This is a read-only archive. Find the latest Linux articles, documentation, and answers at the new!


OASIS: Meaningful open standards or mirage?

By Jay Lyman on May 24, 2005 (8:00:00 AM)

Share    Print    Comments   

Despite some recent criticism -- which included a call to boycott standards based on reasonable and non-discriminatory (RAND) licensing fees -- the OASIS standards body is behind what many in the open source community view as a primary example of a truly open, industry-accepted standard: the new OpenDocument format.

Now that the OpenDocument format is official, and with its recent truce on Web services standards, there is a sense that the standards body is on the same path as the W3C, which is viewed as truly dedicated to open standards. However, OASIS critics still complain of the continued RAND policy and the patent encumbrances that stand to benefit some OASIS member companies rather than the industry at large.

Struggling toward openness

Larry Rosen, a software law expert and leader of the RAND reaction, said the concerns were focused specifically on RAND licensing within OASIS standards. He indicated he was unaware of any patent encumbrances in the OpenDocument format, and was therefore unconcerned about it.

"I see any standard that is not encumbered by patents as being a true benefit to the public," Rosen told NewsForge.

While he continued to protest the OASIS policy that still allows RAND fees, Rosen said the circumstances had changed in favor of open, patent-free standards since February, when a group of open source leaders, including Rosen, Bruce Perens, and others called for action against RAND patent standards from OASIS.

Describing it as vague, Rosen said IBM's patent pledge was nonetheless being applied more recently to OASIS standards, which Big Blue indicated would be free of patent issues, as least those from IBM.

"That, to me, is a victory," he said. "Assuming IBM means what they say, then at least IBM patents are off the table, which is a big chunk of patents. Microsoft hasn't made the same pledge, HP hasn't made the same pledge. Most other companies have not, but I'm not as worried when IBM wants standards free of [patent encumbrances]. I think the IBM declaration was a very big deal."

Rosen still had criticism for OASIS maintaining its policy that allows RAND, and indicated he has been unable to get any movement on the matter. Still, he again stressed the IBM patent pledge was a greater force in favor of open standards, rather than RAND ones.

"Given IBM's declaration, I can't imagine OASIS going ahead with RAND," he said. "They don't want to, even though they refuse to take it out of their policy. We'll see. We will look at [standards] as they come and say, 'OK, OpenDoc, this is good,' and then next, one at a time."

Columbia Law School professor and FOSS advocate Eben Moglen told NewsForge he sees OASIS as, "struggling, primarily successfully, toward achievement of open standards."

Moglen said the FOSS community was welcoming the new OpenDocument format, which he described as "not perfect, but a significant achievement." Still, the software law expert referred to struggles within OASIS and the lingering question of patent encumbrances in standards.

"We're watching as standards-making in industry goes through an ethical change," he said. "It used to be that standards bodies were closed clubs. If there were patent encumbrances, it was just part of the game. With FOSS, a different philosophy has to be taken when it comes to standards."

Moglen credited the W3C for standing up first and fiercest for open standards, which brings up another charge of Rosen and other OASIS critics, who contend those interested in patent profit from industry standards had moved from the W3C to OASIS.

Never used, so never mind

OASIS President and CEO Patrick Gannon dismissed that charge, telling NewsForge the recent decision to proceed with both the competing WS-ReliableMessaging and its own WS-Reliability Web services standards under the OASIS umbrella, for example, was not made because of the OASIS policy on RAND and patents.

"We've already committed to a royalty-free Web services stack," Gannon said, referring to OASIS as more of an application-to-business-problems body compared with the W3C. "They really are picking where best to do that work. The recent decision was not made because of our policy, it's about finding the best fit and bringing standards to market."

Gannon defended the standards body's overall dedication to open standards, indicating that as far back as standards such as ebXML, OASIS has supported open standards and open source.

"The largest majority of our standards are implemented in a royalty-free basis," he said. "We've had it in place. We've had that commitment. We think our record in this area is stellar."

