Andy Updegrove’s Indian Fantasy



This morning, in my usual routine, I sipped a cup of coffee while reading the latest news around the Open XML file formats. I have a few blogs I follow regularly, and I also usually do a few searches to see what has been written in various media.


The 30- day contradiction comment period for Open XML just came to a close this week, so there were several new articles and blog posts about the continuing ISO standards process. (For those who don’t know, the contradiction comment period is when the “national bodies” (NBs) take a look at whether a proposed standard prevents use of any existing ISO/IEC/JTC1 standards, and if so they can submit their perceived contradiction to JTC1 for possible resolution. I’m not going to get into all of the procedural details here, but Brian Jones has a good overview if you’re curious.)


Anyway, in following links between various articles and blog posts this morning I came across a piece by Andy Updegrove in which he wrote this:

“According to one story, at least one of these countries (India) was considering responding by abstaining from voting, in protest over the extremely short amount of time provided to review the voluminous specification. Instead, it appears that it opted to knuckle down, finish its review, and submit contradictions instead.”


Now, this piece rather surprised me. I happened to have returned from India less than 24 hours ago, and among other things, I was involved in some meetings and discussions with the people in India who have the responsibility of deciding whether to submit contradictions. I was very impressed with their thoroughness and interest in hearing all sides of the issues, and I certainly didn’t know that they had submitted any contradictions to ISO.


Well, maybe Andy knows something I don’t, or maybe he’s just quoting somebody who got the facts wrong. There’s been rather a lot of that getting-the-facts wrong stuff lately when it comes to file formats, you know. 🙂


So I clicked on the link he provided, thinking I’d read about India “submitting contradictions” to ISO. But that article doesn’t say any such thing! And in fact, when I followed the link I immediately recognized the article because I happen to have read it in New Delhi on Monday morning, a good 12 hours before India had responded to ISO on this matter. I remembered it because I got a good laugh out of the “against the human spirit” comment in it — the Times of India never fails to make me laugh.


So let me get this right. Andy Updegrove makes up an Indian response to the ISO Fast-Track process, then to support his fabrication he links to an article that was published before India had even responded to ISO, and one which in any event makes no mention at all of India submitting a contradiction. And then, just to help get the word out, an IBM VP links to Andy’s article to help him spread this fabrication. (Presumably they do it this way so that nobody at IBM is actually telling lies, they’re just linking to the lies others tell on their behalf.)


Wow. Desperate times for IBM, it seems. And their respect for the ISO standards process is downright palpable, wouldn’t you say?


By the way … unlike Andy and his boosters, I won’t actually discuss the details of what countries have filed during the 30-day comment period. That’s not appropriate at this time, and certainly not appropriate for those who aren’t actually speaking from firsthand knowledge. But I will say this: India’s not the only misrepresentation on the list they’re passing around today, by a long shot. It just happens to be the country where I’ve spent the last week, so I thought I’d comment on that specific one for now.

Comments (32)

  1. panchi says:

    nice catch !!

  2. ODF User says:

    The link on Andy’s sight now points to a rather telling article of just how the Indian govt. really feels about OOXML. Perhaps a simple mistake was made by Andy? He has, after all, gone far and above most folks in collecting information about both ODF and OOXML.

    I see that Andy has corrected his mistake. Based on your high-horse attitude and lack of respect for anybody who is not drinking the MS Kool-Aid I won’t hold my breath waiting for you to write any kind of retraction or update to your article.

    It’s nice to see that all the astroturfers for MS have are ad hominem attacks on others, while folks like Andy have the opinions and decisions of whole nations to augment his position on standards and what the definition of Open is.

  3. dmahugh says:

    Mr. User,

    The link on Andy’s site still points to the same article it always did, the one that says nothing about submitting contradictions.  I don’t know what you mean by "corrected his mistake" — can you clarify what has been corrected?

  4. anonymous says:

    I think you have misread Andy’s article. He describes the Times of India story as suggesting that India would abstain from voting. He does not imply that he got from that article the information that India submitted contradictions.

    On another topic, your palbable contempt and condescension for the Times, Andy, and anyone opposing MSOOXML offended me. Only after reading most of the article did I notice that you are an MSOOXML Evangelist posting on a Microsoft site. I guess that explains your attitude.

    You talk of making up facts, desperation, lack of rescpect for ISO, and "boosters." Perhaps you are looking in a mirror?

  5. Confused says:

    I don’t understand the outrage at all.  Andy says the story linked to mentions India possibly abstaining in protest to the length of the spec and the short time to review it.  The article does say something like that.  Then Andy assumes that because India shows up on a list of countries which have submitted a contradiction, they’ve decided not to protest.  What Andy says makes perfect sense to me.  His list of countries may be wrong, but I doubt he’s making it up.  He got the list from somewhere and I think you are reading the sentence with the story link as his proof that India submitted a contradiction.  He’s only using the linked story as an example of the discord possibly created by having only 30 days to review a long document.  Wow.  I can’t believe how you can read the simple things Andy wrote and stretch into a conspiracy backed by IBM to misinform!

  6. dmahugh says:

    Anonymous,

    Yes, I have quite a bit of contempt for those who have been spreading FUD about file formats lately and trying to convince people that having freedom to choose is a bad thing.  And I have a hard time understanding what Andy is trying to accomplish in his post.  Has he actually read the comments that were submitted?  If so, then — as will be obvious to everyone after those comments become public — he has misrepresented their content.  And if he has not actually read them, why does he say things like "most or all lodging formal contradictions with Joint Technical Committee 1," which is a very specific characterization of their content?

