An interesting offer: get paid to contribute to Wikipedia

by Rick Jelliffe

My first computer was a Mac Plus. Loved it. My second computer was an AT&T Unix PC running System V. Loved it long time. My third computer was a Sparc running Solaris or SunOS. Loved it. At work I run Linux, Open Office, Firefox, Eclipse, etc. No drama. For the last six years I have been running a little company making Java programs. Love Java. I do a little open source development, in particular with the Schematron program (quite like it!), but I have also contributed some code to the Flamingo/Substance project over at JavaDesktop, which provides novel looks and feels and more modern GUI components.

The only time I use Microsoft products is on my laptop at home (a present from my dear old Dad), but I need it to run the SynthEdit program for making virtual synthesizers. Oh, I occasionally also use a ten year old Microsoft C++ compiler, to make some DSP filter code: I have released about 80 filters open source this way. I'm not a Microsoft hater at all, its just that I've swum in a different stream. Readers of this blog will know that I have differing views on standards to some Microsoft people at least.

As a regular participant at ISO standards, on and off for more than a decade at my own expense, it has always frustrated me that the big companies would not come to the table and make use of ISO's facilities. So I am a big fan of the Mass. governments push that governments should use standard formats only. I know some of the ODF people, I had some nice emails with the ODF editor over Christmas for example, and Jon Bosak asked me to join the original ODF initiative at OASIS (I couldn't due to time, unfortunately.)

So I was a little surprised to receive email a couple of days ago from Microsoft saying they wanted to contract someone independent but friendly (me) for a couple of days to provide more balance on Wikipedia concerning ODF/OOXML. I am hardly the poster boy of Microsoft partisanship! Apparently they are frustrated at the amount of spin from some ODF stakeholders on Wikipedia and blogs.

I think I'll accept it: FUD enrages me and MS certainly are not hiring me to add any pro-MS FUD, just to correct any errors I see. If anyone sees any examples of incorrect statements on Wikipedia or other similar forums in the next few weeks, please let me know: whether anti-OOXML or anti-ODF. In fact, I already had added some material to Wikipedia several months ago, so it is not something new, so I'll spend a couple of days mythbusting and adding more information.


Ian Lynch
2007-01-22 10:31:12
Since you openly admit to being paid my Microsoft you immediately destroy any credibility as a neutral commentator. End of story.
Daniel Carrera
2007-01-22 10:36:50
First, my position: pro-ODF.

I'd like to reply to a few points:

>>>>The OOXML specification requires conforming implementations to accept and understand various legacy office applications<<<<<

I'm not sure if I've heard that one. I know that in the OOXML spec just about everything is optional, so the compliance issue isn't really the point.

A more relevant point is that real-world OOXML files will contain stuff from all over the spec, and even external file formats like RTF. There's no reasonable way that I, as a software developer, can handle that. This is a practical concern, not a legal issue.

>>>>>As I have mentioned before on this blog, I think OOXML has attributes that distinguish it: ODF has simply not been designed with the goal of being able to represent all the information possible in an MS Office document<<<<<

This is inaccurate to the extent that it requires qualification. ODF was not designed specifically for MS Office, but it _was_ designed to be able to represent any office document, including, but not limited to, old, current, and future functionality of MS Office. This, of course, requires an explanation and I will explain.

Before I explain, I want to note that I know the ODF spec well, and I'm familiar with the EOXML spec. The only thing I've noticed in EOXML that can't be done with ODF features is inserting RTF files.

Now my explanation:
ODF can be extended. ODF section 1.5 explicitly allows extensions to the spec, with only minor conditions (essentially, you have to use your own namespace).

Microsoft is free to use ODF features for the bulk document and create their own, separate namespace for functionality they couldn't map. This is permitted. This doesn't make the file invalid, and I know of one other ODF application that does this too.

Of course, it is highly preferable that MS standarize their extension through a standards body, but that is a separate argument now.


Jerry Blogger
2007-01-22 11:39:18
You wrote:

"work-in-progress external specification like ODF"

Er, um, ODF is an ISO standard -- ISO/IEC 26300 -- hardly a work in progress. It already has an extensive track record and is already fully implemented in some 30 packages. It is designed to be neutral, vendor independent and platform independent. I seriously doubt that all of OOXMLs 6000 pages will be fully implemented by anyone (including Microsoft). Some parts of the OOXML spec simply cannot be implemented on non-X86 platforms at all.

Pretty funny stuff really.

Fred Arnold
2007-01-22 11:46:30
It's a bit harsh, but I have to agree with Ian Lynch's comment. Look what Microsoft did in Massachusetts, and is still trying to do- derail ODF at all costs, to the extent of trying to re-organize the state government so that a "task force" would take control away from the Information and Technology Division.
Now what business does Microsoft have interfering in any government body to that degree?

When you take dirty money, you get soiled. Microsoft is not some poor oppressed underdog that needs you to ride to the rescue, but a ruthless convicted illegal monopolist that gets dirtier with age. It can take care of itself- no need to make yourself yet another Redmond casualty.

D.C. Parris
2007-01-22 11:50:00
Had Microsoft participated in the development of the ODF spec, as they were invited to do, their claims that ODF was not developed with MS Office formats in mind would be largely a non-issue. In fact, this point makes Microsoft's claims largely FUD. And I agree with the sentiments that being paid by Microsoft shoots your credibility out of the sky. Honestly, wouldn't it have been better for you to offer corrections voluntarily? Or at least be paid by an independent body.
Peter Yellman
2007-01-22 11:52:34
It seems both silly and redundant to have to point out that you cannot both be paid ("contracted" in your terms) by an interested party for an opinion on an issue, and be "independent" on that issue. I am aware that there are people who would make the argument that this is ethically possible, but note that such arguments are almost invariably made by people who are being paid to do so. Even in a courtroom setting, where the system goes to great lengths to maintain a version of this illusion, expert witnesses are presented as "for the defense" or "for the prosecution".

Peter Yellman

2007-01-22 12:04:37
How *DARE* you clean up anti-M$ FUD on wikipedia! I'M TELLING RMS ON YOU!
2007-01-22 12:28:50
Perhaps I'm missing something but I thought the purpose of a standard was for people to create interoperable implementations. Having something like "strings" spew out bits of ascii text from an ODF document would be "conformant" under your definition it would just not implement 99% of the "bits" and say so. This seems even more absurd.

