ODF Alliance: The widening gap between fact and fiction

I was pointed at a document created by the ODF alliance (with a creation date of Feb 6th) that discusses the recent dust-up on supported file types in Office 2003 Service Pack 3. While this was not one of our shining moments as a product, I was disappointed to see the amount of gross factual error in the report.

I'd like to spend some time illustrating how the report is wrong, and how the ODF Alliance ignored available data in preparing it. By doing so, of course, I hope to illustrate the level of credibility with which such claims should be regarded. All the documents / information I reference were available long before the publishing date of this material.

A Knowledge Base article discusses the actual changes made: http://support.microsoft.com/kb/938810

ODF Alliance Claim

The Truth

“When a user attempts to open one of these older files, they will receive the above dialog box and no alternative actions are given to help users to get access to their information in these “blocked” files.


“When pressed for answers regarding this change, Microsoft eventually admitted that their action was in response to concerns with their parsing of Office 2003 code that presented a risk, but only after they suggested the move was in response to security concerns with the files themselves. Microsoft continues, in our view, to erroneously maintain that files in these formats are creating a “security risk.”


David LeBlanc, a senior developer in the Microsoft TWC group, explains the Microsoft position on the issue, which is that the parsing code for the formats is the general problem that requires resolution. This is consistent with other activities taken by Microsoft on other document format related vulnerabilities: http://support.microsoft.com/kb/935865

While it might be a neat trick for the ODF Alliance to caveat inaccuracy with “in our view,” nobody at Microsoft is claiming that the formats are the problem. If such a claim is made, it is against the guidance of the product support documented here http://support.microsoft.com/kb/938810 and here http://blogs.msdn.com/david_leblanc/archive/2008/01/04/office-sp3-and-file-formats.aspx

For what it’s worth, nobody is discussing “Parsing of Office 2003 code”, what is being discussed is “document format parsing code in Office 2003.” It’s not clear if the author of the document was in a hurry when this was written, or if they just lack a basic understanding of the problem.


“Really, what is at risk is Microsoft's ability to sell more products, namely their new Office 2007 which will lock users into their new file format, Office Open XML (OOXML), which despite its name, is not open. What is at risk is Microsoft's own coding errors.”

The logic here is hard to follow: We have disabled legacy formats of Corel and Lotus. Because of this, you are unable to download the ODF Translator to Save as ODF, and you are unable to use the array of other formats that are supported (that include Open XML, HTML, TXT, RTF, CSV, and a lot of others.)? Really? Or are we adding a FREE update to a product people ALREADY have to give them a way OUT of the legacy formats?

Even better, by disabling file formats created by our legacy products the ODF Alliance accuses Microsoft of locking people into our new products by encouraging people to use file formats that are more open than the previous formats. That list of formats includes Open XML, ODF AND UOF. In fact, the reality is exactly the opposite of what the ODF Alliance suggests.


"Unless you need to work with these very old file types on a regular basis, it's probably not a good idea to keep these file types unblocked for long periods of time." The spokesperson, Microsoft Office product manager Reed Shaffner, fails to mention what should be done if one does need ongoing access to older documents.”

The “Save As” feature works well as a remedy here. After unblocking the file types, one can use the current file formats of the Microsoft Office products, or ODF/UOF if one has installed the translator (or TXT, HTML or a number of others). For the Lotus and Corel formats involved here, one could always use the Lotus and Corel applications that created them.

Again, it’s not clear how Lotus and Corel formats are a means for Microsoft to perpetuate the alleged “lock-in.” This would be perfectly congruent to a claim that says “OpenOffice locks you in because it has the ability to open and save Microsoft Office binary formats.” – the argument is equally absurd in both cases.


“Just like Office 2003's lack of older file format support problems, OOXML compatibility with proprietary components like Windows Meta File (wmf) and Vector Markup Language (vml) have been deprecated because of “security concerns” that might prevent ISO approval next month.”

This is also inaccurate. VML was deprecated for the reasons described here: http://blogs.msdn.com/brian_jones/archive/2007/12/21/500-national-body-comments-posting-today.aspx and have nothing to do with security. WMF was removed at the request of various national bodies who desired to have references to “Windows Metafiles” removed – to make the format more platform-neutral.

I’ve spoken on security as well, and I am waiting to see the ODF Alliance recommendation for the removal of binary content from subsequent versions of ODF: http://blogs.technet.com/gray_knowlton/archive/2008/01/18/noooxml-says-down-with-xml.aspx


“Exchange 2000 (in 2003) supported the WebDAV; today it does not. To use WebDAV now requires the additional purchase of Microsoft Office SharePoint Server.”


This claim is odd (and false), because Exchange is a mail server and SharePoint Server is about portal management. SharePoint is not a mail server, and Exchange is not portal management software. This is a nonsense argument.

The more important part of this discussion, however, is that Exchange 2007 does support WebDAV: https://blogs.technet.com/sayleong/archive/2007/03/25/features-de-emphasized-and-discontinued-in-exchange-server-2007.aspx, which gives WebDAV life in Exchange Server (including product support) until 2016, per the Microsoft support policy. Exchange has deprecated the functionality in favor of other technology, but this is different from “not supporting” Web DAV.