    Confused,

    Yes, 30 days is a short time to read the entire Open XML spec.  But as Andy surely knows, that’s not what this 30-day period was about — it was about identifying contradictions with existing ISO/IEC/JTC1 standards.  The 5-month DIS ballot period is when the spec gets reviewed in technical detail.  The claim that people have been put in the awkward position of having to review the entire spec in 30 days is something that has been repeated many times in recent writings from various people at IBM, and I’m disappointed to see Andy adding his voice to that chorus.

  7. Andy Updegrove says:

    Thanks to those that understood what I wrote when I wrote it (and did not change).  The author of the article I linked to says that India was considering abstaining – he wrote that, not me.  My direct source – someone who knows of a certainty who filed responses, and did *not* abstain – supplied me with the information that India had, in fact responded.  I doubt that I’ve been misinformed, and expect you to read confirmation of the information in due course at the JTC1 site.

    So, unfortunately, the title of this blog entry (and its content) is unfair, inaccurate, and – oh yes – wrong.  

    Perhaps if its author truly has "quite a bit of contempt for those who have been spreading FUD about file formats," he’ll retract his statements, and replace them with an apology.

    That, of course, would only be the ethically consistent thing to do.

    –  Andy

  8. dmahugh says:

    Thanks for commenting, Andy.

    My point is that you didn’t say that India <i>responded</i>, you said they had <i>submitted a contradiction</i>.

    I respect your opinion, but I stand by what I’ve said.  Like you, I believe a full exposition of the facts will prove me right.

    Setting aside the question of what India did or didn’t do, if you were told that these 19 countries all had negative things to say about Open XML then I believe you will appear to have been misled when the facts come out.  And I continue to have contempt for those who engage in such campaigns, regardless of whether you’ve been a victim or perpetrator.

    – Doug

  9. Stephane Rodriguez says:

    Duog,

    I know nothing of what’s going on here, so I’ll leave it for others. But I’d like to comment on one thing you said above :

    "Yes, I have quite a bit of contempt for those who have been spreading FUD about file formats lately and trying to convince people that having freedom to choose is a bad thing."

    Can you name two products that can render Office 2007 documents?

    Thanks

    PS : the point of the question is of course that, by fast-tracking all this stuff, it is obvious that Microsoft has actually called for contempt in the first place rather than allowing a competing vendor to have the time to come and play fairly. Ideally, if there is a non-Microsoft vendor that can render Office 2007 documents (accurately of course), then it is a perfectly fair review of specs being pushed to become so-called international standards.

  10. dmahugh says:

    No, Stephane, I can’t at this time.  And I think that’s an appropriate subject of debate, as is the University of Central Florida’s test suite that demonstrates there are not two separate implementations of ODF that will render anything more than the simplest of documents in a consistent fashion.

    What I don’t think we should be debating at this time is how to best characterize the status of the private inner workings of the JTC1 process.  I’d rather let the process run its course, with public disclosure coming at the appropriate times rather than through leaks.

  11. BS Meter says:

    Funny, you don’t seem to be down on Brian Jones for the same offenses as Andy Updegrove!  He too is describing "the status of the private inner workings of the JTC1 process."

  12. Marbux says:

    Doug, your blog article states:

    <blockquote>(For those who don’t know, the contradiction comment period is when the "national bodies" (NBs) take a look at <u>whether a proposed standard prevents use of any existing ISO/IEC/JTC1 standards</u>, and if so they can submit their perceived contradiction to JTC1 for possible resolution. I’m not going to get into all of the procedural details here, but <a href="http://blogs.msdn.com/brian_jones/archive/2007/01/29/explanation-of-the-iso-fast-track-process.aspx">Brian Jones has a good overview</a> if you’re curious.)</blockquote>

    Given your expressed interest in accuracy, you might consider informing your readers that you and Brian Jones unquestionably have spread inaccurate information on the purpose of the contradiction mechanism at JTC-1.

    Neither you nor Brian has so much as mentioned that there is international law defining the scope of a proper contradiction, so I suspect you were not aware of the law.  Moreover, I have information that Microsoft convinced ANSI-INCITS not to file a contradiction based on the same unquestionably erroneous meaning of "contradiction" in the ISO JTC-1 process. With your concern for accuracy, I hope that you might carry the ball to let the folks at INCITS know Microsoft’s legal advice was wrong.

    You can find my thorough legal analysis of this issue, with citations and links to the relevant law, in my article entitled,  <a href="http://opendocument.xml.org/node/238">"The case for a valid contradiction of Microsoft Office Open XML at ISO has not been rebutted."</a> In summary, according to the international Agreement on Technical Barriers to Trade article 2 section 2.2, the national bodies are <u>required</u> to contradict any fast track proposal that was "prepared … with a view to or with the effect of creating unnecessary obstacles to international trade."

    That legal issue was briefed in the <a href="http://www.grokdoc.net/index.php/EOOXML_objections#Introduction_and_Summary">introduction  of the Groklaw document</a> laying out grounds for contradiction of fast track processing of the Microsoft/Ecma draft standard. Microsoft has yet to respond under the correct legal criteria, which is why I concluded that the case for contradicting Ecma 376 stands unrebutted. I certainly hope that Microsoft will address its formal responses to the national standards bodies’ contradictions to the correct legal criteria.

    The correct legal issue is whether Ecma 376, if adopted as an International Standard, would grant Microsoft a competitive advantage not equally available to all vendors. One need go no further than the fact that Ecma 376 fails to include the specifications for the binary formats to conclude that all of that claimed compatibility with the binary formats is for the benefit of a single vendor, Microsoft.