So the question for me comes down to this: can anyone make a proper program by taking the 6000 pages of specification and implement ONLY what is said in those pages without lots of experiments in a software historical archive or worse?

Put simply OOXML "has simply not been designed with the goal of being able to represent all the information possible in an MS Office document" either. Or perhaps it "represents" them but doesn't define them in the OOXML spec in any way that would let anyone other than Microsoft correctly implement them.

If it said "Here you need to implement this PROPRIETARY feature which we will not document" for the implementation to work properly, I think you would likely reject it. Instead it uses a more subtle indirection, "You need to duplicate what this other program does" without specifying it further, and the program is not something easily or commonly available.

"But the archiving community deserves support just as much as the document distribution community". They deserve a complete and open standard, not one that even if implemented 100% correctly according to OOXML spec (exclusive of the external non-standards) won't properly be able to process files.

Or am I wrong and OOXML is really complete? I don't think so at least where the "contradictions" have been pointed out. But perhaps you can look deeper than I have.

2007-01-22 12:30:13
As far as I'm concerned, this issue can be broken down into a single crux; can a document in OOXML using a legitimate partial implementation be opened in a MS product, saved, and still be opened by the original, non-MS tool? Similarly, can a file created in an MS product be altered by a non-MS tool and still be valid?

I think you'll find that interoperability only works smoothly migrating from MS products to the same version of the same MS product. Don't take my word for it - run your own tests. The spec is all well and good, but as far as I can tell it only looks open - any OOXML document cannot leave the MS silo.

Fred Arnold
2007-01-22 12:35:27
"Or at least be paid by an independent body."

If you can find one that is truly independent, and not a Microsoft-funded shill. :)

2007-01-22 12:58:17
The example of the Chinese WAPI contradiction is certainly a precedent. But so is the case last year where the German and UK NB's raised a contradiction in JTC1 against Microsft's C++/CLI specification. (See for details)

I think the C++/CLI programming language is more akin, in terms of the technology and concerns involved, to document formats than a WiFi router standard would be, don't you agree? This precedent shows that a broader set of concerns can and have be used recently to argue for a contradiction.

As others have already said, Microsoft at any time was welcome to work with OASIS to add support for their concerns to ODF. IMHO, it shows fundamental disrespect to standards in general and the W3C and JTC1 in particular to refuse to work with the ODF, SVG, MathML, XForms or SMIL committees and instead create a new standard for everything. Remember, this is not just a new document standard. This is entirely new protocol stack of XML standards replacing the combined efforts of 100's of experts who have developed the existing standards in an open consensus, community process.

And this is not happening in a vacuum. At the same time Microsoft is pushing OOXML, they are also trying to edge out PDF with their XPS format, and PNG and JPEG with their HD Photo format. And don't get me started about XAML. This is an all-out assault on web standards.

As for the process that ODF used versus OOXML, this is really just a difference in the liaison relationships of OASIS and Ecma. Nothing sinister here. Ecma is a Class A Liaison with JTC1 so it is able to use the Fast Track process. OASIS has liaison status with SC34, so it may use the Publicly Available Specification (PAS) process there. Both processes are similar, but not identical, and are used to "translate" an existing standard into an ISO standard. Which one you may use depends on your liaison status.

Jean Hollis Weber
2007-01-22 13:07:58
This page is gathering a lot of good information.
M. David Peterson
2007-01-22 13:37:28

> I'm looking forward to the next few days.

As am I! In fact, I can't wait to see the transformation from its current state of FUD grafitti, to something more along the lines of, you know... FACTS! ;)

@Ian Lynch,

>> Since you openly admit to being paid my Microsoft you immediately destroy any credibility as a neutral commentator. End of story.

No, I'm sorry ... it's the other way around. It's called disclosure, and even more importantly it's called honesty, and is exactly the reason why trust CAN be placed upon the result of his work. Would it have made him more credible if he didn't disclose this information? You wouldn't have known, so form your perspective his credibility would be exactly where it would be based on the knowledge of who Rick is,

> An honest, hardworking, and experienced industry insider, who understands the ins and outs of technology, standards, and politics, and even more so how each of these come clashing together to form what we will each decide to perceive as "the truth."

Of course, you could state: "Well he could refuse to take the job out of principle for doing the right thing!" to which I would reply,

Job Description: We need a hardworking, honest person with extensive experience in standards bodies and document formats to write a non-biased, non-marketing, fact-based white paper on the Office Open XML Document formats. SPECIAL NOTE: Anybody on the planet will then be able to scrutinize every last detail, re-write, and then re-publish the paper as they see best fit.

The right thing? If you read a job description like the above, knowing full well that there is no such thing as "dry ink" on Wikipedia, and therefore if it turns out that anything you wrote was incorrect, anybody else could step in and fix it, then the right thing to do is *TAKE THE JOB*!

Rick was honest, open, and took a writing job in which he knew anyone could simply come in and erase what he wrote if it was even the slightest bit off target in regards to being factual. You don't think that the Open XML page is going to be monitored like a hawk by every Microsoft basher (and non-basher for that matter) on the planet for the next few days, to then *EXPOSE* every possible point they disagree with as "PROOF that he can not be trusted!"

Sorry Ian...


And *THATS* the end of your so called story...

Oh, one last question: If O'Reilly pays me to write a book, does that mean the contents can not be trusted?

Just wondering, cuz' if it stands, your argument would disqualify nearly every book, paper, and other forms of written publication on the planet if held up to your standards of "credibility".

2007-01-22 13:44:03
My viewpoint: pro-ODF, anti-FUD, pro-open standards.

Are you expecting to continue writing articles from an independent viewpoint? I would think that being a paid advocate for one of the protagonists would compromise your independence. I urge you to turn them down and to present your own thoughtfully-researched (and hopefully unbiased, but if not then state your biases up front) series of commentaries instead.

On the legacy features issue, you know that all it takes is one major implementation of OOXML to use those features and instantly any implementation that cannot handle them properly is seen as non-compliant and lacking functionality. Thus, you must either implement those features or become "almost compatible". Saying it isn't so does not change the facts, and you have already (by stating something that is demonstrably false) compromised your integrity.

Any standard is only open if there are multiple , independent implementations. If "IP" questions and secrecy prevent competitors from implementing the full spec, it is NOT really open.

