Arnaud's Blog

Opinions on open source, standards, and other things

Let’s be clear: The Apache Software Foundation does NOT support OOXML.

OK, I’ll admit that nobody has claimed otherwise. Yet.

But in these days and age you are never too prudent. It wouldn’t surprise me to see this or other similar fancy claim being published eventually.

Indeed, in its desperate and last minute attempts to convince National Bodies around the world that OOXML is happening anyway so they might as well support it as an ISO standard, Microsoft is eager to claim support by as many companies and organizations as possible.

As evidence, in its latest OOXML propaganda open letter Microsoft lists IBM among other companies as having “already adopted (or announced adoption of) Open XML in their products”. This, despite a clear explanation of the contrary by Rob Weir, published two months ago! Does anyone believe they haven’t seen it or heard about this? I sure don’t. And if there was any room for misunderstanding Bob Sutor’s statement filled that in.

A colleague in a foreign country even reported that in a National Body meeting he had been confronted by a representative from Microsoft who was trying to silence him via intimidation and insistence that IBM supported OOXML contrary to what he was saying.

Microsoft’s oversight of IBM’s denials is clearly not accidental. It is part of a well crafted and continuous disingenuous plan to convince NBs at all cost. There is already so much evidence of Microsoft going far beyond what most would consider normal lobbying behavior it is sickening. For one, I’m not ready to forget the case of the NGOs in India. Talk about dirty practices.

But what really is at the bottom of Microsoft’s claims is that basically any software that handles XML supports OOXML. While technically this is true to a certain degree, such a bold claim without any further qualification is pure misinformation. Obviously, one of the advantages of using XML is to make your format, whatever it is, easier to handle, it’s one of the fundamental benefits of using XML. But as I previously touched on in my entry on XML vs Open, there is a big difference between being able to handle XML files at the XML level and truly supporting the particular format at hand.

Supporting OOXML. cannot be merely declared on the sole basis that a software can read OOXML files, or store OOXML files. If that were the case, then any XML parser could be said to support OOXML and the Apache Software Foundation could be said to support OOXML because its XML parser, Xerces, can read OOXML files (one would actually have to unzip them first but it’s not like Microsoft would stop at that kind of detail). But it takes much more than that to really support OOXML.

One has to understand the actual structure beyond the XML representation and the semantic associated to each and every piece of data found in an OOXML file. That’s what the 6000+ pages of the specification are supposed to define, unfortunately they do that extremely poorly.

The good news is that I don’t think Microsoft is fooling that many people. Based on my own observation of Microsoft representatives and the way they talk to people they seem to be completely oblivious to the fact that they appear as if they think the people they are talking to are too stupid to see through their tired arguments. I’ve got news for them: people aren’t that stupid. Thankfully. And I’m hopeful the results at the end of the month will make that clear.

The other good news is that whether OOXML gets approved or not, I believe Microsoft will pay a high price for all of its mischief and its image will come out of this badly damaged, something they can only blame themselves for.

In the meanwhile, don’t take for granted any claims of support for OOXML from Microsoft. The fact that Microsoft claims IBM has adopted OOXML can only make one wonder about all the other companies they list…


March 19, 2008 Posted by | standards | , , , , | 15 Comments

How many bad standards does one need in a given domain? Zero.

A lot of the debate around OOXML has focused on whether it is good to have competing standards or not. The debate started from the simple fact that Microsoft decided to create its own standard for office applications rather than adopt the established ISO standard for office applications: ODF.

While there is clearly a need for evolution and there are times when it makes sense to introduce a new standard to replace an old one there is no doubt in my mind that in general there is much more to lose from having multiple standards rather than a single one than there is to gain.

Interestingly enough I should point out that Microsoft defends that very point at times. In the case of XML for instance, when the W3C introduced XML 1.1 to address some internationalization limitations in XML 1.0, something important to many non-western countries, Microsoft voted against XML 1.1 arguing that the introduction of a new version of XML would be too disruptive!

