Apache up against corporate dominance of fake standards process?

By Rick Jelliffe
March 17, 2009 | Comments: 14

Read the rather startling comments to the article The long-running Sun-Apache dispute at Javaworld.

The background is that Sun twice started the process of submitting Java to become a standard (once through Ecma, once through ISO, IIRC) only to withdraw. Finally it came up with its own Java Community Process (JCP) where a wide variety of Java Specification Requests have been made (equivalent to W3C Recommendations). The JCP is widely seen as a way for Sun to get community involvement and the benefits of openness while retaining an effective Sun veto.

One product of the JCP has been the Java Technology Compatibility Toolkit (TCK): a suite of tests, tools and documentation that determines whether or not a product complies with a particular Java™ technology specification.

It seems that Apache claims they cannot use the TCK to test Apache Harmony (their open source Java platform) because it would impose certain restrictions on use. It seems that passing the TCK is more important than just getting a branding tick: some JSRs are only licensed for systems that pass the TCK.

And here is the startling comment: former Apache chairman and director Greg Stein reports that

The problem? Many of the agreements surrounding the JCK and the JCP, even the operation of the JCP, are covered by Non Disclosure Agreements. This is why Stephen, myself, and others can't provide exact language.


People who think that open processes are not compromised when a single vendor (or a cabal of vendors) can dominate the process, or who think that subversion begins and ends with Microsoft (i.e. if Microsoft is involved it is probably tainted, if Microsoft is not involved it is probably OK), should learn from the inadequate JCP/TCK process.

And legal definitions of "open" need to start include clauses that a standard is not open unless there are no non-disclosure issues with participation: transparency.

You might also be interested in:


Sun is making money on java. it's always has been true. Hence they aren't interested in *any* openness till they forced to collaborate.

So there is nothing to expect till someone (IBM?) buy sun.

Apart from the usual morality play dilemmas of greed and stupid, the challenge of being open and transparent while managing IP contributions without submarines remains.

Technologists have to acquire the patience to read participation agreements and the maturity to accept that in an imperfect world, some rules are necessary for the greater good. As much as we'd love this to work like an Andy Hardy movie, it can't.

But at least some are now beginning to understand it isn't a Microsoft Vs The World problem.

@len : "Technologists have to acquire the patience to read participation agreements and the maturity to accept that in an imperfect world, some rules are necessary for the greater good. As much as we'd love this to work like an Andy Hardy movie, it can't."

Len: please don't erect a straw man. Who said that there should be no rules? Please provide a citation. Greg Stein certainly didn't say that there should be no rules. The Apache foundation itself is all about rules. Rick Jelliffe didn't say there should be no rules. Who are you addressing?

But more important: it's not sufficient to IMPLY that SPECIFIC rules are necessary for the greater good. You must actually MAKE THE CASE. How is it good for us that the Apache foundation cannot even publicly discuss the legal documents that they disagree with? Please explain to me how that is of benefit to anyone other than Sun? We're talking about a specific rule. How is it in our benefit?

Java is a big revenue maker for Sun. They will give it off if Apache Harmony is tested & approved by TCK. Sun is also worried, if J2SE is open and Harmony takes over, the next thing Sun will loose is J2EE and J2ME which will basically destroy Sun's Java business. Hence, no TCK for Apache!!!

Remember Microsoft's attempt at stealing Java called J++? In a perfect world, literally everything could be open and no one would steal it. Openness is good generally speaking but there are two sides to every coin. If Sun is a little protectionistic, I can understand.

Just a reminder.

Abdel: I hope they can find some way around it. If IBM buys Sun, I wonder whether the openness would success or the protectionism would.

A lot of FOSS people would prefer Harmony to Mono, because they don't want to entrench a proprietary technology. I know basically nothing about Mono and .NET, but I have been so disappointed by Sun's mismanagement of Java for the desktop: starting with the decision to abandon HotJava and not to maintain the browser classes.

Jim C: Ah, so Sun is justified in impeding FOSS and it is Microsoft's fault? Sorry, but it sounds a robotic response.

Yes, I remember J++, where Microsoft failed to graft their favoured GUI API onto the Java base. And I remember SWT, where IBM succeeded in grafting their favoured GUI API onto the Java base. Both are against WORA but people don't mention IBM...

