Torn between a lobby and a FRAND

Ahead of a European Commission and European Patent Office conference on intellectual property rights and standardization taking place in Brussels on Monday (22 November 2010), some of the most vocal advocates of openness have determined that FRAND is not a foe of "free" and that both concepts are legitimate in their own right.

I don't just mean the actual business practices of the companies behind the royalty-free/restriction-free lobby. Meanwhile we can even hear it straight from the horse's mouth:

Oracle provides TCK [the official testing kit for compliance with the Java standard] licenses under fair, reasonable, and non-discriminatory terms consistent with its obligations under the JSPA [the Java standard-setting agreement].

Don Deutsch, [Oracle Corp.] Vice President of Standards and Architecture

The above quote is from a statement with which Oracle -- a driving force behind ECIS and OpenForum Europe and an "open standards" lobbying partner of the FSFE -- just responded to the Apache Software Foundation (ASF), one of the two or three most important open source organizations in the world.

The context: compatibility of FRAND with FOSS; field-of-use restrictions

The statement quoted above speaks for itself, but the context makes it even more relevant: the board of the ASF had complained that Oracle "imposes additional terms and conditions [on Java licensees] that are not compatible with open source or Free software licenses." The ASF contends that "Oracle is violating their contractual obligation as set forth under the rules of the JCP [the Java standard-setting process]". It reiterated this view in a succinct reply to Oracle's FRAND statement: "The ball is in your court. Honor the agreement."

The open letter to which the word "agreement" points is more than three years old. At the time, Java belonged to Sun; Oracle acquired Sun last January. Therefore, the letter was directed to Sun, and it stated the following:

[...] The JCK license Sun is offering imposes IP rights restrictions through limits on the "field of use" available to users of our software.

These restrictions are totally unacceptable to us. As I explain below, these restrictions are contrary to the terms of the Java Specification Participation Agreement (JSPA) - the governing rules of the JCP [Java standard-setting process] - to which Sun is contractually bound to comply as a signatory. The ASF has a proud history of support for open software ecosystems in which commercial software can flourish.

However, Sun's JCK license protects portions of Sun's commercial Java business at the expense of ASF's open software. It prevents our users from using Apache software in certain fields of use.

[...] limitations on field of use for our users is contrary to the basic principles of open source licensing, and therefore these limitations would prevent distribution under any open source license, including our own. [...]

There you have the argument I addressed in this recent post ("FOSS can implemented patented standards"). The FSFE also listed the Apache license among FOSS licenses that it falsely claims to be incompatible with FRAND-based licensing. (You can read the truth about patent licensing under the Apache license here).

Oracle contradicts itself and its lobbying fronts ECIS, OpenForum Europe and FSFE

It appears to be a common pattern that open source foundations don't want to become "frandations", so they claim legal incompatibility even though their problem is a philosophical one. But what's really interesting is that Oracle uses two contradictory definitions of "open standards":

In front of policy-makers, Oracle (interestingly, Don Deutsch himself) gives talks such as this recent one in Brussels about all the good that open standards do. Oracle co-authors, finances and lends its name to statements such as this one that argue against FRAND-based licensing because it would allegedly "exclude a broad segment of the industry -- mostly open source software developers -- from implementing that specification in their products." ECIS, which issued that one, is run by Thomas Vinje, Oracle's outside counsel on EU antitrust matters. Oracle is one of its most influential members. Similarly, Oracle is a member of the OpenForum Europe (where Don Deutsch gave that talk), and it uses the FSFE for its purposes.

But when one of the most important open source organizations tells Oracle that FRAND terms "are not compatible with open source or Free software licenses", the answer is just that all Apache will get is FRAND -- take it or leave it. The Register also interprets it as saying "this is Oracle's stance on the matter and it's not changing."

As Apache's aforementioned succinct reply shows, the organization insists that this attitude constitutes a breach of the Java standard-setting agreement and, most of all, its section 5.C.III. You can find that agreement here, and this is a wording the ASF interprets in its favor:/p>

"[...] the [licensor] agrees not to impose any contractual condition or covenant that would limit or restrict the right of any licensee to create or distribute such Independent Implementations."