Gannon said although they are not necessarily publicized, there are more than 10 SourceForge projects, for example, that are based on various OASIS specifications, and the standards body includes high-profile FOSS contributors including Debian and BlueCode, which was recently acquired by IBM.

"We don't issue press releases around them, but they're there," he said. "The evidence is out there. We have been a fertile ground for [open standards and open source projects]."

As for the continued concerns around its RAND policy, Gannon downplayed the issue and indicated the policy is rarely used.

"The instances of requiring license fees have been pretty close to non-existent, even under a RAND-allowing policy," he said. "We expect it to be little-used going forward."

Gannon also said in the absence of a drive by the majority of OASIS vendors to eliminate RAND from the policy, it will likely remain. "If it's an option nobody uses, then what's the harm?" Gannon added.

OpenDoc: strength of open source

Sam Hiser, former marketing lead of, told NewsForge that the OpenDocument's finalization as an OASIS standard is a good model for truly open industry standards. He contrasted OpenDocument with other implementations of W3C XML standards for office documents by software companies whose implementations are not open to free access by all.

"Such implementations of XML have proprietary characteristics which require a license or the purchase of the company's software to enable access to the information in their XML formatted documents," he said. "We hope that decision-makers -- particularly in government as they legislate new IT procurement policies and laws in their jurisdictions -- are taking advice to help them recognize the important differences between OpenDocument and the less-optimal implementations of XML file formats for office documents."

Despite the praise for OpenDocument, Hiser reiterated the RAND cries: "Any objections to RAND itself must be amplified to the heavens! Patents are appropriate for mechanical inventions but miss the original intent for digital inventions, i.e., software. Patents must be kept away from software because software has the nature of being easily duplicated and widely distributed and standardized at zero cost in practically no time in a way that mechanical objects cannot. That alone isn't enough justification for trashing the notion of patents -- which still have relevance in the bulky material world. Now that we have demonstrably superior methods for innovating software through open collaboration, there is no benefit and only harm from applying the philosophy of patents to software.

"Companies like IBM, Microsoft and others benefit from software patents by stopping the free market of ideas from operating in the realm of digital inventions," Hiser continued. "Their efforts at creating and protecting their patent portfolios is an old-world approach to keeping individuals and small groups of people from exploiting their own brain power and digital resources and competing effectively. It's a cynical attitude which, luckily, has little chance of surviving when small groups continue to show superior invention and business service delivery around collaborative works."

Hiser poured on the criticism for OASIS, but also referred to a protection sought and won by and others in the new OpenDocument format.

"The events around OASIS' recent amendments to their Intellectual Policy Rights (IPR) policy are detestable and not befitting a body of OASIS' mandate: 'Organization for the ADVANCEMENT of Structured Information Standards.' This IPR policy change goes against the name," he said. "Luckily, too, has a 'kill' clause which would nullify its distribution licenses under circumstances of change in IPR policy, thus nailing's OpenDocument format permanently into the public trust."

Hiser criticized the OASIS RAND policy and IPR change and credited it largely to Microsoft, but indicated it is no match for the push forward coming from FOSS.

"OASIS seem to still feel that there may exist some proprietary software somewhere which qualifies for special treatment as more worthy of protection from competition than other software," he said. "This fallacious view will be revealed in the light of longer experience and deeper adoption of the unparalleled innovation coming out of open source and free software projects. I would say that our experience with the quality of software (and the speed of its improvement) coming out of open collaboration defies this view which is undoubtedly influenced by the corporate constituents of OASIS who have the most patents and whose approach to business is the most threatened by open collaborative software methods and products."

Share    Print    Comments   


on OASIS: Meaningful open standards or mirage?

Note: Comments are owned by the poster. We are not responsible for their content.

Look deep within "Your Self"

Posted by: Anonymous Coward on May 25, 2005 12:02 AM
If the same attention had been paid to OpenOffice & Sun's use of Java within OOo, as has been paid to OASIS, then we wouldn't have the Java problem in OOo.

Just substituting "Java" for "RAND" in the above post pretty much highlights the issues.

kettle -- black -- pot.