M. David Peterson
2007-01-22 13:59:53

And this is not happening in a vacuum. At the same time Microsoft is pushing OOXML, they are also trying to edge out PDF with their XPS format, and PNG and JPEG with their HD Photo format. And don't get me started about XAML. This is an all-out assault on web standards.

A technology company creating new technologies?!

THAT'S JUST AWFUL! Think of the children!

Of course, to think of the children, would mean to be able to provide for those children, and to provide for those children you would need a job, and if you just so happen to be interested and good at technology, you could always go and get a good paying job for a company that develops technology.

Of course, that would require that technology companies existed in the first place, but sadly they went extinct back in 2014 -- Can't really pay a paycheck to someone unless you have product to sell, and to sell product, you would need to be able to create that product, and to create that product you would need someone with the proper skills, and to gain the proper skills requires education, and to gain an education, you need to go to college, and to go to college means you need to have a way to pay for your college, and since most people on this planet find help in that department from their parents, this requires that their parents actually have a job. Of course, even with a job, their parents would need to actually -- you know -- think about the children, and to think about the children requires supporting them and supporting them...

I'm sure you can probably see the patter here, right??? Please say yes.

TECHNOLOGY IS NOT EVIL! In fact, if you want my opinion, it's the people who believe that the creation of new technology is a bad thing that are evil ones.

"Yeah, but if we could all just agree on..."

Since when, in the history of the entire planet, has *ANY* generation been able to sit still and be happy with what they already have, and have been happy to simply just "work together as a community for the greater good of everyone."

They tried that once... It's called communism.

"I'd like a second helping of technology, please... Thanks!"

2007-01-22 14:09:24
What is the point of having an approved and accepted standard if any large corporation can come along and insist on their own file format being a "standard"?

Surely what Microsoft should be doing is to address the parts of the ODF standard that prevent it from being used as a portability standard for documents created using MS Office.

If ODF is not capable of enabling portability of documents created using MS Office, then it needs to be fixed.

But if it IS, then there is no point in having a duplicate standard.

IOW, the proper path that MS should be taking is to get extensions officially approved and added to the ODF standard - NOT seek an entirely new standard to duplicate most of what the current standard already does and does well.

2007-01-22 14:20:33
The need to keep to archivists happy is easily understood, and I have no objection to the notion of Microsoft creating an XML-based format that can express every single piece of legacy data in their back catalogue. That's their business. But what need is there for it to be an ISO standard?
2007-01-22 14:28:08
Wikipedia entry doesn't cover even half of the problems with OOXML, there's more complete list at
2007-01-22 14:36:19
I'm on your side. I hope you are able to take it on. The FUD-calling is pretty disgraceful, and it bothers me a lot that supposed professional executives for really big brightly-colored organizations (formerly known as the biggest monopoly on the planet until another one arrived) are making a death-match out of this.
2007-01-22 14:52:49
Oh about the maturity of ODF. Let's see, still doesn't produce an ODF document that relies exclusively on the ODF-blessed namespaces. There is still no formula specification for spreadsheets and other formula usage.

There are a variety of other places where ODF is underspecified. (E.g., you can include alternate content, such as some binary alternative to an image -- a feature that OO.o uses -- but there is no provision to say what format that blog is in. And these are the guys that blame MSFT for doing that as part of their legacy-preservation effort. They do say what's in the blob though.)

I suspect that there's a lot to tweak, maybe some to repair, in the ECMA Office Open XML Document Interchange specification.

But it must be closer to something that works than ODF, and we'll see what parts people choose to use the most in useful interchange of preservable documents.

2007-01-22 14:53:52
"but there is no provision to say what format that blog is in"

Sorry, I meant "blob" not "blog."

Frank Daley
2007-01-22 15:04:18
to: M. David Peterson

A technology company creating new technologies?!

THAT'S JUST AWFUL! Think of the children!

However, your basic premise is wrong. Microsoft does not create open standards. It's so called standards are simply a set of specifications controlled by Microsoft.

Upon detailed diagnosis, every one of Microsoft's so called standards are simply part of its ongoing strategy of embrace, extend, exterminate


Dylan Brams
2007-01-22 15:10:52
There is something fundamentally wrong with accepting payment to contribute to Wikipedia. Wikipedia is a consensus of what, under certain guidelines, a very large community thinks. This is not much different from accepting a paid lobbying contract; if anyone is paid to contribute to Wikipedia and not a member of the Foundation itself, it is by nature astroturfing. I hope you can agree that it should be discouraged.

Corporations should not have defenses from the people in the community. They are not equivalent to people, and should not be treated so within that community. The information source was created out of the desire of people who were not paid to share, and injecting thought which is influenced by any monetary bias is by definition sullying a good source of information.

In all honesty, I hope you are blocked from contributing. Accepting changes submitted by people who are - even openly - paid to be interested in the outcome of the changes is an extremely bad precedent.

Troy Roberts
2007-01-22 15:17:25
For one that claims understanding of the ISO standardization processes, you leave much to be desired with respect to the current topic. ODF was accept through the fast track procedure and OOXML is on the fast track procedure. The difference, of course, is that the ODF spec (around 700 pages) is about 1/10 the size of OOXML (about 8000 pages).
2007-01-22 15:22:25
Please read Rob Weir's blog if you haven't already; I have found it to be an excellent eye-opener about just how horribly evil OOXML is. If you can actually stomach reading Rob's blog an still go on with this job... well... actually either way I want to ask: How much money are they paying you and what are the terms? Did you ask if it was OK to blog about the specifics? /amazed
2007-01-22 15:22:41
MS announced years ago that bloggers would be selling vista.
2007-01-22 15:24:09
2007-01-22 15:29:26
On another note, I think that companies paying people to correct harmfully incorrect things about them in Wikipedia is a great idea and makes a lot of sense. I'm sure it happens all the time, but I don't know of any prior cases of people admitting to doing it.

If you're going to try to be honest and transparent about it though, you should of course say on your userpage what you are doing and who's paying you.

Sorry for the rapid-fire comments; I'll stop now.

CD Baric
2007-01-22 15:30:14
How amusing... Microsoft dosen't being the victim of FUD.

Har De Har Har, Har, Har.

What comes around, goes around.

He who lives by the FUD, dies by the FUD.

And you want to run out into the traffic and help out Microsoft, for a fee. Thirty pieces of silver, perhaps?