Yet the tactic of introducing a competing standard to disrupt the status quo is common practice for the Redmond company. For instance, in the Web services management area, an area not so visible to the public as office applications but still very important to the IT industry, they did the same. Microsoft consistently refused to join the ongoing industry effort around Web Services Distributed Management (WSDM) at OASIS for years. They kept claiming that they had no interest in this topic. Yet, in 2005. after WSDM became an OASIS standard supported by a large segment of the industry mind you, Microsoft introduced their own technology named WS-Management, with support from some well chosen partners. Three years later the industry is still trying to figure out what to do with the mess they thus created.

But all this debate around multiple standards is somewhat of a distraction from the real issue at hand. In the end what is really being asked to National Bodies (NB) around the world isn’t to choose between ODF and OOXML, or to choose between ODF alone and ODF and OOXML. The question that NBs are asked to answer is whether OOXML deserves to become an ISO standard in its own right.

The reality is that if the OOXML specification wasn’t of such a poor quality it most certainly would have had a much easier ride through the Fast Track process. If all that could be argued against it was that it was too big, the IP license has gaps, and multiple standards aren’t good, this may not even have made headlines, no matter how true it is.

What is appalling about OOXML is that it is fundamentally a VERY BAD specification and I just can’t understand what process would allow this garbage to even be presented as an ISO standard up for vote. OOXML is from a technical point of view just terrible. All I’ve seen from it and all I’ve read about it only confirms this. And I have yet to meet any technical person arguing in all honesty that OOXML is a good specification. The latest facts reported by Rob Weir speak for themselves.

So, again, the real question isn’t so much whether the world would benefit from having several competing standards or not. The real question is how many bad standards do we need? And the answer is zero.

OOXML must be voted down simply because it is a bad standard.

March 19, 2008 Posted by | standards | , , , , , | 1 Comment

Conflict of interest

I wish I knew how to fix it but I don’t. What I do know however is that the ISO/IEC process is severally broken in that it is riddled with room for game play. I guess ISO/IEC managed to get away with it for a long time but things have changed now that Microsoft has shown how to use every loophole in the process to get to its end.

Hopefully, justice will prevail and OOXML will rightly get voted down at the end of the month but the process shouldn’t have allowed to go that far into this sad farce in the first place.

In the latest demonstration of how broken the process is the US V1 technical committee voted on Friday to recommend approval of OOXML in a 17-4 vote, where Microsoft and no fewer than 11 of its business partners voted in favor of the specification. Am I the only one to see a major conflict of interest at play in this?

Evidently the result of the vote simply reflects the make up of the committee and can in no way be trusted to represent any kind of objective evaluation of the proposed standard.

So, I have to ask: how can we ask someone to vote on its own specification? Has any submitter ever voted against its own submission? I actually don’t know if that ever happened but the conflict of interest is so obvious that I wonder why this was ever allowed in the first place.

My guess is that this same process usually works just fine because normally specifications brought up for ratification by ISO/IEC are the product of a standardization process which involves more than one interested party. In such a case, the standard doesn’t solely benefit one particular vendor but represents a compromise between stakeholders.

In the case of OOXML, the reason the process falls apart is I believe primarily because OOXML is the product of a single vendor and because this vendor is supported by a large ecosystem that can participate in what is nothing else than a hijack of the standards process.

To be fair, it is not clear that other standards organizations are immune to that type of scam but thankfully there aren’t that many Ecmas around ready to make it that much easier.

As I said at the beginning I’m not sure how to fix this problem. Given the possible intricacies of partnerships between companies of our modern world it is going to be hard to figure out a clear way to discriminate among the companies involved and decide whether they have a conflict of interest or not. This doesn’t mean we shouldn’t try though. For one thing, it is not hard to make the call when it comes to Microsoft and OOXML. So, we could definitely do better than what we currently do.