    Since that compatibility is the underlying premise for all Ecma justifications offered Ecma 376’s adoption as an International Standard, it is plain that Ecma 376 would confer on Microsoft an exclusive monopoly on migrating the binary formats to Ecma 376 formats, the very purpose of Ecma 376. That is plainly an "unnecessary obstacle[] to international trade." Therefore, if for no other reason, <a href="http://opendocument.xml.org/node/238">"The case for a valid contradiction of Microsoft Office Open XML at ISO has not been rebutted."</a>

    My article also cites and quotes the law establishing that Microsoft’s "freedom of choice in file formats" position is forbidden by international law. You adopted the same argument in your comment on this page where you spoke of "having freedom to choose."

    Leaving aside the fact that is an argument which would only make sense to someone who has no understanding of the barriers to software interoperability posed by inconsistent file formats for the same software functionality, 103 nations of the world have agreed by international treaty that you are wrong, that there <u>must</u> be only one International Standard for given software functionality.

    So I hope you might correct that misstatement in your article as well. I am truly sorry that you have been deceived by the Microsoft spin doctors, but these issues are far too important to be resolved on the basis of anything but accurate information, as I am sure you will agree.

    By the way, Andy Updegrove <a href="http://www.consortiuminfo.org/standardsblog/article.php?story=2007020812133683">has responded</a> to your criticism, making it clear that he got the list of contradicting nations from JTC-1 and that JTC-1 has now published a revised list adding another country (Italy) that was inadvertently left off the first list, so the total is now up to 20 nations. Unless you have different information, you might consider making an apology for your <u>ad hominem</u> attack on Andy, which was a <a href="http://en.wikipedia.org/wiki/Ad_hominem#Ad_hominem_as_logical_fallacy">logical fallacy</a> in any event.

    I have saved a copy of this post’s text, so that there is a record you have been made aware of these facts. I sincerely hope it will not become necessary to to prove that fact and that you will correct the errors in your article and comment.

    — Marbux

    (a retired lawyer in the U.S.)

  13. Marbux says:

    Doug said:

    <blockquote>No, Stephane, I can’t at this time.  And I think that’s an appropriate subject of debate, as is the University of Central Florida’s test suite that demonstrates there are not two separate implementations of ODF that will render anything more than the simplest of documents in a consistent fashion.</blockquote>

    I think you confuse the ODF standard with its implementations. You also wildly exaggerate the extent of the remaining differences in rendering by OpenOffice.org and KOffice. Perhaps you might provide a link to the results of the tests you refer to? Full interoperability is not an easy task. It takes time for developers to implement it. At least the OOo and KOffice developers are tackling the problem and have achieved far better results than the Microsoft-Clever Age plugin does in either direction.

    Of course, Microsoft might consider abandoning the secrecy of its APIs for converting its binary file formats to Ecma 376 as well as disclosing the specifications for the binary formats, which are still used by the major Office apps for their internal processing.

    Those disclosures would be wholly appropriate for a company whose spinmeisters paint it as a champion of software interoperability. That is especially so because: [i] the disclosure of those items has been required for several years by antitrust orders in both the U.S. and in Europe; [ii] Microsoft now wants Ecma 376 adopted as an International Standard for migrating files in those formats to Ecma 376 formats. Ecma 376 is not open because of Microsoft’s refusal to disclose those formats and APIs.  

    I am told by OpenDocument Foundation developers that they could quickly finish their Microsoft Office plugin to provide full interoperability between Microsoft Office and OOo if Microsoft disclosed the specifications for its binary file formats.

    May I respectfully suggest that interoperability is an issue that Microsoft can not win given its lack of cooperation in enabling full interoperability between MS Office and ODF applications? Particularly given that Steve Ballmer himself has said the Microsoft-Clever Age plugin will not provide full interoperability between MS Office and ODF apps? Particularly given Microsoft’s track record in concealing the information needed to develop full interoperability?

  14. I hope says:

    your not the captain of this sinking ship

  15. dmahugh says:

    Marbux,

    Regarding accuracy of the definition of contradiciton, the only thing that’s relevant is what it says in the JTC 1 directive on this matter.  I’ve seen your analysis suggesting otherwise, with the appropriate caveat in its final sentence (“Nothing in this article is intended or should be understood as legal advice.”)  You’re certainly entitled to your opinion, but that doesn’t change the actual JTC directive.

    Regarding Andy’s comments, I think you’re conveniently blurring the line between comment and contradiction, but I won’t get involved in discussion of the content of comments submitted.  It’s disappointing that some feel a need to go there, rather than respectiing the defined process and letting it take its course, and this is the last I’ll respond to speculation about the content of the comments.  Time will tell, and the truth will take care of itself.

    Here’s a link to the University of Southern California’s test suite comparing KOffice and OpenOffice’s implementations of ODF.  Over 300 specific differences are identified.  I’m not a lawyer like yourself, but it’s not clear to me how my comment that these hundreds of differences demonstrate that “there are not two separate implementations of ODF that will render anything more than the simplest of documents in a consistent fashion” is a “wild exaggeration.”

    > I am told by OpenDocument Foundation developers that they could quickly finish their
    > Microsoft Office plugin to provide full interoperability between Microsoft Office and OOo
    > if Microsoft disclosed the specifications for its binary file formats.