Actually, as I have written before, I think it is a sign that the JAVA API was insufficiently modularized, which rather calls into question the whole WORA idea, for desktop apps at least.

Monolithic technologies, no matter how appealing their promises (interop, stability, reliability), just don't work. They are not organic.

Yes, I remember J++, where Microsoft failed to graft their favoured GUI API onto the Java base. And I remember SWT, where IBM succeeded in grafting their favoured GUI API onto the Java base. Both are against WORA but people don't mention IBM...

You are kidding right? First of all, there was a huge row when SWT came out. Second of all, nothing is grafted onto the Java base at all. Microsoft tried to do much more than "graft their GUI onto the base".

@Robert: Good point. Microsoft J++ omitted Swing and replaced it with Windows-friendly graphics API, IIRC. SWT is not part of a "Java" distribution, so it is strictly an alternative to Swing. And, yes, of course there was a fuss about SWT at the time, but no-one mentions it now.

But my point about grafting was "Both are against WORA..." and that in the context of the larger point that "protectionism" is control is proprietary-ism.

@paul: The question is one of fake standards vs real standards. It is tough to call what Sun and IBM produce real standards. They are controlled access proprietary technology. Otherwise there would be little objection to J++.

The point is not should there be rules or no rules. There are always rules. The question is are the rules transparent, are they applied openly and fairly, and do they protect the product of the process.

Some years ago the argument was anyone should be able to participate in standards processes. Two immediate problems occurred:

1. Some people are not professionally capable of contributing to the standards process. Qualifications matter. Intent matters.

2. Some contributions do not qualify because they are proprietary IP, the so-called "submarine patent".

As a result, it isn't possible to have a legitimate standards process in which all possible or requestors participate. Further, until a standard reaches a point of development in which all of the qualifications are met, it isn't possible to make all work of working groups publicly available.

To esnure that qualifications are public, transparent and are signatorially verifiable, standards organizations and consortia have created participation agreements. These are the contract to which all parties are obligated before they are enabled. My point was it is necessary and required that these be public, be read and be signed.

The case can be made by citing those agreements. If you can, do so. If you can't, this is a waste of time.

Paul, Len: Both of you are saying the process needs to be transparent to be legitimate.

I think Len's point is that to prove anything is actually amiss we need evidence, while Paul's point is that to prove everything above board we also need evidence. But the NDAs stop both, and so we can say neither; or we fall back on our prejudices and make judgements based on anecdotes and experience.

(That is why I avoided any speculation about the alleged information that the NDA protects. It may be that Stein is blowing smoke, for all I know, without disrepect to him. And maybe the Sun people made the JCP NDAs genuinely believing that it was useful and harmless.

But it is not a question of motives, but of category: openness must be open, and people demand open standards now.

@rick: I'm saying it can't be open to anyone who doesn't sign the agreements. Otherwise, the submarine patents ensue and indemnity breaks.

The NDA has to be bounded. The participation agreement has to specify the precise point in the process at which a document is published as an open statement, it's release point to a larger audience that is not enabled to submit material because technical submissions are disallowed at that point.

An open-ended NDA is quite another thing.

So if it is categorized, it is process bound and that process must, as you say, be transparent. But it must also be transparent and agreed that the document is open to non-signatories only at the agreed upon point.

I'm not arguing what people demand. They demand the right to bear assault weapons. They demand the right to unlimited copying of copyright material. Their demands aren't always the right thing and a legitimate verfiable process that is legally binding aren't always the same thing.

Len: Without saying at all that formalized agreements by internal participants in standards groups is not prudent nor necessary, it is a mistake to think that such agreements give protection against unexpected IP. (I don't think Len and Paul were suggesting this.)

Someone who has

a) never applied for a patent
b) never applied for a patent in the US
c) never applied for a patent in the US before 1995
d) never applied for a patent in the US before 1995 which has been granted
e) never applied for a patent in the US before 1995 which has not yet been granted and which would cause a procedural larche

has no chance of owning a "submarine patent" AFAIKS. Increasingly rarer.