At a minimum when there is an obvious conflict of interest, such as when asking Microsoft whether OOXML should become an ISO standard or not, the rules should make it impossible for the company to take part in the evaluation process.

When I was a kid I remember there was a TV show called “Eurovision” in which musical bands from various European countries competed with each other. The bands were given scores by judges from each country and the band with the highest tally won, a bit like what is done for ice skating. Although my memory isn’t very clear on the rules and all I do seem to recall that judges did not take part in the scoring of their own country band. This was clearly meant to lessen any risk of bias due to conflict of interest.

Maybe some similar rule should be used for standards ratification. Of course, the fact that a company can hide behind a pseudo standards organization such as Ecma makes this a bit trickier to figure out but, again, I still think we should try.

Note that the recommendation from the V1 committee does not necessarily translate in the final US vote. This recommendation now goes to the INCITS EB, for a 6-day ballot and resolution meeting, which will decide on the final position.

I can’t wait to be done with OOXML so that we can focus our energy on fixing the international standards process. If anything, what we’ve learned throughout this whole ordeal is that it is in need of a serious scrubbing.

March 9, 2008 Posted by | standards | , , , | 14 Comments

Crowd lines up to complain about the Fast Track process!

It appears that, in spite of Microsoft’s attempts to spin the BRM into a successful meeting which delegations from around the world left “happy” and “cheering” (I seriously wonder what they are on, do greed and deception get one high?), people are really upset about the way the BRM went on and are lining up at the “Fast Track Customer Services” counter to file complaints.

Fast Track Customer Services

I actually took that photo while waiting for my plane in London Heathrow. Even though I wasn’t really paying attention the sign caught my eye and I couldn’t prevent smiling at the whole scene. I took the shot vainly trying not to look too conspicuous to the many puzzled people surrounding me. I didn’t try to explain.

I guess the words “Fast Track” will never be the same for many of us.

March 5, 2008 Posted by | standards | , , | 1 Comment

What does the world have to gain from rushing OOXML through ISO?


I’ve already stated several times before how inappropriate the Fast Track process is, simply based on the size of the specification, the number of reported errors, the lack of time, etc. But as I was attending the OFE conference last week in Geneva I was told something that really made me ponder about the reason this is even taking place.

What I was told is the following: during a break XXX from Microsoft was overheard asking a delegation, that had expressed its dismay over the way the BRM was being conducted, something along the following lines: “you want to throw away all this work? waste it all?” And his tone and attitude was reported to be fairly aggressive.

One can wonder about the appropriateness (and effectiveness actually) of trying to press on delegates by making them feel responsible for wasting everybody’s time but the thought that went through my mind when I heard this story was: “Why? Why does it to have to be a waste? Why does it have to be all or nothing? And why does it have to be decided NOW?? What’s the hurry?

The level of interest, discussion, argumentation, involvement, and passion put into the various debates taking place around the world have clearly demonstrated that OOXML is important to people – whether they love it, hate it, or are somewhere in between -.

Given that, it seems obvious to me that it should be given the time it needs to get proper treatment and that giving OOXML more time should be in EVERYBODY’s interest.

Indeed, more time would allow the malcontents to articulate their point of view better, list all the issues they have (and not just the few they have a chance to raise during the artificially constrained timetable of the Fast Track process). More time would allow the OOXML supporters to address all issues raised, and make their position stronger with an improved specification. Finally, more time would allow those on the fence to more carefully weigh the pros and cons and take a decision in a more serene atmosphere.

More time would most certainly allow to get all parties closer to one another if not to completely agree.

So, why is it rushed? Well, let’s see.

The only ones really pushing for this to happen faster rather than carefully are Microsoft and Ecma.

In the case of Ecma it is easy to see why. Ecma is nothing more than a rubber stamping organization for hire with no soul, which is pushing for OOXML to go through as fast as possible and with as little change as possible simply because this is what it is paid to do. This is what their “value proposition” is all about: ‘timely publication of international standards […] “fast track” […] minimize risk of change‘.