    So they could quickly make the plugin work with the binary formats, even though they don’t know the details of those formats, but can’t make it work with Open XML, which has 6000 pages of published and publicly reviewed documentation?  Interesting.

    Feel free to save copies of whatever you’d like, I stand by everything I’ve said.

  16. Phobos says:

    What I find telling in Andy’s post–something that you have glossed-over completely–is the fact that (now) 20 nations have submitted contradictions with respect to microsoft’s proprietary document format.  (Call it what you wish, but we know it for what it is.)

    This is, in Andy’s own word choice: UNPRECEDENTED; which makes it pretty obvious that "we" aren’t the only ones who know OXML for what it really is: a foundation for undocumented extensibility, in the spirit of microsoft’s past undocumented API bag of tricks.

    Really now.   It’s nothing more than yet another attempt at vendor lock-in and tilting the scales against your competitors.

    Be honest about it, for once.

    You guys need to get yourselves behind native ODF support, quit pansying-around with a 0.5-posteriored plugin, stop abusing the word "innovation" in every other sentence, and quit basking in the adoration of mere paid sycophants.  (That wasn’t meant personally, BTW.)

    "So they could quickly make the plugin work with the binary formats, even though they don’t know the details of those formats, but can’t make it work with Open XML, which has 6000 pages of published and publicly reviewed documentation?  Interesting."

    Might be a pretty good indicator of how intuitively "Open" OXML ISN’T.  6K pages….hmmmmm….and it left out the extension hooks, too, I presume?

    I see this as being an echo of WordPerfect and microsoft’s undocumented APIs, as well as microsoft’s typical "one-way street" interoperability practices.  Perhaps you should do some reading INRE: Comes -v- microsoft.  For us, who have been watching you for a couple of decades, this is nothing new–nor is your response to the situation.

    Typical.

    Let us, indeed, see where the "defined process" takes us, along with the level of back room maneuvering it takes microsoft to get what it wants; or, barring that, how much kicking and fussing will ensue after microsoft doesn’t get what it wants.

    How much hand-holding and drool-wiping protection does the "800lb gorilla" need, anyway?

    The day that you guys can compete on the true technical merits; with equal barriers to entry and participation for yourselves, as well as your competitors; and don’t need to be propped-up by exclusionary OEM channel contracts–without equivocation–is the day you folks will have something to say that is worth listening to.

    The problem, for microsoft, has been that the "geek public" has been aware of this for all too long.  We have learned and are now addressing the public on its own terms.

    The present problem, for microsoft, is that the "pedestrian public" is becoming more aware of this as time passes; and, with the advent of vista, and its native document format changes; amidst this "poorly-timed" discussion of Open Standards; you guys are going to be treated to quite a bit more resentment than even you are used to fielding.

    The evidence of this is in the 20 dissenting nations, with respect to the OXML OSI bid.  In this, we can say that what we have been doing has been effective, and we really don’t plan on stopping now.

    More people are realizing that you guys (presumably) have the chops to kneecap a competitor in a dark alley; but, don’t have the chops to compete on a level playing field.  More people are expressing the sentiment that they really wish that Vito Corleone would tell you FUD-guys to take a long, hot bath; because your verbage really reads more like Bill Cosby’s "Brain Damage" routine, rather than anything else that might be called credible by the most ridiculous stretch of the imagination.

    Thanks, buddy, for giving me a wonderful URL to point people to, saying, "Look at the transparent BS these guys have to keep churning-out to prop up market retention for their shoddy products."

    They understand what Andy is saying just fine; however, when they read guys like you…Whiskey Foxtrot Tango – Over?

    Game over, and we install an ODF-compliant office productivity suite.

  17. dmahugh says:

    Thanks for sharing your thoughts, Phobos.  I think I’ve responded to all of these points in various places already, so I’ll not repeat that here, although I must say that if you’ve really been watching me for the last couple of decades, you’ve been on quite a ride.  🙂

    Hey, regarding abusing the word “innovation,” could you point me to an example of a non-OOXML solution along the lines of what Mindjet has done?

    And I certainly do appreciate all the traffic you and others are sending my way.  Link away!

  18. Stephane Rodriguez says:

    "Hey, regarding abusing the word innovation, could you point me to an example of a non-OOXML solution along the lines of what Mindjet has done?"

    Twice wrong I am afraid.

    MindJetManager creates a DOCM file by embedding predefined macros. This is just one of the stuff that is not defined in OOXML, therefore not interoperable.

    Also, MindJetManager addin is an application running on top of Word 2007, meaning that Word 2007 renders all this stuff. Again, much of the rendering semantics is not defined in OOXML, therefore not interoperable.

    This is certainly a bad example when it comes to interoperability.

    This solution will be worth mentioning when the same code binaries run in OpenOffice. Until then…

  19. Phobos says:

    Tip of the hat to Stephanie Rodriquez on that one; which brings to the fore microsoft’s undocumented API tricks; but I’m certain that microsoft has become suddenly inconsistent with their own past practices (something I have yet to see) with regard to the true openness of everything office XML:

    http://www.groklaw.net/article.php?story=2007020819534335

    These very practices–not technical desirability, or even superiority–are the reason that we are even discussing microsoft and its panic over losing the document format space at all.

    Glad you appreciate the hits; however, since virtually every hit results in a conversion to a native ODF productivity suite, you may not be all that jazzed by the long-term results.

    17 out of 30 JTC 1 Committee P-Countries.  

    It’s gonna take a lot of that microsoft bribe money to get yourselves over that obstacle, isn’t it?

