Standards, open standards and double standards
October 3, 2008 
By Jeff Gould, CEO & Director of Research, Peerstone Research
Some commenters think I was unfair to IBM in my last post. There I took Big Blue to task for its announcement that it intends to wage war against Microsoft in the world’s standards bodies. The motivation for this bellicose declaration was IBM’s stinging defeat last Spring in its battle to prevent the International Organization for Standardization (ISO) from ratifying Microsoft’s de facto office document standard (OOXML). IBM and its supporters are furious because the ISO had already ratified their own preferred standard, ODF (the XML file format associated with OpenOffice). The fact that they now have to share the coveted ISO label with Microsoft sticks in their craw.
Let me be clear. I’m not a fan of Microsoft Office, and I really don’t care much one way or the other about OOXML. But I am not impressed with OpenOffice as an alternative to Office. And I am really not impressed with IBM’s attempt to ram the little-used ODF down our throats. The proponents of OpenOffice have conspicuously failed to make this unimaginative clone of Office a success in the marketplace. So they took it to the ISO and got its format declared a standard, a fact which they expected to use to browbeat certain – um, shall we say, “easily influenced” – customers into picking OpenOffice. Everybody knows that the biggest users of OpenOffice are European governments (see the deployment numbers on the OpenOffice wiki), and most of us know that these governments are rarely unhappy to see successful American companies stub their toes.
IBM charges that Microsoft won at the ISO only because it packed the national standards organizations that make up the ISO membership with its pals. The suits in Armonk are shocked (shocked!) to discover that Microsoft actually tried to influence the outcome of a debate where its vital interests were at stake, namely its ability to sell Office to the world’s governments. I don’t doubt that there is considerable truth to IBM’s claim that Microsoft played hardball. But the examples it holds up (a rogue Microsoft employee in Sweden asking partners to vote for OOXML, “committee stuffing” in Switzerland, a mailing campaign in favor of OOXML in India) amount to pretty thin gruel. I mean, people, get serious. We aren’t talking about Franklin Delano Roosevelt’s attempt to pack the U.S. Supreme Court here. Strenuously lobbying for one’s interests is a normal part of the rough and tumble business of setting international standards. I’m pretty sure that IBM has lots of friends in the world’s standards bodies too, probably a lot more than Microsoft. Hmm, wouldn’t it be interesting for someone to throw a little light on that?
But the thing that galls me about IBM’s position – and the reason I wrote my post – is not its goody-two-shoes stance about lobbying. No, it’s the flagrant hypocrisy behind this whole open standards campaign. In a nutshell, Big Blue conspicuously fails to practice what it preaches. On the one hand, its open standards guy Bob Sutor piously affirms that he wants nothing more than to promote standards which will “ensure that each of us can easily purchase and interchangeably use computing technology from multiple vendors”. This is what they say when they are talking about a market where they are losing, such as office software. On the other hand, when it comes to a market where they are winning, such as the mainframe, it’s a different story. Here IBM uses threats, lawsuits and buyouts to shut down tiny competitors who want nothing more than to pay for the right to run Big Blue’s dominant z/OS mainframe operating system on non-IBM hardware. What part of “easily purchase and interchangeably use technology from multiple vendors” do IBM’s mainframe lawyer hounds not understand?
Once upon a time, say around two years ago, IBM had the following noble declaration on its web site:
"IBM has an open approach to patent licensing for products in the Information Technology (IT) field and is generally willing to grant nonexclusive licenses under reasonable and nondiscriminatory terms and conditions to those who in turn respect IBM's intellectual property (IP) rights."
Well, that page is gone now (though the WayBack Machine has preserved its memory). In its place, in the far less public venue of its court filings against its former competitor Platform Solutions Inc. (now a wholly owned subsidiary of – heh, heh – a certain large computer company based in Armonk, NY), Big Blue sings a very different tune about openness:
"IBM has refused to agree to license its patents to PSI and its copyrighted operating systems and other software for use on PSI's emulator systems despite PSI's demands because, among other reasons, IBM has no obligation to do so..."
As a free marketer my response to this is “fair enough.” Everyone has the right to play hardball with their own proprietary technology, as long as they don’t care about bothersome little details like their reputation for openness. I fully understand and accept that business is based on greed. What I don’t understand is how IBM can talk from both sides of its mouth and expect no one to notice. I mean, either you’re for openness or you’re against it.