You do understand that Microsoft is using it's financial mass and influence to corrupt the Commonwealth of Massachusetts' stated objective of only employing truely open document formats but you want to help them with a PR problem.

Thanks for pitching in.

CD 'Bar' Baric

Joe Buck
2007-01-22 16:18:22
Feel free to change the wording as long as you don't hide the problem. The difficulty is that, as written, the spec allows a compliant document to contain any Microsoft legacy format, and the resulting document is still considered an OOXML file.

In my view, a standard document format should be just that, with sufficient information to allow programs to be written to fully parse that document format. Old formats should not be considered OOXML.

2007-01-22 16:27:29
Orcmid wrote:

I suspect that there's a lot to tweak, maybe some to repair, in the ECMA Office Open XML Document Interchange specification.

But it must be closer to something that works than ODF

To the first: if that's the case, then I think you'd agree that Ecma 376 is definitely not a good candidate for a fast track. As to the second point: What is your basis for this claim? There is exactly one implementation of Ecma 376, and it is hardly mature or widely used. There are several implementations of ODF, with a much longer track record.

2007-01-22 16:34:43
Quite apart from the MS propaganda/"fact-correcting" issue, the author seems to be pointing a finger at Wikipedia - regarding the inaccuracy of information on its pages.

It's an old argument but it tends to hold firm - if you read an article and think it is wrong then don't moan about it. Change it. If you can't put in any new information then at least remove the incorrect information and note why on the discussion page.

2007-01-22 16:54:12
re: M. David Peterson's "his credibility would be exactly where it would be based on the knowledge of who Rick is"

Well, prior to reading this I did not know who Rick was, so his credibility in my book has gone from unknown to incredible (double meaning intended).

re: Dylan Brams' "There is something fundamentally wrong with accepting payment to contribute to Wikipedia." and "it is by nature astroturfing"

That is just crazy talk. There is nothing wrong with accepting payment to contribute to Wikipedia. 'Astroturfing' doesn't really apply either, as long as he is documenting (on his wikipedia user page) who his employer his. It certainly would seem, to me anyway, dishonest if somebody were to do something like this and _not_ disclose that they were being paid... but DISCLOSURE is exactly what this blog post is and Rick should be commended for that.

It is a given that commercial entities have always and will always try to sabotage the wikipedia in underhanded ways, but paying someone to correct inaccuracies is not underhanded. Paying someone to correct inaccuracies does nothing to damage the consensus model. Wikipedia has means to deal with actual sabotage, funded or not, and they work pretty well.

The problem here is that, from his gross misconceptions about ODF and OOXML already demonstrated in this post, I find it very hard to believe that Rick is the right person to fairly do this kind of job. So, I hope that if/when he gets caught inserting demonstrably intentional falsehoods he _then_ gets blocked, with extreme prejudice. As in, please go away and never come back; you are banned for life. Getting blocked just for being paid to edit wikipedia would set a horrible precedent, though.

Obviously, given the unknown (but large) number of paid editors already there, Wikipedia clearly needs more paid editors who disclose that they are paid!

Peter Yellman
2007-01-22 19:31:27
Re: Finite: "That is just crazy talk. There is nothing wrong with accepting payment to contribute to Wikipedia. 'Astroturfing' doesn't really apply either, as long as he is documenting (on his wikipedia user page) who his employer his. It certainly would seem, to me anyway, dishonest if somebody were to do something like this and _not_ disclose that they were being paid... but DISCLOSURE is exactly what this blog post is and Rick should be commended for that."

I agree: Rick is to be commended for his disclosure. Keep in mind, however, that disclosure was clearly Rick's idea. No one in his right mind would believe that Microsoft generally favors disclosure! It is a dirty, barely hidden secret that "disclosures" such as Rick's and the recent "free laptop" incident represent the small tip of an iceberg of paid Microsoft marketing and FUD, channeled through bloggers, tech writers, etc., the VAST majority of whom represent themselves as objective analysts. You apparently agree with me that Rick is exceptional, since you are ready to give him a commendation for revealing that he is being paid by Microsoft for his activities. In any case, the issue of disclosure is irrelevant to my criticism, which is that Rick juxtaposes his disclosure with a reference to his own "independence". Rick can claim that he can be objective or even truthful on this subject while being paid by Microsoft and the rest of us can make our own call on that, but the definition of independent as he used the word cannot sustain a paid relationship such as the one he describes.

Peter Yellman

El Cerrajero
2007-01-22 19:55:05
Let's see, OOXML is a huge piece of crap. That's a fact.

If you write about it objectively MS ain't going to be very happy ^_^

2007-01-22 20:07:05
For the record, although pseudonymous (so you'll have to take my word for it) I feel I should note at this time that I am not employed by any players in this field :) I just happen to really hate fake openness, crap standards, Microsoft, _and_ shills*... so I've got a strong personal interest in this matter.

Peter, yes, I fully agree that it is impossible for him to be an independent voice on this issue while being paid by Microsoft.

Note that M David Peterson earlier had the audacity to raise the straw man argument of "If O'Reilly pays me to write a book, does that mean the contents can not be trusted?" (He obviously missed the key word *neutral* in the thread's first comment, either accidentally or intentionally).

* By the definition of shill I don't actually believe Rick is (or will be) technically a shill per se, so long as he discloses his relationship (on wikipedia in addition to here). But considering that Rick did not say that Microsoft encouraged him to disclose their relationship, and we'd be foolish to assume this is the first time Microsoft has paid people to edit blogs and wikis, this issue is still very much shill-related.

For a nice long list of links documenting Microsoft's history of employing shills, see the digg page about this blog post:

Also relevant:

Rick Jelliffe
2007-01-22 20:47:24
Wow, so many comments. Here's a few quick responses.

Ian: I have not started or been paid by MS yet. It is only three days work we are talking of, as an independent contractor not as an employee. My opinions are long-held and on public record. I don't have to get approval for any changes I make to MS or Ecma or anyone, so the improvements I make would be my own, not imperitives from the Borg. The job relates to Wikipedia and not to my blogging or other forums. And I am doing it openly, so that suspicious people can judge whether I have Stockholm syndrome. But at a certain point, grownups look at arguments rather than teams.

