the World Wide Web, a success that was far more important than any single vendor’s market position or ideas for what was right for the world.

We can do this again but we need to do it with care. Texas is in a position to demonstrate to its citizens and the world that this success is repeatable and that it intends to be a leader. Texas can show that the phrase “open standard” means more than what a corporate marketing department says it is.

Mr. Chairman and members of the committee, open document standards are insurance policies for your documents, versus a history-losing accident waiting to happen. I thank you for the opportunity to testify in support of the bill, and would be pleased to answer any questions you may have.

I have a longer discussion of the tax points in the original blog entry “Avoid the single supplier document taxes.”

Addendum: Avi Alkalay of IBM Brazil has posted a translation of this into Portuguese.


  1. This is a wonderful letter.

    I would like to translate it to portuguese and publish it in my blog, with your permission.

    We in Brazil are living a great moment for ODF. Yesterday was just announced the ODF Alliance, chapter Brazil (more on the website). The initial components of this independent organization are IBM, Sun and Red Hat.

  2. Avi, yes regarding the translation and congratulations on the Brazilian chapter of the ODF Alliance!

  3. Bob, I’m so very curious to learn whether the committe asked you any questions and, if so, what they were and how you answered.

  4. Nathan, it’s hard to remember, but I did answer some questions about the positions of some of those opposed to the bill, which I refuted. I mainly recall my very last statement at the end of the House hearing where I addressed what was happening with ODF and Accessibility. The basic response there was that what was a liability in 2005 is turning into a very successful global community effort to provide state of the art accessibility support in a true open document format, namely ODF.

  5. Bob, Thanks for your reply and the experience and agile intellect you bring to bear in advocating open standards like ODF. I hope the parliament (I’m not sure of the correct term in Texas) makes a wise decision for the benefit of the people of that state rather than one for the benefit of a few.

  6. I second Nathan’s words. When the time comes, you will have to come to California and help us out as well.

  7. Bob –
    Great response. One item you mentioned but did not highlight is the long-term access. I know HIPPA (the Health Insurance Portability and Privacy Act) requires storage of records / documentation for at least 30 years.
    With a proprietary format, if the vendor changes the format, you have to change also. Now you run the risk of not having software to access your older records, keeping copies of the old software around (which may involve licensing issues, as well as possibly require keeping old hardware around), or having to pay the vendor to create translators – which they may not do. Even worse, the vendor may go out of business and you could lose access to the data you’re required to be able to access – a catch-22. You are at the mercy of the vendor.
    With open formats such as ODF, the open data format ensures you can always create a product that accesses your data – even years old. Of course, this only works if the data format is completely specified (no “do it like xxx product” fields). ODF meets this criteria.

  8. Bob, this page (and probably all others) looks bad on a wide screen with IE7.

    The right sidebar its not on the right. It starts only after the comment form, aligned on the right.

  9. Is that any better? There was a URL in a comment that was pushing out the right column.

    I’ll save the conspiracy theory about this particular entry not looking right under IE7 (grin).

  10. Bob, have a look at this one:,id,102,site_layout,sdaindia,news,16489,p,0.html

    “In response to an attendee’s question on this issue, Hilf clarified that the file format was a part of the software, which is why they could not adopt ODF as the file format for subsequent versions of MS Office.”

    And no, although it’s April 1st, this is serious…

    “part of the software…”

    Give me a break… :-)

  11. Well, even if OOXML is not “part of the software” as such, it does read like it might well be closely based on it; how else would you get such a variety of element name styles, for instance? If the element names are copied from the corresponding variable names, then it makes sense. (And testifies to the poor quality of the code itself, but that’s a different story.)

    Of course, if it were true that OOXML is “part of the software”, it would amount to the harshest criticism of OOXML I’ve ever read.

  12. Excellent!

    You made a very good case for ODF. As a native Texan and Free Software advocate, I very much appreciate what you are doing.

    I realize you have to be brief, keep it simple, and communicate the essence of the subject in terms that can be understood by the senators and congressmen in Texas.

    There are so many good points to be made on the subject of open document standards that I’m sure it is difficult to pick out the most important ones.

    Cross-platform interoperability would have been a point I would have emphasized more, but I realize these congress-critters might not have as much reality on that as on taxes. I also would have emphasized that all of the taxpayers in Texas have a right to be able to access government documents, not just the taxpayers who can afford to pay the monopoly corporation’s tax.

    It might also have helped to mention that plug-ins are already available where documents can be read and saved in ODF from within Word.

    Anyway, great job and keep up the great work!

  13. My question is this: How much modification of old documents is being done when a researcher or database searcher accesses old files. The only one I can think of is the recent troubles of a document being released with parts being blacked out for security reasons (wrongly I might add because an astute user of word could still get the blacked out portions by using word’s history function). I can see being able to use cut and paste on the document as a real boon to usefulness. Last years document where one just changes the year from 2006 to 2007 and your new document is done is a very useful function but really old documents almost need to be somehow certified as being as they were in 1948 or whatever because someone is appealing for the information in the document to be the truth because witness is long gone.
    That said, Is ODF lacking if very old wordferfect or word document doesn’t format exactly as it would in the appropriate time period application. Although from what I know the appropriate time period application won’t run on a current computer anyway. Microsooft’s Open XML is not going to be any better in that way either though if one uses that.