Apparently the ASF believes that this means its open source implementation of Java (the Apache Harmony project) must not be restricted. However, as I've already shown, Oracle says that FRAND licensing is "consistent with its obligations under the [standard-setting agreement]."

IBM -- the largest member of ECIS and OFE and financier of the FSF/FSFE -- supports Oracle

So Oracle believes FOSS can be reconciled with FRAND. And guess where IBM -- the other large company supporting the same European lobbying entities (ECIS, OFE and FSFE) -- stands: firmly on Oracle's side.

For a long time Big Blue supported not only the ASF in general but also its Java implementation, Harmony, in particular. Last month, however, it defected and now supports Oracle's OpenJDK, a GPL-based Java implementation.

IBM's open source VP Bob Sutor wrote on his blog that this switch of allegiance "will help unify open source Java efforts" and that "customers will benefit by having first class Java open standards developed collaboratively and constructively".

The Guardian's technology blog, however, calls this move "as much divisive as unifying."

The decisions and positions taken by those companies completely undermine the efforts of their "open standards" lobbyists in Europe such as IBM's Jochen Friedrich, who advocates extreme positions, and Oracle's Trond Undheim, who (as I mentioned in a recent post) referred to a group of EU officials as "rats" transmitting the "RAND disease" (RAND is synonymous with FRAND).

Upcoming Brussels conference on intellectual property rights and standardization

This endorsement of FRAND's compatibility with open source by Oracle and IBM has interesting implications to the debate taking place in Europe over open standards. Like I said at the beginning, the European Commission and the European Patent Office are going to host next week a conference on "Tensions between intellectual property rights and standardisation: reasons and remedies".

Some of the players I mentioned will speak there. Oracle's friend of FRAND, Don Deutsch, is on a panel on "ex-ante commitments to licensing terms". Thomas Vinje, counsel to Oracle as well as ECIS, will talk about the "certainty of availability and continuity of essential IP rights for licensing". An IBM patent attorney, Nicolas Schifano, is also on that panel. Finally, the FSFE's Karsten Gerloff will give a speech on "open source, freely available software and standardization".

I believe Don Deutsch won't be able to avoid discussing the Java situation. That one is important in and of itself, but it also plays a role in the patent infringement suit Oracle filed against Google and which draws more attention right now than any other patent suit in the industry (although there are so many going on, especially concerning mobile devices).

In its answer to Oracle's complaint, Google references the ASF's criticism and Oracle's obligations under the Java standard-setting agreement to allow independent implementations. Oracle claims seven of its Java patents are infringed by Dalvik, the virtual machine for Google's Android mobile operating system. Dalvik is derived from a part of the Apache Harmony code. So Oracle's denial of a different license has ramifications way beyond the ASF's desire.

The field-of-use restriction Apache complains about relates to mobile devices. Oracle allows independent Java implementations, but it draws a line in the sand where smartphones and tablets are concerned, for commercial reasons. This is a perfect example of a restriction that has nothing to do with royalties. Those who oppose FRAND often try to narrow the issue down to license fees while I advocate a more comprehensive approach.

Political debate and mixed-source reality

The conference "is part of an open dialogue process that the Commission is undertaking with key stakeholders [...] The EPO and the Commission hope to organise further events and meetings on issues relevant to IPR and ICT standards, with the objective of improving transparency and predictability in this crucial field."

While several of the speakers have a propensity for politicizing this topic, the conference program also lists panelists who operate on a day-to-day basis in the mixed-source reality of the IT industry. For instance, SAP is a markedly Linux-friendly vendor of proprietary software and should be able to tell us about how to cross the span. I also see major telecommunications equipment companies on the list. Nokia co-founded the MeeGo project, an open source mobile operating system based on Linux. But Nokia instigated patent litigation against Apple and is having an argument over FRAND licensing. This seems to me a perfect example of a company that cares about open source as well as its intellectual property.

In terms of an additional type of stakeholder that would have been great to have on one of the panels, I wish there were some SMEs (rather than just an association claiming to speak on their behalf) with mixed-source expertise and a complete focus on meeting customer needs by combining the best of both worlds. There are many of them out there.

That said, I really look forward to attending the conference next week and will report on it here (in compliance with applicable house rules, of course).