For the record, I think that both z/OS and Windows should be open sourced by their respective owners. I’m not saying they should do it under the GPL. That would no doubt be asking too much. It’s only natural that these companies should want to preserve and expand the huge business empires they have built around what are two of the most useful and valuable pieces of intellectual property of all time. But the example of Sun and Solaris shows that they may have less to fear than they think. There is life for big ticket proprietary software after going open source.
Of course I’m not holding my breath while IBM and Microsoft make up their minds about open sourcing their crown jewels. But I do think we can demand a certain minimum amount of fair dealing from both sides. I’ve written elsewhere about my beefs against the new interface in Microsoft Office 2007 (in short, I hate it). But I can’t see what’s bad about the ISO’s ratification of the Office 2007 file format. After all, the standardization process has obliged Microsoft to dump its old policy of keeping file formats secret. It has exposed OOXML to intense public scrutiny and, in the opinion of at least some open source advocates, resulted in real improvements to a standard made extremely complex by the requirement to accommodate billions of legacy Office documents. While it would probably be prohibitively expensive (read: unprofitable) for a competing vendor to build a full-scale office suite around OOXML, Microsoft has at least put the format out in the open for all to see, and offered a grudging but seemingly far-reaching promise not to sue people who try to implement it.
Of course the IBM camp complains that this promise is ambiguous because, well, some developers may not be intelligent enough to understand the inevitably convoluted legal language of Microsoft’s promise. Since I’m not a lawyer, I wouldn’t claim to understand it myself. But I do know this. If IBM wants to show that it really cares about openness, it should extend the same “promise not to sue” to customers who want to run bought-and-paid-for copies of z/OS on alternative mainframe platforms offered by T3 Technologies, Fundamental Software and the open source Hercules emulator, those micro competitors who surely do not represent a threat – mortal or otherwise – to IBM’s vast mainframe money-spinner. If it does that, I promise I’ll write an open letter to Steve Ballmer urging him to port Office and SQL Server to Linux.
IBM Open Standards | 2 References
Reader Comments (19)
Just curious, If you're not a fan of MS Office and don't like OpenOffice either, what's your favorite? Abiword? Koffice? WordPerfect? Something else?
Microsofts de facto office document standard (OOXML)?
Pray, tell me which Microsoft product supports OOXML?
Can it be implemented in my own software? Last time I looked there was no specification online besides an old draft which has not much to do with what got the ISO badge besides lots of defects. And even the ISO specs aren't published, yet. Contrary to the ISO rules by the way.
From somebody earning a living with digital archives, OOXML is a dead duck.
I think the main reason is the fact ISO disregarded legitimate complaints about OOXMLs shoddy spec, I'm sure you've heard of them, references to undocumented functions in older MS closed formats etc. Sweden has accepted it separately itself due to dis-satisfaction with it.
http://www.metamorphosis.org.mk/content/view/1247/4/lang,en
It's not as if you have to give up MS Office anyway, MS had the chance and always do to fully join the ODF Board and shape it differently if needed but they chose not to. The idea of ODF is in 100years time with a printout of the spec you can still open a document, with the binary blobs in OOXML and reliance on undocumented MS features this is not possible.
Even today Norway has highlighted how they were cheated by their own standards group.
http://www.noooxml.org/forum/t-93970/norwegians-leave-their-standards-body-in-protest
It's a dirty tricks campaign which does nothing but hurt us as users int he future. ISO is meant to stand as an impartial judge in this matter but clearly it was not capable. Never has a standard of 6000+ pages been fast tracked like OOXML was.
It's poor show from all sides.
"Little used ODF rammed down our throats."
Oh my god, and I thought that Microsoft spinmeisters where the chamions of inside out logic.
Let us see.
ODF: vendor neutral standard. Supported by a range of products from a variety of vendors.
MSOXML: Microsoft, holding a monopoly in the desktop and office suite market rejects ODF and pushes forward its own format, a format that can only be implemented 100% by Microsoft.
Who is pushing what down our throats exactly?
"Little used ODF rammed down our throats."