    Remember, Doug, microsoft has never won anything because they were objectively better on the technical merits.  They’ve won {x|y|z} simply because they have always been the more ruthless criminal gang.

    Truth, ethics and a level playing field really are meaningless to you guys.

  20. Marbux says:

    Doug, you said, regarding the accuracy of my legal analysis  on the proper scope of “contradiction” within the context of the ISO standardization process, you responded, “the only thing that’s relevant is what it says in the JTC 1 directive on this matter.”

    I note that the term is not defined in JTC-1 Directives.

    But were you correct, that erroneous position would have several repercussions, notably:

    [i] Microsoft’s “a contradiction is only proper were it impossible for an ODF file and an Ecma 376 file to coexist on the same system” argument would be wrong; there would be no limits whatsoever on grounds for valid contradiction because the term is undefined in the directives;

    [ii] were Ecma 376 to be ultimately adopted by ISO under procedures that forbid national bodies from complying with the cited treaty, ISO/Ecma 376 would not be entitled to the status of an International Standard under the treaty; and

    [iii] national governments would then be forced to enforce the treaty’s requirement that international standards not be applied if they would have the effect of creating unnecessary obstacles to international trade; in other words, we are all off to the World Trade Organization dispute resolution process.  

    You also have apparently overlooked the portion of my article pointing out that the JTC-1 rules for appeals of any JTC-1 act or failure to act — by their own terms applicable to contradiction decisions — specifically provide that such an appeal is improper unless effects on competition are involved. So JTC-1 Directives agree with my legal analysis and not with Microsoft’s position that to my knowledge has never so much as mentioned the controlling law. I note that you adopted the same tactic.

    I also note that JTC-1 Directives do not provide for “comments” rather than “contradictions” being submitted to the process by NBs during the contradiction phase of a draft standard’s processing. May I presume that Microsoft is preparing to challenge one or more documents because they did not use the word “contradiction?”

    Your refusal to discuss the contents of the contradictions is also troubling. The customary practice of the JTC-1 Secretariat has been to make contradictions public and distribute them to JTC-1 P-Members promptly after the close of the contradiction period. But she abruptly decided to delay publication of the Ecma 376 contradictions until after Microsoft/Ecma have responded to them.

    Perhaps you might be so kind as to explain whether Microsoft/Ecma requested that delay and the reasons? I note that the departure from customary practice has the effect of reducing the amount of time available to people other than Microsoft to review the contradictions. Was that the goal?  

    On your exaggeration of the rendering differences between KOffice and OpenOffice.org, it was an exaggeration because you attempted to leave the misimpression that the differences were due to the weakness of the ODF standard.

    In fact, if the reader checks the summary of differences on the site you link, they will learn that the rendering differences are almost entirely due to the differences in feature sets supported in the two office suites.

    The reader will also learn that you did not link to the University of Florida site, but to the results of tests posted at the OpenDocument Fellowship site where ODF developers are working on improving interoperability between their applications in broad daylight.

    It would be useful if Microsoft were to publish its own results of testing using the University of Florida test suite for Word-OOWriter interop using the Microsoft-Clever Age plug-in for Microsoft Word.

    I am sure the world would appreciate hard data rather than market-speak on that issue. The KWord and OOo developers have made a public record of how well they are progressing at implementing full interop. Will  Microsoft do the same or will it insist that someone do it for them?

    The only barrier to to full two-way interop with ODF is correct implementation of the same feature sets in the applications whose developers are working to achieve full interoperability.

    That barrier has almost nothing to do with weaknesses in the ODF standard. (I say “almost” because the two development teams have — as was expected — encountered a very few areas where the standard needs further detail added to reduce opportunities for inconsistent rendering. Those edits to the standard are being worked on at OASIS.)

    But in sum, the ODF standard is robust, completely XML,  and completely open. That contrasts markedly with Ecma 376, which does not specify the functionality for numerous proprietary binary blobs that appear in Microsoft’s own version of Office Open “XML” but not in Ecma 376 and Ecma 376’s lack of specification for the functionality of many of the XML tags, which are just two of the numerous and documented major barriers to full 2-way interoperability Microsoft included in Ecma 376.

    The indisputable fact that Ecma 376 would, if adopted by ISO, create large advantages for Microsoft not available equally to all competitors renders Ecma 376 ineligible as an ISO standard. International standards are about leveling the competitive playing field, not tilting it as Microsoft’s managers obviously believe.

    Your example of Mindjet Manager as a sterling implementation of Ecma 376 indicates the opposite of your point.  That application’s relevant features enable only 1-way importing of its output into a single vendor’s application where the output is refactored by scripts into a rendering that can not be exported back to the originating app. The specifications for the proprietary embedded scripting language are not not disclosed in Ecma 376.

    Mindjet Manager illustrates that Ecma 376 is the (mostly) XML equivalent of Microsoft’s Rich Text Format, featureful enough to send data to Microsoft Word but not featureful enough to complete the round-trip to the originating non-Microsoft app.  

    In conclusion, I have no way of knowing whether your response to my post was a deliberate evasion of the constructive criticism I raised or was instead based on faulty information you have been provided. But either way you are wrong. (And either way, your position bespeaks a lack of sound judgment in your well-publicized efforts to have the OpenDocument pages on Wikipedia edited to your liking.) Your article should be corrected.

  21. dmahugh says:

    Hmm, Phobos, I guess I’ll take that as a “no.”