  14. This may be a little off topic but I wonder why the United States is taking such a pro Microsoft approach to the approval of OOXML. I suspect I know the answer but I am interested in your thoughts.

    The Contradictory Nature of OOXML (Part II) – 19 Nations [make that 20] Respond

  15. Well, some US corporations (notably Microsoft and anyone in their ‘feed chain’ of distribution partners who get money when someone buys a copy of Office 2007) are obviously well-funded and for OOXML.

    Other US corporations (notably RedHat and Novell, who get money when someone wants to buy services related to open-source software) are obviously against it; the view is that there is no point in ISO standardising OOXML since it does not enable any additional interoperability; ISO ODF XML has been designed by many from universities and corporations (i.e. not-for-profit experts and for-profit experts) and should be allowed to do its job.

    I’m slightly biassed here, becuase of who pays my salary, so I’ll confine my remarks to the obvious, that OS/2 and Lotus SmartSuite have sold practically all that they are going to sell, and belong in the museum.

  16. Chris, I don’t understand your comment about OOXML not enabling any additional interoperability beyond what ODF provides. Based on everything I’ve seen, ODF looks like a great way to enable interoperability in the desktop-software business, but most people I know aren’t in that business. What about people in other types of businesses — how does ODF enable interoperability for them?

    For example, suppose there’s a schema that I use in my business every day, and I’d like to create documents that use that schema to transfer data between diverse systems that I use to run my business. Where can I learn how ODF supports that scenario?

  17. Oh, I meant that OOXML only enables interoperability between one Microsoft-provided installation and another Microsoft-provided installation. You write a document with Microsoft Word, send it to me in an email, the only thing I can do with it is buy a Lenovo-compatible Personal Computer and copy of Microsoft Windows ; then I either download the no-charge Microsoft viewer and look at it, or buy Microsoft Word and revise/collaborate.

    ISO giving it a rubber stamp doesn’t enable any other vendors to get in on the act; for example I won’t be able to get it to a Sun system running Solaris and IBM Websphere, and understand it in any useful form. Or, domestically, get a PC out of the recycling skip (saving on landfill), ask Mark Shuttleworth to send me a Ubuntu, and be able to do something with the document. Or get a Playstation 3 with Yellow Dog Linux, ane deal with the document there. Whether I can do anything with an Apple Mac is by grace and favour of whoever succeeds Bill Gates at the helm of Microsoft.

    It would be nice if Microsoft would sell a Word viewer for Linux, and/or a full version of Microsoft Office for Linux; but they are currently making a commercial decision not to do so. And no-one can force them to; it’s a commercial choice.

    ISO rubber-stamping ODF XML does enable interoperability. We have several vendors (and open-source distributors) who have implementations of it; the spec you can get at no charge from ISO (which isn’t the way ISO normally operate, normally they sell the specs, so someone must have been sugar-daddy and given them a bucket of money to make it a no-charge item); so you can implement machinery which understands and creates ODF XML documents, either buying a pre-existing one, or hiring someone who can do it, or doing it yourself.

    Within your business, what you do is clearly for you to decide. You look at the requirements, and you implement a solution which meets those requirements to the best of your ability. Or you buy a solution from a vendor who offers terms and conditions that you like.

    But an open standard is more important when it comes to communications between ‘governments’ and ‘citizens’; and where it comes to important records (such as archives, minutes of meetings, medical records, long-term contracts between corporations), where it’s wise to be in charge of your own destiny and avoid forcing whoever you want to communicate with into buying from a particular vendor.

    It’s worth a little more investment up-front to make sure that you succeed in communicating with your audience.

  18. Dick, ancient applications can still run on today’s computers, if an emulator exists for the old platform.

    Emulation used to be too costly to be viable to support older hardware, but these days, when you’re emulating an 8-bit or 16-bit processor, the emulator overhead isn’t enough to slow down the application to usable responsiveness, and today’s emulators actually include additional delays so that when you press a key in WP 5.1, your screen isn’t instantly filled with repeats of that letter.

    That being said, there are still support issues, and emulators aren’t perfect. I have heard that printing – a task fundamental to use of ancient documents – can be especially problematic.

    As a tech, I find the concept of certifying a document as being from a particular time to be… interesting. One could always disconnect their computer from the network, set the time to whenever, and edit the document, without disturbing that certification. A simple time correction, and reconnect back to the network, and nobody’s any the wiser. Depending on ones computer configuration, the disconnect from the network may not even be needed. Some operating systems have detection or even blocking of changing the clock back to that extent – however, all one needs to do is get the document on a system without those restrictions, and you’re free to do as you wish.

    I find the fact that there may be unmodified versions of the file in circulation somewhere that could be used to refute your tampering to be sufficient safeguard against this. However, if that isn’t good enough, one can digitally sign ones documents. It may not certify that the document was, in fact, last edited in 1972, for example, but it will help establish that the author was, in fact, the last person to edit it. (One can still break digital signatures, but it significantly raises the bar to do such a behind-the-scenes edit.)