Looking deep, finding a troll

Posted by: Anonymous Coward on May 25, 2005 07:18 PM

FUD and rubbish, the issues are not even vaguely similar. If they were, GNU/Classpath would not have been able to build a full set of classes so that can run in a completely Free environment.

But they have. And they will continue to do so. Take your anti-Sun trolling elsewhere.


As long as it uses Java

Posted by: Anonymous Coward on May 25, 2005 08:49 PM
it isn't completely free.

And with Sun's funding of SCO, with Sun's trashing of Red Hat, Sun's continuing attacks on Linux, Sun's pushing of Solaris as an alternative to Linux when the world is going to Linux, I find it ironic that you are associating Java/Sun with Free.

Sun is doing everything it can, Sun is saying everything it can to drag Red Hat/Linux through the mud. Sun is injecting FUD into the "StarOffice is protected on patents, can't say the same for OpenOffice use in business", basically Sun needs Solaris to succeed at the expense of Linux. Or they will die and they know it. And most everyone else knows it. Including the analysts.

As for trolling, just look at the news recently on Sun's injection of Java into OOo via its programmers working on the distro. You might have legitimate grounds for the charge if this were a non-issue. But it is an issue, an issue that has been dogging OOo, an issue that is under constant discussion on the various tech news sites. As much as you wish this issue to remain under the radar, it isn't going to go back under the radar. The community is very well aware of the problems of Java in OOo.

So keep supporting Sun, fanboy. SCO has its fans as well. Keep supporting the anti-Christ of Free Software.


Reminds me of IETF + M$'s SenderID and VRRP

Posted by: Anonymous Coward on May 25, 2005 12:14 AM
The IETF lost a lot of respect from me personally when it was ready to allow patent-encumbered technology (actually, patent pending in Sender ID's case) as official standards. The W3C, too, was about to cave into M$ and Apple on this, until major pressure from very important groups like the Apache Software Foundation told them Hell, No. I believe that, had not Apache had over 60% of the HTTP server market, the W3C would've rolled right over, and our objections would've been viewed simply as white noise. This certainly happened when the IETF approved the VRRP "standard" even though Cisco then claimed, and still claims, patents on it. The result was OpenBSD's Common Address Resolution Protocol (CARP), which I can confirm works very well.

What that means is that we need to make sure that our Free Software continues to gain market share, just as Apache has. We as businesspeople and engineers need to support only software that is unencumbered by patents or "patent-pending" claims. As Free Software continues to gain market share, groups like OASIS and the IETF will be essentially forced to change their attitudes toward RAND policies, as the W3C has, or face irrelevance and eventual extinction. If, on the other hand, we keep supporting platforms like Windows, MS Office, and closed UNIX, that won't happen, and it will be our fault.


We need clarity here

Posted by: Anonymous Coward on May 25, 2005 04:15 AM
As for the continued concerns around its RAND policy, Gannon downplayed the issue and indicated the policy is rarely used.

If it's rarely used, what's the point of having it?

A standard must never be encumbered by patents. It's as simple as that. No ifs, buts, maybes, or other loopholes.

There is no such thing as "reasonable and non-discriminatory licensing fees". Any licensing fee discriminates against the Free Software community. That's precisely why companies like Microsoft want them.


time is a factor too

Posted by: JelleB on May 26, 2005 07:28 PM
If nobody is now collecting licencing fees for OASIS standards, does not mean that it will never happen. Most standards don't make it at large, so collecting fees at the beginning will slow or stall any emerging 'standard'. But when the standard is finally accepted by a large userbase, the company that initially refrained from collecting, can now collect ten times as much, as it is very hard to move away from the standard once you are committed and have actual code. That is the problem with patent encumbered standards, the fact that now few cases exists does not men we should walk with open eyes into this ransom trap.
Some glimmer of hope is that swpats may not be allowed in europe, so you yanks still have a sane place to migrate to anyhow.


This story has been archived. Comments can no longer be posted.

Tableless layout Validate XHTML 1.0 Strict Validate CSS Powered by Xaraya