“Office 2003 supported Reference Schemas that Microsoft offered to make public to governments requesting openness in file formats. Today, Office 2007 has obsoleted these schemas. Further, it is unknown what file formats Office 2009 - a product Microsoft is currently developing - will support”


Wrong again. The XML Reference schemas of Office 2003 are supported in Office 2007. And, while the attempt at FUD relating to the future of Microsoft Office is interesting, let’s wait and see what Microsoft communication that the ODF Alliance will cite as the basis for the claim.

“Recent events with Office 2003, and the examples above, should act as a cautionary tale of proprietary product, vendor, and platform lock-in. Although individual users are likely to experience frustration with compatibility issues seen in Office 2003 and OOXML, businesses and government agencies face a much more serious set of problems due to the ever increasing demands for document retention. Imaginethe expense that a government agency using OOXML will incur for the initial conversion of documentsand subsequent conversions that would be required whenever Microsoft is inclined to change their"standard".”


From this I guess we should assume the ODF Alliance is saying two things:

ISO26300 will be the only version of ODF that ever exists, and the cost of migrating documents to open formats isn’t worth the effort.


We disagree on both fronts, as ODF already has three versions, PDF has many (1.0-1.7, plus PDF/X, PDF/A and others). The whole point of having open file formats is to encourage a migration to interoperable solutions. It’s not clear why the ODF Alliance argues against this.

“The OpenDocument Format (ODF), an ISO standard, employs this type of design. Additionally, as ODF is supported by several vendors, platforms, and implementations, no user of ODF documents is at the mercy of any particular vendor.”

Last I checked, this list: http://www.openxmlcommunity.org/applications.aspx is a lot longer than this list: http://opendocument.xml.org/products.


And as we just saw in the prior argument, I guess we should take this to say that ODF 1.0 should be used, and ODF 1.1 and 1.2 should never be considered, given the (claimed) difficulty of migrating to an evolving standard.



At what point is the author held accountable for this level of factual error?

Comments (10)

  1. Anonymous says:

    Binary documentation and translator project. On Friday Brian Jones covered the availability of the Office

  2. GrayKnowlton says:

    Well, we might be interpreting that differently, but by "parsing of code" I think of compiling a program in preparation for execution. the object of the parsing in question is the document, hence the clarification. if the comment were ‘parsing of macro code within the document’ or ‘parsing of morse code within the document’ i’d agree, but we’re talking about how apps read document formats.

    as for identificaiton, we see this popular trend in the blogs of brian jones and bob sutor, and in some cases for good reason. you do bring reasonable questions / comments to my blog (which are appreciated), but identity brings accountability, and my name / picture are present here, so it’s only fair.

  3. GrayKnowlton says:

    Hi Dave,

    I’m not sure I understand the relevance of your comment, but now that you brougt it up, I’ll point out that yes, "OpenDocument" originate within OASIS as "OpenOffice XML Format."

    If you could do me a favor and visit http://www.openxmlcommunity.org and find the "anti-ODF" content there, you’ll get your answer on the organization that is doing the smearing here.

    And for what it’s worth, my objection here is the level of factual error in the ODF Alliance content. This document is so far off the mark that (if they value their credibility) they should pull it down.

    I’d prefer "Dave S." that if you are going to continue to comment here, that you please identify yourself.

  4. Anonymous says:

    Binary documentation and translator project. On Friday Brian Jones covered the availability of the Office

  5. Anonymous says:

    Patrick Durusau has picked up on a thread that I regard as somewhat important (and underplayed) in the

  6. GrayKnowlton says:

    Hi Anonymous,

    I am hoping that somewhere someone will have a site dedicated to the good things about ODF, rather than using comparisons to self-validate.

    In the future please identify yourself.

    Hi Jimmy,

    I’m not disagreeing with you here, but I am saying that all opinions are considered when we discuss Open XML; we’ve done quite a bit to accommodate a lot of input from a lot of people. But when it’s this bad, and we’re just flailing around without the facts to support the position, sometimes it’s worthwhile to set people straight.

    As with Dave, please do identify yourself 🙂

  7. Anonymous says:

    It is sad that the odf "alliance" spends so much of its time worrying about open xml and doesnt seem to be interested in odf at all. hopefully now they’ll see the cost of being so negative.

  8. Dave S. says:

    In a manner of speaking, a file is code for the appropriate interpter. Doesn’t MSO-XML encode a document according to the requirements of the MS Office 2007 application?

    It is sad that the MSO-XML supporters spend so much of their time worrying about odf and not seem interested in getting along. Hopefully now they’ll see the cost of being so negative.

  9. Jimmy B says:

    Is it really the ODF alliance writing this or is it just the ODF alliance publishing IBM written/funded documents.

    And allthough the ODF alliance claims to have many members (free memberships ahoy) isn’t it just an IBM funded front for IBM’s own public relations machine?

    Who is really behind this ODF alliance?

    Just a bunch of anonymous publications and almost no info on their internal structure.

    Not exactly an Open organization…

  10. Dave S. says:

    “For what it’s worth, nobody is discussing “Parsing of Office 2003 code”, what is being discussed is “document format parsing code in Office 2003.” It’s not clear if the author of the document was in a hurry when this was written, or if they just lack a basic understanding of the problem”

    The document is encoded, so parsing the Office 2003 code is what is being done when an Office 2003 file is opened for reading.

    As to ID – what difference could it possibly make in an open exchange of ideas? Like you on ODF, I’m just trying to keep the technical part of this honest and accurate.

Skip to main content