Daniel, I asked the editor of ODF a question to the effect of whether they ever looked at MS Office in designing ODF. His answer was to the effect that they hadn't and that it was up to MS to join ODF if MS wanted to influence it to be more MS Office-friendly. An entirely reasonable position for a standards working group. So until someone goes through both specs and finds otherwise, the claim that ODF necessarily can handle everything that is in OOXML is some combination of hype, optimism and intent.

Jerry, work-in-progress in the sense that parts are still being developed, such as the spreadsheet parts IIRC.

Fred, I am not an anti-trust policeman, nor a business practices policeman, nor am I motivated by compassion for Bill Gates. My experience in the XML/SGML document world in the last 18 years has been that the presence of MS Word in a production chain has been an enormous stumbling block for putting together working systems. And since it most deployed system in the kinds of places that put these production systems in place, it has been a major problem indeed. In the early 90s, and David Sklar of EBT pioneered a thing called the "Rainbow DTD" which provided a very close-to-the-metal conversion format for RTF, FrameMaker MIF and Word Perfect: in fact it seems to be a general architectural principle that for complex data conversions you are better off with a pipeline where the first step is to bring out an XML-ized view of the binary format or database, then convert that. For example, the Open Source OOXML<->ODF converter is based on this kind of pivot. So OOXML is a great step forward, even for making ODF import/export practical. There are a lot of people who have to use Microsoft products in their work; I don't care whether Microsoft benefits or not, I don't see that MS money is any more dirty than any other big company's (I have some horror stories from other big companies too!), but if someone is going to be paid to improve Wikipedia, I would prefer it was me rather than some marketing type.

D.C. Parris: err which body? If ODF or IBM turned around tomorrow and also offered me money with the same lack of strings, I would be happy to accept it. But imaginary bodies frequently don't pay up. :-)

Peter Yelleman: Yes, but everyone has an angle. I run a company that sells XSD and RELAX NG and Schematron software; and I write a blog saying Schematron is good and RELAX NG and XSD have their place but that grammar-based schema languages are double-handling and lack power and so will eventually die out. Does my blog reflect my company position? Independent does not mean "doesn't have a POV" nor does it mean contrarian (having a point of view deliberately different from everyone elses.) You may agree with my POV sometimes, MS may agree with my view sometimes, ODF may agree with my vie sometimes; independence comes from persuing your view no matter what other people think. So for me to say "I won't say X because corporation Y also thinks that way today" is actually to surrender my independence. What is important is transparency: when you read an item from MS or ODF or consortium org, you need to know "who pays the bills?" But the purpose of this is not to dismiss an argument because it comes from the nasty people in black hats; it is to help detect FUD or spin.

Anonymous: :-)

tz: Since OOXML is the native saving format for Office 2007, I don't know how you can say that it hasn't been designed to support all the information in an MS Office document? What magic captures the other information if not the file format?

William: that is an interesting claim. Would you care to share a scrap of evidence, even the tiniest, tiniest, tiniest bit, or do you go on the FUD pile? In fact, as the examples a few months ago on my blog show where I have both ODF and OOXML (and HTML and others), the structure is unusual but regular.

Fred: Microsoft shill? Is this name calling now? Since I don't have to get approval for what I write to Wikipedia (and I am not being contracted for any other work or lobbeying) why should I care who pays?

Rob: (I think this is Rob Weir from IBM, an MS competitor: he has an interesting blog) Thanks for the tip on that case, I'll certainly have a squiz. Of course, just because broader issues (err, is that a euphemism for FUD? :-) ) were raised, it does not mean that the issues were decided based on them. ISO committees attract people who like to cross the t's and dot the i's, and rightly so. Which makes them impatient about extraneous issues. But does that make the problem merely one of naming? If the Ecma standard was accepted at ISO under a less similar name and with a more differentiating pre-amble, would that then satisfy your "contradiction" claims?

Jean: thanks for the link. Of course, don't expect Groklaw to be impartial! The benefit of that site is that it is not impartial.

David: thanks for the friendly comments. I wonder whether Rob Weir, who is full time, gets demands to know what his salary is, too. Or whether Groklaw posts funding arrangements. In ODF there are many volunteers, but I don't think anyone would be working supporting OOXML in a voluntary capacity: it is a different beast. But that is probably the same as the 802.n specs: industrial standards are underwritten by the industrial bodies who will benefit from them.

W^L+: To say that a standard is not open unless it has multiple open source implementations is goal-post shifting, isn't it? None of the definitions of Open Standard on Wikipedia have this constraint: ITU, EU, Danish govt, Perens', Kerchmer's, and even Bob Sutor's. And whether the "Open" in "Ecma Office Open XML" means "Open" in the sense of "Open Standard" or a lesser sense

Dave: Yes, ISO standardization is not a court where the other actions of a company are judged. Especially by a jury stacked with its competitors :-) Rob complains about MS's non-standard formats, but where has IBM's support for ISO SC34 been for the last few years? IBM used to be very involved at SC34: Charles Goldfarb, Sharon Adler, Ander Berglund and so on. But they moved their focus of attention to W3C then OASIS etc. They suddenly rediscover ISO when there is some marketing advantage, and feign outrage when Microsoft do the same. But all companies are welcome at ISO JTC1 SC34, even prodigal sons! Maybe I am being too harsh.

Standards Fan: on other hand, no-one is forced to use a standard. The question becomes "which standard is more appropriate for my use". The users have a richer set of choices. I fully expect that ODF will be the format of choice for governments who need a level-playing field and public data interchange, as a matter of public policy: Microsoft simply are not pitching OOXML as a solution in this area AFAICS. Hence their underwriting of the ODF converter. Ultimately, they don't care what format is used as long as they are not locked out from markets and as long as they can try to lock users in by dependence on features; which is fine. (Lock-in by features is different from lock-in by formats. But the rise of the web application suggests that there are compelling reasons for some people to avoid lock-in by features too. However, this has nothing to do with standardization processes.)

Alan: Many government organizations have preferential policies, so that if there is an ISO or national standard, that must be preferred to a consotium standard or proprietary technology. So it reduces the amount of work for organizations looking to adopt OOXML, in that they don't have to justify it against ISO PDF, ISO HTML, ISO ODF, etc. as much.