The case of Microsoft isn’t as simple on face value. If, as it claims, Microsoft really wanted to be more open it would have a keen interest in improving the quality of the specification and ensuring maximum interoperability. The world at large would also gain from having a better specification that actually enables competition – be consumers or vendors -.

So why is Microsoft favoring time to “ISO standard” over quality of the specification? The only possible explanation is that its claims about openness are mere pretense. If it’s only pretense then everything becomes obvious.

Microsoft has an inherent conflict of interest in the OOXML standardization process. Not only it does not have any interest in improving the quality of the specification, in fact, the worse it is the better for Microsoft. The lower the quality of the specification the more difficult it is for others to implement OOXML and actually compete with Microsoft products.

All Microsoft really cares about is the ISO standard label, so that it can declare that it’s now safe for everybody to buy its product because (in theory) it is based on an ISO standard. The faster Microsoft can get it, the better.

Unlike every other vendor Microsoft doesn’t need a good specification to develop its product, it already has a product. This really is why the OOXML specification is of such a poor quality. It is not accidental.

Now, what I don’t understand is why should anybody else play Microsoft and Ecma’s game. National Bodies have nothing to gain from playing this game.

NBs shouldn’t feel intimidated by Microsoft’s implications that they would be responsible for wasting a lot of effort by saying No to OOXML. NBs shoud instead tell Microsoft to stop wasting everybody else’s time for its own self and sole interest and request that Microsoft demonstrates genuine interest in opening up its format by taking it to the normal ISO standards process so that due process can take place.

March 4, 2008 Posted by | standards | , , , | 2 Comments

Can anyone be more disingenuous?

There is not enough time to blog about everything that should be said about Microsoft’s indecent tactics to win the ISO vote on OOXML at all cost. I explained in my previous entry how misleading the disposition of comments document is, claiming agreement even when they disagree just so that the response can be seen as positive.

For another demonstration of pure disingenuousness let’s look at Microsoft/Ecma’s response to China’s comment.

China filed along with their No vote in September a comment explaining that “we found it is a very complex technology […] We think the fast-track procedure is not suitable for this DIS” and ending with “more time is necessary and essential to conduct a credible and responsible evaluation.

Seems reasonable enough and it’s fair to say that this sentiment is shared by many.

In response, Microsoft/Ecma’s wrote the following:

DIS 29500 is indeed a large specification. However, there is precedent for ISO/IEC specifications totalling housands of pages. Examples include:

  • ISO Standards for the Exchange of Product model data (STEP):
    • ISO 10303-210:2001 – 4,515 pages
    • ISO 10303-214:2003 – 3,529 pages
    • ISO 10303-212:2001 – 2,808 pages
    • ISO 10303-218:2004 – 1,837 pages
  • MPEG 2 Standard (ISO/IEC 13818) – 1,558 pages
  • MPEG 4 Standard (ISO/IEC 14496) – 4,415 pages
  • ODA – 1,244 pages
  • POSIX (ISO/IEC 9945) – 3,549 pages”

On that basis, many will think “Oh. So, that’s really not a rare occurrence. So, what’s all the fuss about OOXML being so big then?”

Well, here is the catch: What Microsoft is conveniently omitting to indicate in its response is that none of these standards have gone through the Fast Track process! They all went through the normal 5 stage development process over several years.

For instance, documentation on MPEG 4 (ISO/IEC 14496) indicates that “The total documentation package for ISO/IEC 14496 is extensive; 17 parts have been published from 1998 to 2004, with more to come.” This makes it very clear that while the specification is indeed several thousand pages long it was developed in several parts over 6 years!