This is either disingenuous or naive unbecoming a serious journalist. ODF started it's Oasis standardization on: December 16, 2002 and it wasn't complete until it went through 3 years of refinement. It also reused already existing standards as much as possible. At that time, it went to ISO and was passed unanimously without controversy or contradictions. There were some deficiencies relating to Accessibility and Formulas, but since there were no existing standards in this area yet anyway, this issue was put off until future versions of ODF (they're now part of the standard)
OOXML, OTOH was rammed through ISO using political games (e.g. kicking out most of those who opposed then voting or denying legitimate appeals from even being heard or bulk voting and outright lying to all involved on the meaning of each vote or even violating ISO's own rules to get OOXML passed). This is even after the chapters of contradictions and not-invented-here and vendor specific undocumented holes were found. Maybe, OOXML should have been eventually approved once OOXML was fully cooked, but OOXML is a junk standard that even MS (who knows the details of the undefined sections of the standard) can't implement until their next release (although they are able to implement ODF).
The practices involved in getting OOXML approved are beyond playing hardball. They're outright fraud.
OOXML defacto standard?!! Not even the latest Microsoft office product support it. Even if you count the very similar XML based document format that is used in Microsoft products as similar enough to the actual ISO standard to count as OOXML, you would find that there are far more users of ODF. The defacto standard, if there is one, is still the old Microsoft doc format.
The problem is that, the original suggested OOXML standard had references to Microsoft products saying things like funktion X should behave like in MS-Office 97, without any formal specification on how it worked in that program. In reality that meant that nobody but Microsoft would be able to fully implement the standard. No wonder IBM was against it.
Second, it is sort of stupid to have two standards for the same thing. If the current standard isn't good enough it is better to improve it than to enter yet another standard. Microsoft had every opportunity to influence ODF so that it would fit their technical needs, but didn't want to do that. A standad would be a threat to their near monopoly position on office software.
When ODF caught on, and finally got its ISO status that cat was out of the bag. The Microsoft reaction is to release their file format for ISO standardisation well knowing that it would be very hard for third parties to implement in real products. Fortunately ISO caught the worst part of this, and Microsoft is now in a situation where their own product doesn't support the OOXML standard. In fact, Microsoft will support ODF before they get to support their own standard.
>Little used ODF rammed down our throats.
Im not too familiar with this term. Can you please detail how this ramming is achieved?
As for OOXML being defacto standard, please indicate what programs use this so called standard.
As someone who read Andy Updegrove's blog, your writing seems to come from a parallel universe.
I can see from your article that you are just that kind of self-centered US-citizen.
I disagree with you. "We" European don't dislike Microsoft because its American. "We" dislike it because it did not use open standards, or even well documented file formats.
You have handily blurred the lines between standards, copyright, patents, and licenses, as well as redefined de-facto to suit your own purposes. You have also blurred the distinction between document formats and office software. And while IBM has recently led the charge against ISO and its practices, you have completely ignored the four national government body's complaints against the ISO procedures and outcome of the Office Open XML standardization process and the very recent resignation of 13 Standard Norge members over the mishandling of that country's vote for OOXML because they did not want their personal reputations sullied by their association with that organization.
Let's start with a definition of terms:
De facto: Actually; in fact; in reality; as, a king de facto, -- distinguished from a king de jure, or by right.
Standard: That which is established by authority as a rule for the measure of quantity, extent, value, or quality; esp., the original specimen weight or measure sanctioned by government, as the standard pound, gallon, or yard.
Copyright: The right of an author or his assignee, under statute, to print and publish his literary or artistic work, exclusively of all other persons. This right may be had in maps, charts, engravings, plays, and musical compositions, as well as in books.
Patent: To grant by patent; to make the subject of a patent; to secure or protect by patent; as, to patent an invention; to patent public lands.
License: Authority or liberty given to do or forbear any act; especially, a formal permission from the proper authorities to perform certain acts or to carry on a certain business, which without such permission would be illegal; a grant of permission; as, a license to preach, to practice medicine, to sell gunpowder or intoxicating liquors.
Proprietary: Belonging, or pertaining, to a proprietor; considered as property; owned; as, proprietary medicine.
With these definitions in hand, let's examine your remarks.