AragOrn: I am glad to see lists with real issues, not FUD. This is something that I know some of the key players in the ISO ODF process were worried about too: that there was too much FUD and too few facts. I am not being paid by MS to go through that list on my blog, just to edit Wikipedia, so forgive me if I don't do a point by point on them! I'll just make a general observation again, that the point of OOXML is that it publicly exposes the format of a particular real piece of software. If you designed a data format from scratch, you would clearly do things differently; but that simply isn't the point of OOXML, so comments based on "X could be done better" are irrelevant it seems to me. It is ODF's burden to be ideal, it is OOML's burden to be real. I think Bob Sutor made a nice comment that "OOXML is the past, while ODF is the future", and that encapsulates a lot of differentiation between the two: people with legacy .DOC etc issues may be more well-served by OOXML than by ODF.

Orcmid: well, I think IBM is playing hard on this, but there is a lot of anti-Microsoft sentiment out there. And there are legitimate questions to be asked and answered. The thing for decision makers in national bodies to do, if they are responsible, is to not accept one side's claims unless they have heard and weighed a response. IBM or the others are doing anything wrong in putting their POV vigorously, and aligning their corporate imperatives behind opens source and open standards is a great lark and good for users I am sure, but this is an issue with billions of dollars of sales at stake, and so decision makers shouldn't accept everything that anyone writes uncritically, especially when some ODF proponents think it is OK to make "broad claims" regardless of how wild they are.

Frank: Of course MS wants ODF-supporting companies to remain just small enough to forestall anti-trust efforts. What public company would want to give up market dominance? But so what? The points are whether MS has satisfied ISO procedure, whether the standard is contradictory and so stalls at the fist round, whether it seems to be of enough usefulness and quality to be voted at the second round. ISO is a standards body with its own rules about what makes a standard acceptable, whether this accords with "open standard" or not. Clearly I think that it is a disaster to have encumbered and RAND standards; free access is important.

Dylan: Do you think the Wikipedia entries have not been edited by people in full-time employment by large companies, in actual marketing positions, who have to justify what to write? When Wikipedia references Groklaw quoting a guy who works for OASIS' lawyers, is that therefore not a corporate influence? But, ultimately, if you see some change made in Wikipedia that is wrong, please fix it up! In the council of many is wisdom. Unless they are parrots.

Troy: Do you know the difference between PAS and Fast Track? See
for more explanation, but not from an unbiased source. ODF was accepted through the PAS procedure not Fast Track. (I was at SC34 meetings where this was discussed.) Yes, my knowledge in lots of areas leaves a lot to be desired; I hope I don't compound my ignorance with rudeness though.

Finite: Child slavery is horribly evil. A file format is not. But it is a good idea to put note up on my user page (I didn't know I had one.) I am not "admitting", I'm out and proud baby! More seriously, yes, I think transparency is good; indeed it is a form of openness. But I haven't actually made any changes yet, responding to this blog is taking all the time when I could be getting paid.

Ed: I am not selling vista. What on earth is your point?

CD Baric: Yes, it has its amusing aspects certainly :-) But from the ISO POV, the issue is not who has been a naughty boy but whether member nations would be better off with OOXML with an ISO number on it or not.

Joe: Interesting point. I think it is the same issue as embedding Java programs or .COM or VB controls or any non-standard/non-markup fragment, though. Both OOXML and ODF allow that: organizations adopting both of them would have to have some policy and profile in order to restrict it. The issue is moving away from markup and standard formats in any area, embedded controls as well as display material.

Nige: Oh, I'm not pointing the finger at Wikipedia. Its characteristics are well known. I use it all the time. It is great. Look me up in it! It is interesting that MS thinks that it is a kind of opinion leader that they feel is important enough to contract an outsider to look at. If I don't find it is a hotbed of FUD, I'll just improve the entries with more examples or references.

Finite: please let me know what these "gross inaccuracies are"? And "get caught"? Just correct the entry if it is wrong. As far as my competency, I don't want to blow own my horn, heaven forfend, certainly not. But Tim Bray (Director of Web Technologies at Sun) has said some nice things at

Brad Eleven
2007-01-22 21:24:16

I deeply respect your disclosure. There is nothing quite so attractive as being vulnerable, nor so telling as the response of others to vulnerability. I don't intend to make you wrong, let alone attack you. I want to show you a different side of the dodecahedron that represents the trade-offs inherent in being paid to provide one's opinion.

You had me at "AT&T UNIX PC". That is, I've read your article with the generosity that comes with feeling related to another person. I get your position, and it looks to be built on shifting sand. You seem ready to enter into this agreement with the best of intentions, and I believe you.

Consider that someone, right this minute, is having sex with a stranger for money, and that this person did not plan to be doing so. He or she started down some path with the best of intentions. Then there was a tiny compromise. And another. By now this person has been paid by the stranger, and is off to trade the cash for something that he or she is absolutely convinced is justifiable, if not necessary.

I'm not saying that this agreement will turn you into a crack whore. I'm not even saying that you're prostituting yourself in any way at all at this point.

I would ask you to consider that your assertion that MS wants to pay you "just to correct any errors [you] see" is, by definition, open-ended. Today your paradigms limit your perception of errors, probably well inside anyone's definition of objectivity.

How might this agreement affect your paradigms--even if the agreement doesn't change? It is my opinion that the agreement will, in fact, change. As several have pointed out, MS has many brilliant and non-evil people. These, however, are not the people at MS who thought up this idea, and they're most certainly not the ones with whom you've made the agreement. I know this because MS is not going to give a brilliant, non-evil person control of the disbursement of money spent in its self-interest.

I hate that corporate influence has infiltrated every aspect of our lives. By that I mean that I despise it. I loathe it. I blame it for the decline in our industry. I fault it for the commodification of every last technical position available in these United States and beyond. From my Congressional representatives, who have come to depend on corporate money in order to raise the vast sums required for re-election, to my neighbor who quotes talking points like, "Is it somehow illegal or immoral to make a profit?", I find it difficult to believe that anyone believes that a corporation has the same rights as a citizen. In particular, I think it's a steaming crock of sewage for corporations to claim the right to petition their representatives. I might rethink this if corporations had the same responsibilities--and growing liabilities--as citizens do. I find it regrettable that we, as citizens, have stood by and allowed for this corruption to advance. We get what we tolerate.

Of course it's not illegal, or even immoral to make a profit. It is when making a profit becomes a "win at all costs" proposition.