    And I probably need to go look up “ad hominem attack,” so I can better understand what all these anonymous commenters are so upset about. Ah, Wikipedia seems to have a reasonable definition. Applying that definition to the thread above I’ll leave as an exercise for the reader. 🙂

  22. dmahugh says:

    Marbux, your claims of 1-way interoperability between Mindjet and Office will come as a bit of a surprise to all the people who’ve seen it in action.

    Why just last week, speaking of India, I demonstrated at Linux Asia how I could export a Mindmap to Word, add a new topic in Word called Linux Asia with a bunch of notes attached, then import the result back into Mindjet and see my original Mindmap, 100% fidelity (based on the DOCM and nothing else), with a new cloud showing the Linux Asia notes I had entered.

  23. Marbux says:

    I just realized that I overlooked responding to one of your points, the one that evaded rather than addressing the secrecy of the binary file formats and Microsoft’s APIs for adding native file support to Office apps.

    Yes, while the documentation Microsoft has released on its XML office file formats has been helpful, there remains the problems posed by the secrecy of the relevant Office APIs and the binary formats, the many  binary blob dark objects whose functionalities are not defined in Ecma 376 but are open books for Microsoft only.

    Microsoft is simply ignoring that the disclosure of the APIs and the binary formats is required by two antitrust injunctions, one in Europe and one in the U.S. Instead, Microsoft has attempted to sneak an "open standard" through the ISO process that discloses neither the APIs nor the specifications for the binary formats, even though the migration of documents stored in those formats to XML is the sole justification Ecma offers for Ecma 376 being adopted as an ISO standard.  

    It is undeniable that Microsoft seeks through the ISO process to establish an exclusive monopoly on migrating those billions of legacy binary documents to XML and only to Microsoft’s flavor of XML + binary blobs.

    Will Microsoft ever address the multitude of secrecy issues raised by Ecma 376 or will it continue its attempts to corrupt the ISO standard development process?

    That is the real issue. If you do not address it, then you take the side of the Microsoft spin-meisters rather than engaging in principled discussion. It is your credibility at stake. It is your decision whether to protect it. All I can do is to point out the choice you face. Your readers will know what your choice is by your response.

  24. dmahugh says:

    > It is undeniable that Microsoft seeks through the ISO process to establish an exclusive monopoly

    > on migrating those billions of legacy binary documents to XML and only to Microsoft’s flavor

    > of XML + binary blobs.

    Now, I’m just some guy who works here.  And I wasn’t around when a lot of these decisions were made.  (I’m new enough to Microsoft that I received an email when the Wikipedia story broke from old friends saying "wow, you work at Microsoft now?")  So the following is just my personal opinion, and I have no idea whether it’s consistent with what others might say on this topic.

    But it seems to me that Microsoft has a unique and important responsibility to those people who have invested in the creation of billions of Office documents, nearly all of which are currently stored in the legacy binary formats.  As the vendor of the software that they used to create those documents, we are obliged to help them move those documents forward into open standards-based formats, and to make that process as simple, straightforward, and inexpensive as possible.  This responsibility doesn’t fall on IBM, Sun, Oasis, the OpenDocument Foundation, or anyone else — as the vendor who enabled the creation of these documents, that responsibility is ours alone in my opinion.

    And we’ve addressed it, by providing a clear upgrade path to the Office Open XML file formats that is simple and reliable.  Open XML is the only available file format that provides this full-fidelity upgrade path, and I sure don’t see that as a problem.  Rather, I see that forward migration as a messy and difficult task that we must lead, and we’ve accepted that responsibility and given it our best effort.

    If you take a 100-year view of where we’re at, proprietary binary document formats are on the way out, and open standards-based (and generally XML-based) formats are on the way in.  And Open XML’s uniquely high level of compatibility with the vast majority of existing documents is going to help expedite this transition to the standards-based future, which is a good thing for all involved.

    P.S. I’ve enjoyed the dialog, but I must warn you I won’t be so quick to release comments from the moderation queue during the next day or so.  We have guest in from out of town, and showing all of those folks a good time is going to take precedence over document-format discussions for a while.  Enjoy the weekend!

  25. Marbux says:

    Doug, I assume the accuracy of what you say and therefore apologize for my misconception that round-tripping had not been achieved by Mindjet Manager.  That said, perhaps you might also discuss my point that it is dependent on a Microsoft proprietary scripting language that is not specified in the Ecma 376 draft standard? Do I have that wrong too? So far as I am aware, the only featureful word processor that provides VBA support is WordPerfect. I am aware that Novell is working on an open source clone VBA for OpenOffice.org.

    However, I note in your earlier article that Mindjet Manager also is dependent on MS Office 2007 "ribbon" APIs. Microsoft has a patent pending on the ribbon tool and has specifically disallowed its emulation by products that compete with Microsoft Office. See <http://en.wikipedia.org/wiki/Ribbon_(computing)> for  a concise explanation.

    Would you please explain how Mindjet Manager’s feat might be replicated in applications that do compete with Microsoft Office? –Particularly given that Microsoft has granted no rights whatsoever for other developers to implement Ecma 376 unless they obtain a separate license from Microsoft? For the relevant legal analysis of Microsoft’s Open Specification Promise, see <http://www.grokdoc.net/index.php/EOOXML_objections#Patent_rights_to_implement_the_Ecma_376_specification_have_not_been_granted&gt;.

    As I said before, the reasons for Ecma 376’s ineligibility  as an international standard are legion. The relevant barriers to other developers’ full implementation of Ecma 376 run the gamut from merely technical to legal.

    While I apologize for my apparent error, I still believe Mindjet Manager speaks more to Ecma 376’s weaknesses as an international standard than to its strengths.