"Some commenters think I was unfair to IBM in my last post. There I took Big Blue to task for its announcement that it intends to wage war against Microsoft in the world’s standards bodies. The motivation for this bellicose declaration was IBM’s stinging defeat last Spring in its battle to prevent the International Organization for Standardization (ISO) from ratifying Microsoft’s de facto office document standard (OOXML)."
IBM is not simply waging war against Microsoft. It is waging war against the ISO and any other vendor that would attempt the same kind of manipulations Microsoft did. Also, it is not just IBM that is waging war. Four official national government standards bodies waged war before IBM ever took a stand: Denmark, Brazil, India, and South Affrica. And their complaints were not aimed at Microsoft; they were aimed at the ISO.
Their legitimate complaints of mishandling of ISO standard procedures during the process of ratifying the OOXML draft, properly formed and submitted as official complaints, were mishandled as badly as the ratification process itself.
IBM's war is also about maintaining the credibility of the standard-setting process and of the organizations entrusted with setting standards. There has been much discussion, both inside, and outside, of IBM about the credibility of the ISO as a result of this one proposed standard. And this is unprecedented.
IBM and others opposing the ratification of Microsoft's OOXML format specification were not simply seeking its defeat. They wanted the ISO to acknowledge that its more than 6,000 pages of draft were inappropriate for the fast track process and needed careful review and revision before being published.
Microsoft's format specification is not a de facto standard by any measure. First, it has not been implemented by any software vendor, not even Microsoft. Second, until it is implemented, and until it is, in fact, widely used, calling it a de facto standard is an exercise in fantasy.
"Let me be clear. I’m not a fan of Microsoft Office, and I really don’t care much one way or the other about OOXML. But I am not impressed with OpenOffice as an alternative to Office."
Here, you are blurring the distinction between document formats and office software. Perhaps the fact that document formats have traditionally been closely associated with particular programs and have been kept closely guarded secrets has confused you about the nature of the ODF document format standard. The difference between ODF and OOXML here is that ODF is not proprietary, whereas OOXML, although it has been ratified as a standard, contains numerous references to proprietary specifications.
This renders OOXML useless as a standard. Standards are rules that define the measure of a thing, so that when there is more than one of a given thing, people can evaluate its quality and conformance to a reference. Microsoft admits that its own file formats do not conform to this standard. And since the standard refers to undefined proprietary bits, it cannot be implemented independently of Microsoft.
ODF, on the other hand, has already been implemented independently of OpenOffice.org numerous times. It is the default file format of KOffice; it is an option in AbiWord and Gnumeric; it is now supported by Apple's TextEdit software; Correl Office is implementing it; even Microsoft has announced it will be available in the next version of its office suite ahead of OOXML.
"The suits in Armonk are shocked (shocked!) to discover that Microsoft actually tried to influence the outcome of a debate where its vital interests were at stake, namely its ability to sell Office to the world’s governments."
There was never anything preventing Microsoft from selling its office software to the world's governments other than Microsoft's own stubborn refusal to conform to the requirements these governments imposed for purchasing software that reads and writes document formats that are freely implementable and officially sanctioned standards. They could, after all, implement ODF in Microsoft Office without loss of function. This would have satisfied all concerned. They stubbornly refused to do this.
"Strenuously lobbying for one’s interests is a normal part of the rough and tumble business of setting international standards. I’m pretty sure that IBM has lots of friends in the world’s standards bodies too, probably a lot more than Microsoft. Hmm, wouldn’t it be interesting for someone to throw a little light on that?"
There has been plenty of time for Microsoft to do just that, if it could find anything egregious to throw light upon. But the fact is, this is not just about IBM's finger pointing. It is about several government standards bodies pointing the finger. Not at Microsoft for strenuously lobbying for its own interests, but at the ISO for caving in to pressure from Microsoft when it would clearly harm their own reputation and that of the standard setting process and would not benefit anyone but Microsoft.
"In a nutshell, Big Blue conspicuously fails to practice what it preaches. On the one hand, its open standards guy Bob Sutor piously affirms that he wants nothing more than to promote standards which will “ensure that each of us can easily purchase and interchangeably use computing technology from multiple vendors”. This is what they say when they are talking about a market where they are losing, such as office software. On the other hand, when it comes to a market where they are winning, such as the mainframe, it’s a different story. Here IBM uses threats, lawsuits and buyouts to shut down tiny competitors who want nothing more than to pay for the right to run Big Blue’s dominant z/OS mainframe operating system on non-IBM hardware."