If you'd like to be updated on patent issues affecting free software and open source, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents.

IBM discredits the open standards lobby

On Thursday and Friday of last week, I saw hypocrisy of the worst kind: two IBM vice presidents preaching open standards values to EU decision-makers and FOSS community members instead of practicing them at their own company, which would really need that kind of lecturing.

IBM advocates openness and interoperability but actually refuses to be interoperable with open source only to keep customers locked in.

The first thing those guys should learn is that interoperability, by definition, always goes both ways. This isn't only a problem for them and the customers they're milking and the companies they don't allow to provide interoperability. As the EU is finalizing and will later be implementing its European Interoperability Strategy and Framework, IBM's open hypocrisy harms the cause of open standards.

There's increasing awareness for the fact that IBM, the biggest and most vocal backer of open standards lobby organizations in Europe (OpenForum Europe and ECIS), does in its core business the exact opposite of what it demands from others in their core business.

Something like that can work for a while. But once the decision-makers find out, being exposed as a hypocrite is a surefire way to lose in politics.

Neelie Kroes (European Commission Vice President): we must practice what we preach

European Commission Vice President Neelie Kroes, formerly known for her relentless pursuit of the Microsoft and Intel competition cases and now the Digital Agenda commissioner, addressed the OpenForum Europe conference on Thursday and said:

I am still a big fan of open standards. I believe in openness, and I believe in practicing what one preaches.

Google's chief legal officer David Drummond and Oracle vice president Don Deutsch gave examples of how their companies contribute to the cause. In light of what they said and what Mrs. Kroes had said before, I put a question to both of them: Aren't you concerned that IBM's use of patents against open source (in order to prevent interoperability) creates a credibility problem for the lobbying effort in which IBM is your partner?

Google's chief lawyer: you won't see us use patents against open source

David Drummond pointed out that I can't hold them responsible for what IBM is doing, and I didn't mean to do that. But through OpenForum Europe they walk side by side with IBM on this particular issue. They meet politicians together. They publish joint position papers. So a credibility issue for one of them in this particular context jeopardizes their collective effort.

After clarifying that he didn't want to comment on a particular company, David Drummond said something great:

"Anyone using patents against open source is a bad idea -- you won't see us do it."

This is obviously not a formal and detailed promise of non-assertion. But for a quick statement on a conference panel, and coming from Google's chief legal officer (the lawyer who advised Google's founders back in 1998 when they started), this was clear enough. We need more such commitments from large companies. And we need them to adhere to that principle.

IBM vice presidents afraid to talk

Later that day, IBM's vice president of technology and strategy, Michael Karasick, gave a presentation on openness in the cloud (here's the conference program).

There's the Simple Cloud API supported by (among others) IBM and Microsoft, and he talked about that initiative as well as other examples of open interfaces for cloud computing. It was designed to be a pitch to EU decision-makers with a special focus on e-government. He defined the cloud as a different form of "consumption" and talked about virtualization and "fundamental platforms."

So I asked him why his list of open interfaces doesn't include the mainframe instruction set, given that governments around the globe use mainframes for very critical purposes. Otherwise, IBM would transform its current mainframe hardware monopoly into a new cloud-related monopoly.

He simply didn't want to answer. Instead, he chose to talk about an earlier question that had already been answered by another panelist. How weak.

Silence at LinuxTag

The following day I was in Berlin at LinuxTag, where I gave a talk on EU processes relevant to Linux (and Free and Open Source Software in general). Before my own session I went to a presentation by IBM vice president and standards advocate Rob Weir on "the mutual advantage of open source and open standards".

It was bewildering to see him list (on page 8 of his presentation) the four freedoms from the Free Software definition. Software can be "open" even with patents, but it certainly can't be free as in free speech (and hardly free as in free beer) with them. If IBM truly believed in those freedoms, it would have to oppose software patents instead of lobbying for them, piling up more of them than anybody else and using them in really bad ways.

Rob Weir talked a lot about the Open Document Format and warned against the dangers of a standard being controlled by a single owner. I told him that he was probably preaching to the choir at LinuxTag when it comes to open standards, but I asked whether it wouldn't be a good idea for him to do the same kind of advocacy inside his own company, given the way IBM fights against open source and interoperability only to preserve its mainframe monopoly. Like his colleague the day before, he didn't want to answer the question.