Consider the recent revelations about the patently illegal behavior of Hewlett-Packard's Board of Directors. They're just the ones who got caught, you see. MS is no different; it is not in any way open to full disclosure of anything, unless it is forced to do so. Even then--like a spouse caught cheating--a corporation will only disclose what it must.

I don't know you at all, Rick, except for this wonderful and open article that you've written. I can tell that you are talented, insightful, and that you care about the technical topics in this article, and probably about many more.

You seem to have been lulled into a sense of complacency about the role of money in what many of us consider to be a merit-based body, i.e., technical standards. The very presence of corporate patronage indicates that a tumor has taken root. Tumors are very interesting: They trick the body into nourishing them, with the explicit goal of taking over as much of the body as possible, simply to expand. Unchecked, they weaken the body and prevent it from its original purpose. Ultimately, they kill the body.

"If O'Reilly pays me to write a book, does that mean the contents can not be trusted?"

Yes, if the topic of the book is one in which O'Reilly has a financial interest.


No, it's not. Neither are guns. Either can and has been wielded by human beings with evil intent, often in the guise of righteousness.

Stephen Samuel
2007-01-22 23:14:06
You said: "yet, if I understand matters correctly, ODF was submitted in a fast-track procedure that didn't even allow these kind of objections."

Simply put the OASIS process is a very robust process which is reasonably close to the demands that ISO has for creating a proper standard. The ECMA process, on the other hand can be used to create a 'standard' which is little more than a formalized dump of a vendor's proprietary product.

These differing levels of robustness require a different level of examination when

It's like you're a hospital in Chicago and you've received CVs for two nurses -- One trained in Oregon, The other in the (then) Belgian Congo. Now, the Congolese nurse might be an excellent and capable nurse, but the chances are that, as a Chicagoan administrator, you're likely to more trusting of the Washington training -- simply because it's probably closer to the standards in Illinois.

That you would attack the difference in the treatment of these two standards raises some serious questions about just how much time (or was it money?) Microsoft spent indoctrinating you into their way of thinking before they set you loose to change the notes on Wikipedia.

As for UUIDs and death squads, I would note that Microsoft is being accused of enabling death squads not hiring them. It's people like Putin and Bush who are more likely to be accused of actually using death squads (depending on the accuser).

In any case, the fact that Microsoft paid you to 'correct errors' in a supposedly independent manner isn't a source of crisis of trust for me. On the other hand, you talking like a Microsoft evangelist after taking their cheque does raise some alarm bells for me.

2007-01-22 23:40:15
Rick, thank you for your response. I really hope that once you have had some time to think this over, you decide to turn it down. You sound like an interesting, intelligent, self-directed sort of guy. I really would like to hear your own opinions of the formats, and how each can be improved. I do not want the things you say to be subjected to "he was paid to say that," even if you finally decide to say things I disagree with.

I did not say "open source" implementations, because I did not mean that implementors must be open source projects. If an open source project can implement a standard, especially if they become the reference implementation, so much the better. Without Tomcat, for example, would everyone in the world offer Java servlet containers and application servers? Most of the ones I have seen use Tomcat (or Jetty, another open source implementation) internally. Still, in the context of the office applications market, enterprises want something with a paid license and support (which very well could be open source, or could be proprietary).

People like me want multiple choices within each category (IE/Firefox/Opera, for example), and we want the ability for those choices to work with the same data without a lot of extra work. The reason I can use multiple browsers is because they all work with a more-or-less open standard. That is the same thing I want with office applications.

Someone compared this to the free laptops for bloggers thing, and I think the comparison is apt. Joel Spolsky referred to "pissing in the well," and I think he has a point.

Rick Jelliffe
2007-01-22 23:49:19
Brad: Thanks for finding me attractive. I am even more attractive in person, actually. I have been involved in standards work at ISO, W3C, Standards Australia, IETF and a regional body, and there has always been a corporate influence. Standards committees are not always a paradise of humanitarians and altruism, and certainly not of commercial dis-interest. SGML came out of IBM. XML came out of Sun's Jon Bosak's organizing ability. And so on. The idea that there has not been corporate involvement in standards is not real. Some standards committees, such as W3C Internationalization, have a large corporate influence but are highly benign and fun. Other groups are just a hard slog, with the big boys trying to sabotage each other at every step.
2007-01-22 23:55:16
It's unfortunate for Rick and Microsoft that he announced before doing the job because the more he will do the harder the Wikipedia community will fight back.

I do appreciate that you finish your note by saying "its information should be taken with a grain of salt" because you seem to acknowledge that even though wikipedia content is great it should not be taken as the holy grail of content. I'm saying that because I've found disturbing reading some highly notable people praising for Wikipedia as if it was al good.

There are too many examples of the Wikipedia "ministers" making History as they see fit rather than as what actually is. Too many articles about existing people that contain wrong information that actually become information. 1984 is not far ;)

Anyway it would have been interesting to see what would have happened if Rick had done the job, then announced it. I'm sure we would have seen those articles being reverted really quickly.

Wikipedia is nice on the outside but is too political on the inside for my liking.

2007-01-23 00:02:25
OOps wrong copy paste.

That response was meant to Dare's blog note here about this article:

Sorry folks.

- Sylvain

Rick Jelliffe
2007-01-23 00:17:28
Stephen: Actually, I haven't read any material from MS on ODF versus OOXML yet. My thoughts so far are unfortunately all my own.

W^L+: MS is underwriting that ODF converter, so the OOXML standard does not prevent anyone from using ODF or, more to the point, any government from mandating it. I have a long-standing bias toward plurality (indeed, I spoke at WWW7 in I guess 1998 or so on the subject, opposing a MS guy...) so if MS has shifted around to my position, for whatever motives, I'm not going to suddenly change out of childish reaction. For example, here is a post of mine from 1999 on the subject:

The relevant lines are "There are some people for whom standardization means homogenization: in each area, we should only have one technology. There are others for whom standardization means that there should be at least one high-quality choice. I subscribe to the latter, and to the view that "enabling" standards are more important than "constraining" standards.

Sylvain: Actually, I have had email from at least one Wikipedia person who seems to think it is a good development. I may be hopeless naive here, but I don't expect to write anything that would be likely to be reversed. And Dare's blog is always fun.