Here, you are equating standards with licenses and copyrights. A standard provides a reference to serve as an environment in which companies may compete for business. A license defines the terms and conditions under which one may or may not use another party's copyrighted or patented work. They are not interchangeable, and they are not mutually exclusive.
IBM can fight for standards and the standard-setting process. And they can protect their copyrights and licenses. And they can do both without hypocrisy. You are suggesting that IBM, and hence, anyone else, cannot promote standards and at the same time protect their interest in copyrighted and licensed works. This is preposterous.
"What part of “easily purchase and interchangeably use technology from multiple vendors” do IBM’s mainframe lawyer hounds not understand?"
What part of copyright law and legally-binding license do you not understand? IBM's mainframe law suits have nothing to do with easily purchasing and interchangeably using technology from multiple vendors. IBM's (and Microsoft's) software licenses are restrictive, no doubt. And enforcing those licenses and protecting their copyrights is their right. IBM's software licenses restrict users from installing it on hardware from other vendors. So does Apple's software license for Mac OS X, by the way. But that does not prevent anyone from purchasing competing products, and it does not prevent those competing products from inter-operating with IBM's products.
Microsoft's products are another matter, though. They have fought hard to prevent competitors from creating competing products that could successfully inter-operate with their own.
They have crafted and thrust upon hardware vendors licensing terms that prevent them from installing competing software on the hardware products they sell. And the linchpin of their ability to keep the market to themselves is their proprietary office file formats.
"Once upon a time, say around two years ago, IBM had the following noble declaration on its web site:
'IBM has an open approach to patent licensing for products in the Information Technology (IT) field and is generally willing to grant nonexclusive licenses under reasonable and nondiscriminatory terms and conditions to those who in turn respect IBM's intellectual property (IP) rights.'
"Well, that page is gone now (though the WayBack Machine has preserved its memory). In its place, in the far less public venue of its court filings against its former competitor Platform Solutions Inc. (now a wholly owned subsidiary of – heh, heh – a certain large computer company based in Armonk, NY), Big Blue sings a very different tune about openness:
'IBM has refused to agree to license its patents to PSI and its copyrighted operating systems and other software for use on PSI's emulator systems despite PSI's demands because, among other reasons, IBM has no obligation to do so...'"
Here, you are blurring the distinction between copyrights and patents, as though they were interchangeable. IBM has demonstrated their willingness to license their patents, having contributed several for use by free software developers, as well as donating code to the open source world and contributing to large free software projects, including the Linux kernel and OpenOffice.org. That they protect their copyrights and enforce the licenses that apply to their software has nothing to do with open policies surrounding patents or their advocacy of open standards.
" While it would probably be prohibitively expensive (read: unprofitable) for a competing vendor to build a full-scale office suite around OOXML, Microsoft has at least put the format out in the open for all to see, and offered a grudging but seemingly far-reaching promise not to sue people who try to implement it.
"Of course the IBM camp complains that this promise is ambiguous because, well, some developers may not be intelligent enough to understand the inevitably convoluted legal language of Microsoft’s promise."
Wow. calling developers stupid. There's a sure method of swaying people to your point of view. Nevertheless, you, yourself, admit that implementing this standard outside Microsoft is probably impractical. Of what use is it, then?
The only reasonable answer to that question is that it is of use only to Microsoft. It is a tool to allow Microsoft to offer a format to governments who demand a de jure file format standard be used by software they purchase, while still maintaining their control of the file format by making it prohibitively expensive for anyone else to implement.
That IBM, Denmark, Brazil, India, South Africa, and 13 members of Standard Norge are taking a stand against ISO and Microsoft says that it is no longer acceptable for a single vendor to forcibly control the entire market for office productivity software and to hold public records hostage to a single vendor's file format.
First, pick the least controversial events from the ISO proceedings and act like nothing else, more sinister actually happened. Then, spin IBM's involvement in trying to straighten out that mess as a BAD thing. Throw in a good measure of the same tired lines, "MS Office is used the most, so it must be the ONLY good; if everyone else is doing it, why shouldn't you?" Finally, try to make everyone that has a valid complaint against the OOXML ISO standards are mere snot-nosed sniveling kids, and that's the idea of this one, folks!