Still at LinuxTag, one of IBM's leading mainframe Linux experts, Karl-Heinz Strassemeyer, was asked by RadioTux (a web radio program on Linux and open source topics) about the Hercules open source mainframe emulator, and he said he doesn't comment on "legal affairs".

Instead of muzzling its employees, IBM should practice what it preaches. They won't do that just because I say so. But they should heed what the European Commission says.

If you'd like to be updated on patent issues affecting free software and open source, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents.

OpenForum Europe and ECIS want software patents for their members but oppose those of others

Before I report in more detail on the OpenForum Europe Summit 2010 that took place last Thursday in Brussels (and from which I went directly to LinuxTag 2010 in Berlin for a presentation on relevant EU processes), I'd like to provide an overview over the key players of those organizations and their intentions.

In a nutshell, they're in favor of their own software patents but don't want to pay royalties for those of others. Their story is that some other companies' patents are needed for free for interoperability's sake, instead of advocating the abolition of all software patents including their own.

They don't want to address the root cause of the problem because they want to have their cake and eat it, too. The only way in which this approach is open is that it's open hypocrisy.

Two organizations with a mutual nucleus of three companies

There are two organizations in Brussels, the de facto capital of the EU, using the noble cause of open standards and interoperability as a pretext for a rather different agenda: OpenForum Europe (OFE) and the European Committee for Interoperable Systems (ECIS).

OFE is a lobby organization that now plans to work more closely with academics to give itself a think tank image.
ECIS brings about antitrust complaints and also engages in lobbying.

OFE's current list of members includes IBM, Google, Oracle, Red Hat, and Deloitte.
ECIS' members are Adobe, Corel, IBM, Nokia, Opera, Oracle, RealNetworks, and Red Hat.

So there are three companies who are members of both organizations: IBM, Oracle, Red Hat. Here's their background concerning software patents:

• IBM is the world's largest patent holder and one of the most ruthless ones, and has recently been in the news for betraying its own "patent pledge" by infringement assertions made against open-source startup TurboHercules.


• Oracle took a great position a long time ago, opposing the patentability of software with very good points, but later (maybe due to the rise of FOSS) determined that they are now useful to them.


• Red Hat lobbied to keep a proposed EU software patent law alive when we had already formed a majority for its rejection, and partners with IBM on a number of initiatives that appear to protect FOSS but are either ineffectual or even potentially harmful. Red Hat's CEO recently made a negative statement about software patents but they aren't doing anything to push for abolition. Last year there was some debate over Red Hat's related actions and positions. TechRights has multiple articles on this (1, 2, 3, 4, and 5).

Discontent with European Commission's Digital Agenda and draft European Interoperability Strategy/Framework

Right now they're unhappy that the EU isn't inclined to become their gofer. ECIS admits it without mincing words while OFE tries to spin-doctor around and play European Commission Vice President Neelie Kroes off against her colleagues. Others have previously attempted to play commissioners off against each other, and failed.

In her speech at the OpenForum Europe conference on Thursday, Mrs. Kroes certainly expressed a personal preference for royalty-free standards. OFE claimed to welcome that in its usual hypocritical way. I like that preference for fundamental reasons: I would prefer to see software patents abolished, which would take care of the royalty problem. Broadbased support for the Defensive Patent License could also have that effect.

But no matter how much we want that, it's simply not accurate to claim or imply that Mrs. Kroes supports the OFE's demands. Politicians have lots of preferences for what companies should do. What truly matters is whether or not they use their decision-making power to impose conditions. OFE would like the European Commission to use its European Interoperability Strategy and Framework -- a set of procurement guidelines with political implications going well beyond -- to require all vendors who want to do business with the public sector to adhere to rules that would force them to make licenses to their interoperability-related patents available on a royalty-free basis. And that's what the Commission is absolutely not inclined to do. Instead, the Commission fully accepts the notion of patent-encumbered standards.