    Oh, and will you respond to my other points in my prior post?  As it is they stand unrebutted.

    Best regards,

    Marbux

  26. Marbux says:

    Doug, I am very pleased that you are at least grappling with one of the major issues, the secrecy of the binary formats. That is a welcome sign that you are open to principled discussion.

    But Microsoft’s obligation to migrate its customer’s legacy files to XML is an insufficient justification for the secrecy. First, we are discussing a proposed International Standard for migrating those legacy documents to XML, not one vendor’s obligations. An International Standard must not, under the relevant treaty, bestow a competitive advantage on one vendor not available to all vendors.

    Second, there is unquestionably a market requirement for migrating those legacy files to XML, but no market requirement at all for migrating them to Ecma 376 XML. There are no featureful applications that implement Ecma 376 in two-way round-tripping mode.  

    I can not, for example, set an instance of MS Office to generate pure Ecma 376 files. In Office, I can only generate files that are compatible with Ecma 376 but include binary blob dark objects that only Microsoft Office can understand and utilize without reverse engineering.

    The only relevant market requirement is for migrating those legacy binary files to the existing International Standard formats, the OpenDocument XML formats.  See e.g., the linked page that tracks government decisions to adopt OpenDocument worldwide. http://www.opendocumentfellowship.org/node/91

    As explained in some detail in my article linked in my first comment, the relevant treaty forbids inconsistent standards that inconsistently duplicate or overlap with existing international standards. There is no market requirement for Ecma 376 to be adopted as an International Standard purely as a matter of law.

    Microsoft has no legal duty to implement OpenDocument. The company is free to market a product that does not conform to international standards and to tout the claimed benefits of its non-standards-based approach.  But it thereby forfeits the right to market its product to those who are required by law to procure only software that conforms with International Standards, such as all governments that are party to the international Agreement on Government Procurement, governments that are commanded by the same Agreement’s safety valve not to apply International Standards that do give one vendor a competitive advantage not available to all competitors in the relevant market.

    The long and the short of it is that Microsoft is legally required to work with the OpenDocument standard if it wishes to regain eligibility in the government software procurement market.  As it stands now, it is illegal for governments signatory to the Agreement on Government Procurement to procure or renew Microsoft Office licenses.  

    As to your point that Ecma 376 is the only specification able to deliver full fidelity in migration of legacy Office binary files to XML, the OpenDocument Foundation’s plug-in has already achieved above 98 per cent fidelity in migrating the relevant word processing documents to what is expected to be adopted as the OpenDocument 1.2 standard.

    The argument that OpenDocument can not fully express the range of features in MS Office has never been made in a manner that would allow rebuttal, i.e., by identifying any single feature in MS Office that can not be expressed in OpenDocument formats. Even if there were, the OpenDocument specification is easily extended to accommodate such issues. We are discussing, after all, an eXtensible Markup Language ("XML").

    In short, you’ve been misled by Microsoft’s spin-meisters. There is no market requirement for Ecma 376 justifying its adoption as an International Standard.    The spin-meister’s argument that OpenDocument can’t meet Microsoft’s needs is a closely-related chestnut that does not withstand scrutiny and appeals only to the audience’s ignorance. The OpenDocument Foundation’s plug-in for Microsoft Office proves otherwise. (Microsoft observed its demonstration to the European Commission’s IDABC experts.)

    I realize you are in a difficult position where it may not be wise for you to cross swords with the company’s spin-meisters. But you might consider the ethical alternative of avoiding repetition of their deliberate disinformation.

    Best regards,

    Marbux

  27. Stephane Rodriguez says:

    Doug said "Why just last week, speaking of India, I demonstrated at Linux Asia how I could export a Mindmap to Word, add a new topic in Word called Linux Asia with a bunch of notes attached, then import the result back into Mindjet and see my original Mindmap, 100% fidelity (based on the DOCM and nothing else), with a new cloud showing the Linux Asia notes I had entered."

    What’s Linux Asia has got to do with it? I have lost you here. You are not implying that saying the "Linux" word twice in a sentence is an example of interoperability with Linux, right?

    What you are talking about here is how a Mindjet blob (apparently a piece of custom xml, correct me if I am wrong) can be read and updated within Word 2007. What is the enabler here Doug? The file format, or the macros?

    Of course the answer is the macros. What we are talking about here is applications, not file formats. So taking the Mindjet example in this file format interoperability thread is totally out of context.

    In my previous comment, I explained what it meant to take Mindjet in the file format interoperability context. Making clear that it did not much. I thought that was what we are all talking about.

  28. Stephane Rodriguez says:

    Doug said "as the vendor who enabled the creation of these documents, that responsibility is ours alone in my opinion."

    Then why do you constantly use the word "Open", if all there is here is one vendor doing their thing in private?

    Doug said "And we’ve addressed it, by providing a clear upgrade path to the Office Open XML file formats that is simple and reliable."

    Not just binary blobs are rewritten (new coordinate systems, new binding, new semantics), the MS Office team managed to introduce new ones where there was not before. For instance new VML blobs in several places. Outside Word, Excel and Powerpoint, only Internet Explorer can render VML blobs (and please, don’t say that VML markup is documented in ECMA 376, the rendering semantics isn’t).

    Doug said "Open XML is the only available file format that provides this full-fidelity upgrade path, and I sure don’t see that as a problem."

    Not true. Create a chart with an older version of Excel, then open it in Excel 2007. It does not look the same.