2007-01-23 01:17:11
Thanks for responding, but I don't feel like my question is quite answered yet :)

If there is a clear requirement in a given organisation to maintain complete fidelity of every legacy nuance (however invisible and irrelevant to an information consumer) of an imported binary document, and only one modern format is capable of expressing such nuances, then surely that format will sell itself to the organisations that require it, regardless of the "preferred" status that any inadequate formats may enjoy.

On the other hand, if it is the content of a document that is relevant rather than the excruiating detail of its interactions with its originating software, then a one-time conversion into an application-neutral format which leverages widely-accepted standards would seem to be a sensible way to proceed. Otherwise it appears that the tail is wagging the dog, does it not?

Preservation of the minutiae of years of legacy (as represented by the "billions of documents" we keep hearing about) really seems to be most appropriately a matter between Microsoft and their customers. It seems sufficient -- and under different circumstances, perhaps even admirable -- that they have publically released the specification. Why now drag the entire world into it via the ISO process?

Incidentally, I can see your angle on plurality, but in terms of EOOXML, it seems a little obtuse.

2007-01-23 03:00:26
>> "The OOXML specification requires conforming implementations
>> to accept and understand various legacy office applications."

> But the conformance section to the ISO standard... specifies
> conformance in terms of being able to accept the grammar, use
> the standard semantics for the bits you implement, and document
> where you do something different.

Congratulations - by nit-pickery and careful definitions you've managed to avoid or ignore the problem.

The problem is that nobody can support a "standard" in any meaningful way if they don't support the whole thing. I can write a web browser that only recognises the "p" and "i" tags, but that doesn't mean it conforms to the XHTML spec in any *meaningful* way, even if the spec says this is acceptable.

The acid test is whether it *works* or not for normal users:

If you ignore all the legacy cruft in the OOXML spec, then the vast majority of the existing MS Office documents out there will simply fail to render properly when opened by application X. Users see their files' formatting screwed up, decide application X is useless, and Microsoft retains their Office monopoly.

If you want application X to reliably render any/all existing documents, you have to emulate the entire set of MS Office behaviours, including the legacy gunk, old bugs and details which aren't even explicitly detailed in the spec (IIRC, you have to ask Microsoft or third-party vendors for that information).

So... yes, maybe according to the spec you can write a "compliant" app using only the information in the spec and ignoring every legacy wrinkle and bit of still-proprietary information not included in it... but this app will be *useless*, by any sane definition of the term.

Sure, it must have seemed like a great wheeze at Microsoft HQ - "So, we release this byzantine format filled with binary blobs and legacy bug-replication wrinkles, but we don't document them, and then make all the nasty complex legacy bits optional. We use the fact it's optional to make it as hard as possible for anyone who's trying to write an Office app to understand and parse them, but without that functionality their app frequently breaks on existing files, making the whole thing useless and hopefully securing another nice, juicy decade of office productivity software monopoly for us."

One of the most attractive things about an open format or protocol is that it helps ensure against vendor lockin. Exactly how is OOXML doing this when it serves mainly as a way to unofficially-but-effectively degrade the ability of third-party vendors to compete with Microsoft is beyond me.

Unless, of course, one were being paid by one side of the debate to ensure everything's "fair".

2007-01-23 03:22:31
Why didn't Microsoft create the OOXML file format so that it wasn't dependent on previous file formats?

Opening a file saved in a different format should use a filter to convert the contents of the file to the new file format.

The conversion filter should be handed the job of maintaining document fidelity. The new standard should define document attributes (generic attributes like "line spacing" not "Word 97 line spacing") that can be set during the conversion process.

Once a document has been converted to the new format, there is no need to retain legacy baggage.

I believe doing this would allow the file format to be converted completely and cleanly.

Why didn't Microsoft create the OOXML file format so that it wasn't dependent on previous file formats?

To maintain its monopoly.

2007-01-23 04:06:16
I must say that especially IBM seems very intent on trying to create a smearcampaign against OOXML. (with of course Groklaw following IBM as always). I wish they would try and put their efforts in improving the ODF spec instead as at the moment I find it hardly workable to use the ODF spec for use in applications as it seems impossible to create compatible documents from just using the specs. (something in which OOXML also isn't great).
The fact that odf fans just jump on any comment coming from IBM and try to add it in wikipedia is just pathetic. Both specs are at best moderatly workable and it seems wiser to let OASIS and Ecma first improve to the next versions in stead of trying to discuss the merits and flaws of both formats.
People are actiing like standards are something solid and set in stone but actually development on both formats will continue for many more year so I think people should move forward more and if they think OOXML is flawed then come up with improvements and simular for ODF.
2007-01-23 04:15:55
Must admit I find Microsoft's move a little unnerving, but that's probably not too rational. I've no doubts about your integrity and I don't think that's undermined in this case by your being paid by an interested party, especially since you have publicly declared this. If anything, if ODF is better than OOXML (for any definition) then clearing up misinformation will be doing ODF a favour, because arguments can be based on solid premises. I bet there's also something of a (spontaneous, not-paid-for) counter-initiative, MS's move will add motivation to people wanted to clear FUD on ODF.

Could you please clarify one point:

"ODF has simply not been designed with the goal of being able to represent all the information possible in an MS Office document; this makes it poorer for archiving"

Is it then the case that OOXML can faithfully (reversibly) represent all the information possible in an ODF document?


For what it's worth I personally think it would be insane, all things being equal, for anyone to choose a format that's been designed with proprietary interests in mind over one which is grounded in the open source community.

I use OpenOffice if anything fatter than HTML is needed, not least because it saves me a lot of money over MS Word. I've done stuff recently for a publisher that insisted on Word, but it turned out OO's saving in Word format was perfectly adequate.

Nicolas Mailhot
2007-01-23 04:26:40
You're assuming OOXML and ODF can coexist with very different data models because they'll each be used within their own vertical isolated silo.

Unfortunately people (and apps) are increasingly exchanging documents, often Office documents (one can argue this is not the right way to do it but Microsoft has been pushing Office as an information hub for years), so all those differences are going to be very expensive to work around.

One can not think in terms of pre-1995 isolated PC anymore.

Likewise the long OOXML specification is not going to help the archiving community. Indeed the archiving community needs a simple specification more, because it'll have to manage/index large pools of archived documents to be of any use to its users. We're not in the microfilm era anymore. Documents live even after they've been archived. Reproducing Word Perfect artefacts accurately is less important than being able to reprocess any old document safely nowadays.