But that’s not all. Microsoft/Ecma’s response then goes on into citing how PDF went through Fast Track and is “a specification of over 1300 pages, which was submitted to ISO in June 2007 and ratified as ISO 32000 in December 2007.” Putting aside the fact that 1300 pages hardly compares with the 6000+ pages of OOXML, once more they omit some fundamental pieces of information:

PDF did go through the Fast Track process but it has been a de facto standard which had been publicly available for many years before it was submitted to ISO. Only 205 comments were filed during the review, and all of them were resolved. In the end it was approved unanimously.

Unlike PDF, OOXML is a brand new format. It is NOT the format that everybody is using today. OOXML is neither a de facto standard nor a format that has been publicly available for years.

If OOXML had been developed the way MPEG 4 has, it wouldn’t be of such a low quality today and, its standardization through ISO wouldn’t be so controversial.

February 22, 2008 Posted by | standards | , , | 2 Comments

When “Agreed” means No.

Speaking of agreement, I should point out how misleading Microsoft/Ecma’s Disposition of Comments is.

The Pro-OOXML lobbyists a la Rick Jeliffe were prompt to demonstrate with fancy graphics how “The Editor (Rex Jaeschke on behalf of ECMA TC45) has accepted the lion’s share.” Graphics that are now being reused by Microsoft in other documents prepared to convinced NBs that everything is fine, Microsoft/Ecma accepted most comments.

Of course, Microsoft conveniently forgets to mention that even Rick Jeliffe himself admitted in a follow up comment that “ISO-ese ‘Agreed’ is sometimes more like the Japanese ‘Hai’ or English ‘OK’ rather than the English ‘Yes’.” and that “indeed [this graphic] is of limited usefulness“.

Ironically enough, Rick ends up explaining that it’s because ‘People want slogans and headlines and they prefer “MS agrees to nothing” or “MS agrees to everything” to “Editor agrees to most but not all issues raised and makes suggestions in response usually prompted by the suggested solutions of the national bodies.”‘, that he made this post, all the while producing one of the most bogus and misleading piece of information for Microsoft to reuse as a headline and slogan.

I have personally read my share of comments and responses from Ecma and I had myself made this puzzling observation that while many responses start with “Agreed.” they follow with an explanation for denial of the request. Rick claims that in this case “Agreed” means “OK” but my reading of this is that it often merely means “No”.

While Microsoft/Ecma’s response is impressive in size, its content is far less impressive, and this kind of tactic can only lead to conclude that it is designed to mislead people into thinking they have agreed to most comments.

I wonder how that plays with the notion of “standards professionalism”…

February 21, 2008 Posted by | standards | , , | 2 Comments

Can you live with it?

In this latest entry Jason Matusow started the final phase of Microsoft’s pro-OOXML lobbying by saying how much “I really like [ Patick Durusau’s ] comment in his first footnote on the first page:

Granted, I have a number of issues with the current OpenXML proposal but experts do disagree in good faith even within open standards development projects. If a proposal cannot progress until we all agree, then we risk proposals being held hostage to whim and caprice.

Jason then insists on “The fact that there may be issues that remain open and/or in contention does not necessitate the perpetuation of a “no” following the BRM.” and that “[he is] hopeful that the same spirit of standards professionalism exhibited in Patrick’s letter will be reflected across all NSBs.

The sole irony of being lectured on “standards professionalism” by a Microsoft employee could have prompted me to write about this just because, having spent the greater part of my career working on standards, I find the whole OOXML debacle truly appalling and a total disgrace to the standards community. But this is not why I’m writing about it.

The reason I’m posting this entry is because, while I agree that total agreement isn’t always achievable, it is a mistake to conclude that it is therefore unreasonable, or unprofessional, for a National Body (NB) to insist on disagreeing by maintaining a No vote, and that instead they should accept that not all of their requests can be fulfilled so they should let go.

Anybody who’s participated in the development of standards knows that it’s all about compromises. It is a give and take process. Nobody gets exactly what he wants, but nobody gets nothing. In some cases, all agree, in other cases, they don’t. The question then is the level of disagreement you have and how you deal with it.