    Full-fidelity means a lot for customers out there, especially those expecting pixel-precise rendering. In fact, with the new chart drawing engine in Office 2007, you really fucked up.

    And that’s just one example.

  29. Luc Bollen says:

    Doug, Marbux,

    First, let me state that while I enjoy to read the point of view of both sides, I’m better convinced by Marbux’s arguments than by Doug’s ones.

    This being said, I start to be fed up by both parties standing on their positions and playing the Bugs Bunny game "I’m right – Your wrong – No, you’re wrong – No, I’m right…"

    It seems to me that ODF proponents have offered at several occasions to cooperate with OOXML proponents.  Up to now, my understanding is that OOXML proponents response is limited to "OOXML is necessary for billions of legacy docs. Point. ODF has no value for these billions of legacy docs. Point."

    When will a real discussion and cooperation start on how to really achieve interoperability and to offer a real choice to the customers : the choice of the applications they want to use, whatever file format is used to store the result of their work ?

    Thanks to both of you for providing a positive and constructive response, not an additional round of "I’m right – Your wrong" game.

  30. dmahugh says:

    OK, I’m going to respond to various open issues from the thread above, and then I won’t be responding here again today, perhaps not until later this week in fact.  Thanks for all the feedback, but I’ve given this thread over an hour of my time this morning, and that’s all the time I can spare for now.

    Marbux,

    Yes, Mindjet’s solution  uses VBA and ribbonUI, because their goal was to leverage the rapidly growing installed base of Word 2007.  As for how that could be replicated in an Office competitor, I’m not aware of the options there: are there Office competitors that offer something like the ribbonUI interface, so that a document can automatically customize the editor’s UI?

    Regarding the OSP, I’m not a lawyer but it’s my understanding that the OSP allows implementation of all details in Ecma 376 with no need for a license from Microsoft.  That’s the consistent message I’ve heard from our execs and spokespersons on that issue (which is certainly not my area of expertise).

    As for whether there is "secrecy" around the details of the binary formats, I see the core question here (regarding documentation of Office’s legacy formats) as "how big should the spec be?"  The Open XML spec has already been criticized for being too large, and if it included full documentation of every detail of the binary formats, that would probably make it an order of magnitude larger still.  Would that really be a good thing?  I think not — converting those documents is a one-time problem, and not something that should complicate the spec or its future implementations.  Yes, I agree that Microsoft has an inherent advantage in helping people convert those documents, just as IBM has an inherent advantage in helping people convert proprietary NSF files (say) into an open format, but I don’t think the spec itself should try to eliminate that advantage because it would add a huge amount of complexity with no payoff for anyone except during the one-time conversion of existing documents.

    When you say you can’t set an instance of MS Office to generate pure Ecma 376 files, I’m not sure I understand.  Can you give me an example of a document that doesn’t take advantage of any proprietary feature in Office (such as VBA macros or the ribbonUI extensibility) and yet Office still generates binary blobs that aren’t documented in the spec?  I can’t think of anything like that.

    The OpenDocument Foundation’s plug-in does seem to provide some forward migration for the binary formats, but I’m not yet familiar with the details of what it offers.  My attitude is that the market should be free to choose.  If the ODF plug-in is good enough for most people, then they’ll use it; if people find that Microsoft Office’s conversion works better, they can use that.

    As for your points made in prior posts, I just scanned through for questions that didn’t seem to be rhetorical in nature and I saw these:

    "Your refusal to discuss the contents of the contradictions is also troubling."  Not sure what to say to that.  I’ve been told (not by anyone at Microsoft, but by persons involved in managing the JTC1 process) that it’s not appropriate to publicly discuss the content of comments before they’ve been officially made public.  I don’t have any knowledge of why there’s a delay, and I don’t have any prior experience in these matters so I don’t know what’s typical.

    "It would be useful if Microsoft were to publish its own results of testing using the University of Florida test suite for Word-OOWriter interop using the Microsoft-Clever Age plug-in for Microsoft Word."  I agree, that would be interesting.  I’m not involved with the ODF translator project myself, but perhaps you could suggest this on their team blog.

    "Your well-publicized efforts to have the OpenDocument pages on Wikipedia edited to your liking."  I posted the full text of what I asked Rick Jelliffe to do, which I believe made it clear what I was asking for.  I’ll leave it up to readers to decide whether that was a request for him to edit pages "to my liking," and in any event, the concept of editing OpenDocument pages never came up.

    Stephane,

    I mentioned Linux Asia because that’s where I did the demo I described.  Feel free to not read anything into that detail.

    As for Mindjet’s use of binary blobs, those are for automating Word’s (proprietary) UI, which is separate from the question of interoperability between Mindjet’s file format and the Open XML format.

    I use "Open" because that’s its name; it’s a proper noun, as I’ve pointed out before.

    As for limitations on the fidelity of forward migration for binary documents, you make a good point that I probably shouldn’t say "full fidelity."  But I do think it’s accurate to say that the Open XML formats offer the most accurate/lossless upgrade path for the majority of existing documents, and for most documents there is no noticable loss of fidelity.  And I think the market will react predictably to those facts.

    Luc, thanks for your comments.

  31. Stephane Rodriguez says:

    Doug said " And I think the market will react predictably to those facts."

    That’s a political answer. If you keep repeating the Microsoft party line, we are not advancing a bit.

    The technical answer is to plug the old chart drawing engine when a document is open in compatibility mode.

    Microsoft really fucked up here. They must fix this.

  32. Stephane Rodriguez says:

    …or document the changes with the older chart drawing engine. That goes without saying.