    Doug, OOXML may not be able to be implemented by anyone by Microsoft. At the least, producing another implementation of it will require significant help from Microsoft, as there are details that just are not mentioned in the ‘specification’. If nobody but Microsoft can implement it, it provides no interoperability.

    Contrast this with ODF, where there are currently multiple applications which support it, and you can open your ODF files in any of them and have them work just fine.

  19. Dick, since every office application reformats documents for use with the selected printer, even if you had WP5.1 or such running on an emulator, it would have a different appearance depending on which printer you selected. If you are not intending to edit the document, it really needs to be converted to PDF for long-term storage. If you are going to edit it, then it needs to be in ODF (or plain ASCII, ANSI, or UTF-8 text), so that you will have continued access to it in the future (without regard to your chosen operating system or applications).

    So your old contracts, including your mortgage agreement, should be moved to PDF. Things that you need to modify from time to time need to be moved to ODF. Things that you do not need to have available could stay in “legacy” formats, at the risk of someday being unable to open them. I would avoid “open but proprietary” formats (such as ECMA 376) entirely, at least until we know how long the vendor will support them.

  20. So well-spoken, Bob, thanks for sharing, and thanks for all your hard work on behalf of open source, promoting open standards, and ODF. Ultimately,

    “…open document standards are insurance policies for your documents, versus a history-losing accident waiting to happen,”

    sums it up for me.

  21. Chris, Ed, thanks for your comments. It does sound like we’re looking at interoperability a bit differently.

    I’m thinking of interoperability between disparate systems (a back-end LOB system and a word processor, say), and to build those types of solutions you need to break down the divide between documents and data. That typically requires custom schema support, so that the documents can become containers/interfaces for pre-existing business data that happens to exist in XML instances that have nothing to do with document formats.

    Conversely, if you’re looking for ways to interoperate between different instances of the same application type (such as two word processors that happen to run on different platforms), then custom schema support is irrelevant and other things such as simplicity of the spec become a bigger factor.

    So since we’re looking for different types of interoperability, we’re looking at the comparison between ODF and Open XML. With the DIS ballot period having started yesterday, this distinction will probably be one of the central themes of the debate ahead.

  22. Here’s the deal, Doug. When it comes to documents and interoperability, we’re talking about a single concept. We’re not talking about a “custom schema” that encapsulates a particular vendor’s implementation and then a general one for everything else. We’re talking about everyone working on and moving to that one format, and that one is ODF.

  23. I think you missed my point, Bob. In the Open XML spec, “custom schemas” have nothing to do with document formats — the schemas in Open XML are the “reference schemas,” and the “custom schemas” are the non-document-format schemas that people use to implement interoperability with their own systems.

    For example, the schemas Mindjet uses to describe the details of their mindmaps, the schemas Dassault uses to describe the details of their 3D drawings, or the schemas the Florida House of Representatives uses to describe the details of proposed legislation. Those are the types of things people are doing with Open XML’s custom schema support, and the ability to embed arbitrary XML instances in a document, separate from the visually oriented information that embodies the look of the document, is what makes interoperability with other systems possible in all thoses cases.

    I won’t belabor the point here; I understand the responses I’ve received and appreciate the feedback. But it does seem clear that the type of interoperability ODF enables and the type of interoperability Open XML enables are two quite different things. And I just wanted to make clear that custom schemas have nothing to do with Microsoft’s implementation of Open XML — you’ll never find a custom schema in a document created purely by Office, which only writes the Open XML reference schemas.

  24. Doug, thanks for the clarification, and I do agree that you folks have a different idea about what interoperability means (grin).

  25. We have a little work to do with the ‘Institutions’.

    The UK Institute of Physics (a non-profit ‘trade association’ for physicists, analogous to the IEEE for engineers in the US) is generously giving material to an outreach program, in the hope that physicists from industry will venture into school classrooms and help with the lack of physics teachers. Many UK school physics classes get taught by biologists or non-scientists (and I think the situation is similar in the US).

    Some of the material is here
    all well and good, vendor-neutral, except that there is a Powerpoint presentation and a Word document. That rather forces me to buy a copy of Microsoft Windows … the ‘Guillaume Portes’ effect … if I want to do anything with the outreach. In turn, that puts the Institute of Physics into Microsoft’s commercial distribution channel; probably in this case as an unwitting volunteer.

    Now, it would be perfectly simple for the IOP to supply the material in ISO ODF XML format (as well as the Microsoft format) ; place a link to people could download the open-source document tools if they didn’t already have anything suitable; and the outreach could outreach through more of us. I can pitch Impress presentations, and access Writer documents, in more than just the Microsoft contexts.

    Loads of web sites already do this for ‘published’ documents; put them up in PDF format and give a link to if you want to pick up the Acrobat reader. Not open-source, but it is a no-charge item and it runs ‘everywhere’.

    And now, with the ISO standard and the collaboratively-developed open-source implementation, we can do this for the ‘reviseable’ documents too.

    So how do we get the message across without appearing to be ungrateful ?

Comments are closed