I think it serves to look at the W3C in this regard, because W3C is an organization that strives to be open and to not favor any particular stakeholder. For this reason it has set as a goal for itself to make decisions by consensus as much as possible.

Consensus does not mean everybody agrees, but it sets the bar higher than having a majority. A practical way to judge whether consensus is reached is not only to ask whether everybody agrees but also whether anybody strongly disagrees. The actual question asked to judge the latter is: “Can you live with it?”

What’s interesting is that the result can be completely different. In a situation where there are two options A and B, where the majority favors A over B but somebody can’t live with A, while everybody can live with B, B is the chosen option.

Getting back to OOXML, the question is not simply whether there are remaining open issues and whether NBs agree or not. The question is the number of issues that remain open, the importance of these issues, and ultimately whether they can live with it or not.

Given that NBs were never given the time necessary to review the whole specification, that the little review that could be done has revealed a low quality level, plus the fact that the BRM is limited to discussing the issues that were raised from partial reviews, no matter what happens at the BRM, there is plenty left for one to say “I can’t live with it”. It would actually be unprofessional not to.

February 21, 2008 Posted by | standards | , , , | 2 Comments

How to make a fool of yourself – Ecma’s lesson #1

One of the comments India’s NB registered with its NO vote on OOXML in September was:

The name “Office Open XML” is often mistakenly called ‘Open Office XML” implying a connection to the OpenOffice project which does not exist. This naming confusion has been documented and has occurred numerous times, including by analysts and even in Microsoft press releases and blogs.

It then requested the name to be changed to something that would be less confusing.

Ecma’s response is to deny the request, explaining that there is no need to change the name because “we believe there is no confusion“.

Yet, in the very same document, in response to another comment from India’s NB mind you, Ecma wrote (emphasis is mine):

There is no specific definition of macro languages in the Open Office XML specification […]

You’ve got to love it. 🙂

February 8, 2008 Posted by | standards | , | 5 Comments

Microsoft wants National Bodies to wear blinders

It is unknown at this point how exactly the convenor of the upcoming OOXML Ballot Resolution Meeting (BRM) will run the meeting and how he will make the call as to whether there is agreement or not on each issue being discussed.

My understanding is that he can simply make a judgment call based on the apparent consensus or lack of it, or he can call for a vote. Either way this means that countries present at the BRM need to be prepared to express their position on all the proposed dispositions of comments.

Of course, with thousands of proposed dispositions of comments and only a few weeks before the BRM, this is an impossible task. Microsoft knows that just as well as anybody else but they are not going to let this kind of hard reality stop them from pursuing the sacred grail – the ISO Standard label -.

So Microsoft is instructing countries to only worry about their own comments, not all comments. If followed, this has several advantages for Microsoft: it makes the task of reviewing the response countries received less daunting and it increases the chance that a country will be satisfied by the response, and consequently decide to vote in favor of OOXML.

The only glitch in Microsoft’s take on this is that it really doesn’t make any sense for a country not to worry about all the issues that were raised.

I’ve participated in many standards meetings. Never have I seen issues being segregated by their origin. Once issues are raised they all end up in the same pool for everybody to consider, regardless of their origin. And then when a resolution is proposed everybody gets to express his or her opinion on it, not just the people who raised the issue. This is simply because everybody’s interest is in the standard as a whole, not just particular sections of it.

The fact that it’s an impossible task for countries to merely read about all the issues that were raised and their associated proposed dispositions only proves, once more, that the fast track process is totally inappropriate for OOXML. But it’s not like we didn’t already know that and even if common sense would dictate to stop this non sense right now I don’t expect it to. The OOXML train already got seriously dinged in September but it will keep running its crazy course until it crashes.

January 23, 2008 Posted by | standards | , , , | 29 Comments