Hmm... Microsoft has said it does not now support OOXML and it has no plans to support it. Further, it said it will add support for ODF.
Microsoft has chronic problems playing nice. It always has. ISO is just the latest to be favored by its gentle touch. ISO has been pretty much destroyed. Not much is happening there any more. Out of the smoldering ruins rises the OOXML standard, a document that nobody wants, not even Microsoft.
IBM vs. Microsoft is an interesting match. IBM has in the past shown Microsoft how to establish and maintain a monopoly. Microsoft learned their lessons well. IBM has changed over the last couple of decades to a much less predatory company. Maybe Microsoft will mature too. Time will tell.
Ever think of helping yourself to some anger management?
Hardhitting responses to dirty journalism, it is good to see.
That done, how about the one correct observation, on IBM having firmly reestablished its illegal monopoly on mainframes ... any thoughts on what may or should take place in that substantial portion of the industry?
It appears the article author thinks -erroneously - that ODF is "little used" in the market place. To clear his head with facts, the American Library Assoc. has implemented OpenOffice in most all public libraries in USA. That is tens of thousands of PCs running Open Office.
Furthermore, he is obviously ignorant about the "formal" adoption of ODF by my most European Union Countries, the governments of korea, Taiwan, South Africa, most all South American Countries, and is being strongly considered by California, Indiana, Oregon, Texas and other states. There are plenty more.
Maybe he gets his info from the Microsoft FUD marketing division. He needs a "reality check".
W. Anderson
wanderson@nac.net
Hold on Jeff. I’m also upset by IBM monopoly and propaganda machine. Judging by the comments posted here, the IBM propaganda is a very efficient brainwasher: they wear a tiny ODF cloth and they’re forgiven all their monopolists sins and abuse of reason.
Open your eyes: Microsoft is not the only evil in the IT world.
Supporter: That there are evils committed by IBM, or anyone else, does not excuse those being committed by Microsoft. Hijacking the standard-setting process and destroying the credibility of the largest standards organization in the world just to protect one's monopoly market share is an inexcusable Microsoft sin. Sorry, but there is much more than IBM propaganda to the charges against Microsoft and the ISO.
It is not propaganda that work on several ISO standards has ground to a halt because of Microsoft's unprecedented stuffing of committees with their own OOXML supporters, who have since ignored all ISO requirements to participate in ongoing work on other standards:
Exhibit 1
Exhibit 2
Exhibit 3
It is not propaganda that Microsoft will not implement their own standard until after they implement ODF. Therefore, it is also not propaganda to counter the assertion made here that OOXML is a de facto standard.
The complaints of several countries against the ISO's processes do not come from IBM, hence, they cannot be credibly attributed to IBM.
Do you have anything to offer, other than excuses for Microsoft's behavior and unsubstantiated claims of IBM brainwashing?
You accuse those commenting here of being brainwashed with nothing to back up your remarks. It took a lot of space to untangle the muddled use of terms in the original article and to respond to them. The original author's article was clearly shown to be propaganda itself, and he has not had the courage to answer a single rebuttal.
Scott,
this post by Jeff is about IBM and ALL comments are about ODF/Open XML and Microsoft.
By focusing your comment on Microsoft, you blur IBM's behaviour in the mainframe area and accept their blund mainframe monopolistic practices.
IBM is much larger than Microsoft. Their mainframes control a much larger part of the world information. As explained by Jeff, they are guilty of sins against IT freedom.
Comment IBM's behaviour here and keep your comments on Microsoft for posts about Microsoft.
You are off topic.
Frequency of the following terms in the original article:
Microsoft: 14
OOXML: 6
(Microsoft) Office: 7
Total for MS-related terms: 27
IBM: 21
OpenOffice (Sun Microsystems): 6
ISO: 7
ODF: 2
Based on the frequency of Microsoft-specific vs. IBM-specific terms alone, it is safe to say the article is as much about them as IBM. There is nothing off-topic in correcting gross miss-statements. But like the author, you refuse to deal with facts. And since you are not the topic police, screw off.
You're losing your temper, Scott.
Jeff Gould is clearly a Microsoft shill.
Jeff, do you not have any shame? Or are you just totally ignorant?