However, even you do have a submarine patent, participating in a standards process to the end without disclosing this would raise its own set of problems/defenses/remedies. Fraud at worst; I have blogged earlier about the (scant) case law, but I think you would expect a voiding of the right to require licenses even if the submarine patent was upheld. Andy Updegrove's site has some good material.

None of the standards bodies assert that their standards are free from unknown third-party patents. Participants must disclose, and in the case of ISO the editor makes a reasonable effort to track down known patent issues (for example, I contacted the one known patent in the area of Schematron, and we worked out which aspects the patent claimed to relate to, and I made sure the wording of Schematron avoided any overlap; and I did an online patent search, trawling through.)

But there is always a possibility that new standards could have something nasty in the IP attached. If it happens, and things cannot be resolved adequately, we make a new standard as an alternative. (And, following the PNG/GIF example, the existence of the new standard may make force the patent holder to become more reasonable.)

Furthermore, if you look at the patent allocations that the big boys give (OSP, covenant not to sue, etc) you see that they apply only to the standard in question not to alternatives.

It always amazed me during the ODF/OOXML fights, that people who made the claims that we shouldn't use OOXML because it might have all sorts of a patents applying didn't see that the same applied, even worse to ODF. If MS has a patent on some office application thing that OOXML used, the chances are that it would also apply to ODF, which had (then) no license.

That this was not brought up indicates to some extent that the issue was FUD, or at least that people who found it easy to think ill of Microsoft also were not inclined to follow logical consequences that might seem to soften ODF's crunchy goodness.

So standards bodies have to muddle through. Knowledge in the marketplace is never perfect. There is always risk. An NDA does not increase knowledge in the marketplace, it decreases it.

It would be great to have standards organizations on a better footing. I think the footing has to be on competition law, and I think it has to be (an incremental) part of wholesale IP/Patent law reform.

In particular, I think that once a patented technology achieves market dominance, the patent licensing rights must be revoked for public policy reasons: to establish appropriate market/bazaar conditions. Once a technology is dominant, the patent owner still has first mover advantage.

I don't think this is as fanciful as it seems. The writing is on the wall: the move by Europe to require Microsoft to de-proprietorize their formats (to the extent that it is possible using the standards mechanism) is just the start of this. I think many comments by various anti-trust regulators around the globe indicate that this will increasingly be happening.

At the moment, it will be case-by-case, but I think that as the movement for "open standards" increasingly rejects its role as monkey for some corporate organ grinders and stands up for itself, then the connection and necessity for clearer positioning of standards w.r.t. anti-trust and IP will become conventional wisdom.

"...necessity for clearer positioning of standards w.r.t. anti-trust and IP will become conventional wisdom."

One hopes. On the other hand, I'm watching the deja vu all over again over in the virtual worlds industry where absolute dumb prevailed. The new kids didn't take precautions and now some old kids are taking their assets to the bank with the usual "only the laywers profit" inevitabilities. In the participation rules we wrote for Web3DC, they only apply to signatories, to submissions made, and they effectively close the working groups to members only who are signatories. Cold but rational. One side effect was IBM's "3D Standards initiative" included all of those who wanted proprietary patents in their work and actively shut out those who didn't. It was and probably still is a thinly disguised initiative to seize as much market-the-IP high ground and commons as possible.

The side effect to Linden Labs is they are now in active conflict with their users for that and other initiatives (get porn off the main continent). It ain't pretty if something starts with one set of goals and values and then has to shift hard to stay in business. So I advise anyone with a so-called 'open' initiative to look very hard at those agreements. As the twig is bent...

No, we can't always know what else is out there and patents assume a best faith search which experience reveals not only doesn't always take place, there are examples (eg SUN) where the lawyers told them not to search, not to review prior art, in other words, to "stick with their lie" as the commedienne said. Mammals... relentless.

The NDAs cast too wide a net but I can understand how someone thought it a good idea only in terms of participation. I was just looking at the OSLO M site and noting some rules for protecting its development through who joins and what they agree to for the privilege. This does seem to be the way some are choosing to work online these days possibly because of conventional wisdom.

What do you think those NDAs were there to achieve really? IOW, is it a case of drinking bad water or of coming back and poisoning the well?

Frankly, old friend, I despair of the open standards processes these days.


With an existing international standard and processes in place, this is what we contend with.

News Topics

Recommended for You

Got a Question?