In his summary of the OFE event, the OFE's chief executive Graham Taylor wrote: "We could not have hoped for a better keynote speech from European Commission Vice President Neelie Kroes." But in her speech, Mrs. Kroes actually said: "I have nothing against intellectual property being brought to the standard-setting table, but it must be disclosed." She also talked about a new legislative initiative on interoperability that "would likely involve some form of pricing constraints." That's another way of recognizing that there will be royalties. If something is royalty-free, you don't need to talk about "pricing constraints."

So the OFE's summary spins the commissioner's words in a way that I consider unreasonable. Then the OFE tries to drive a wedge between Mrs. Kroes and the other commissioners: "We at OFE believe there should be a complete re-write of the EIF. If this doesn't happen Vice President Kroes may not achieve the goals she is working so hard towards."

But what Mrs. Kroes said, especially when paying attention to the things I quoted from her speech, is simply consistent with the current draft of the European Interoperability Framework.

The Commission is a lot more consistent than the OFE.

OpenForum Europe lobbied European politicians for software patents, falsely claiming to speak on behalf of the open source community

When I first heard about OpenForum Europe six years ago, there was an extremely bad reason: OFE's chief executive (back then and still today), Graham Taylor, had lobbied European politicians to support software patents, and he falsely claimed that he represented the open source community.

That claim was refuted by Bruce Perens, the author of the original Open Source Definition, in an op-ed published by TheRegister. Bruce wrote that Graham Taylor "does not have the credentials to represent the Linux, Open Source and Free Software developer communities, especially when he contradicts our extremely strong opposition to software patenting."

The FFII knew that Graham acted as an astroturfer at the behest of IBM and other OFE members, but tried to engage in a facts-based dialog with Graham Taylor. The FFII wrote two open letters to him, and the introductory paragraph of an FFII page on OFE starts with the following strong and accurate statement:

Open Forum Europe is lobbying the European Parliament in the name of "open source companies" in order to make software directly patentable and to ensure that interoperable software may not be written.

Graham Taylor is not the only OFE executive to have been criticized harshly by the FFII. Paul Meller, now research and communications director at the OFE, was probably disliked by the anti-software-patent movement more than any other Brussels-based reporter writing in English. The FFII accused him of repeatedly spreading "extreme misinformation", alleging "a large number of errors and lack of objectivity", and even less flattering things were said on certain mailing lists.

While I also found that his articles were in most cases overly sympathetic to the positions of those pushing for software patents, there are explanations. Some of his articles appeared in the New York Times, others on IDG's website, so with a view to professional audiences he felt he had an obligation to listen to both sides of the argument. Unfortunately, one of those sides (the pro-patent camp) made a very professional PR effort from the beginning whereas the FFII didn't always handle its communications very well (as its core activists admitted at the time). I had several facts-based, constructive conversations with Paul at different points in time. Now, of course, he's no longer independent.

ECIS: ten commandments for the EU but none for its own members

The key person driving all of ECIS' efforts as its lawyer and spokesman is Thomas Vinje, a partner at Clifford Chance. He was also invited to speak at the OFE Summit on Thursday and unlike the host organization, he made it very clear that the EU's Digital Agenda and the current draft of the European Interoperability Framework fell short of his expectations. He didn't speak there in the name of ECIS but again, he is ECIS whether or not he formally introduces himself on that basis.

Thomas Vinje went on to propose ten ways to modify the document in order to make it meet the objectives ECIS considers important. A single one would have been sufficient: abolish software patents. Given where ECIS' members stand, it's no surprise he won't say that.

But he certainly forgot to insert an extremely important commandment in front of all others: do as you say.

Why doesn't ECIS hold its own members to the same standards it advocates?

Either there isn't any code of ethics in place for its members or it isn't any good, because otherwise IBM would have to be excluded from the organization given that it promotes interoperability only in markets in which IBM stands to gain from it but refuses to be interoperable and open where it generates half of its profits.

The same could be said about OFE as well: if you're serious about openness and interoperability, you have to demand that IBM provide interoperability right here and now, or you have to exclude the black sheep of the family. IBM probably pays you (and ECIS) a lot in membership dues but that's only your problem. If you don't get your house in order, you can't expect political decision-makers to buy your claims.

If you'd like to be updated on patent issues affecting free software and open source, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents.