Saturday, July 31, 2010

IBM's patent holder rhetoric against open source innovation

I previously wrote that IBM's reaction to the European Commission's antitrust investigation boiled down to diversionary tactics (some of them implausible, all of them irrelevant) and the usual rhetoric of patent holders. Now I'll comment a little more on the latter part, which is of a kind that's been annoying me ever since I became involved with patent policy years ago.

It's a problem that isn't unique to IBM: every time someone sues someone over a patent, you can hear similar things. But no other company has used that kind of rhetoric in such an obnoxious and hostile manner against free and open source software (at least not recently).

Likening an alleged patent infringement to piracy

This is how IBM's reaction to the announcement of TurboHercules's EU antitrust complaint in March 2010 began:
TurboHercules is an "emulation" company that seeks a free ride on IBM's massive investments in the mainframe by marketing systems that attempt to mimic the functionality of IBM mainframes.
Hercules is an independently-developed, 11-year-old open source project. If that amounts to free riding, then all of FOSS does in one way or another. Linux, OpenOffice, MySQL... you name it.
This is not really any different from those who seek to market cheap knock-offs of brand-name clothing or apparel.
That's just so wrong. What IBM describes there is called "trademark piracy" (although the use of the term "piracy" is debatable in that context). Someone who intentionally makes illegal use of a trademark such as Armani or Boss knows exactly what he's doing: infringing on someone's rights. But if software developers infringe a patent, it's inadvertent most of the time. Someone writes code and someone else obtained a patent that reads on it. No wrongdoing. That's why I commented negatively on the inclusion of patents in the Anti-Counterfeiting Trade Agreement (ACTA).

An Armani or Boss pirate will also try to fool customers, at least the credulous ones, with respect to the origin and quality of the product. TurboHercules doesn't do that. It clearly says that its software is the Hercules open source mainframe emulator, and every single person who downloads and installs the program knows that it's software -- not a mainframe. You can tell by the weight :-)
TurboHercules is a member of organizations founded and funded by IBM competitors such as Microsoft to attack the mainframe.
On the Microsoft conspiracy theory I commented in the previous posting. The notion that TurboHercules aims "to attack the mainframe" is so absurd. Those guys are fans of the mainframe, and by making mainframe software run on other platforms they certainly don't harm the mainframe.

TheRegister's mainframe expert Timothy Prickett Morgan pointed out that the mainframe ecosystem stands to benefit from TurboHercules, calling it "perfectly understandable to want the Hercules emulator to be available as a true alternative to IBM's mainframe iron running its mainframe software, and a perfect fool as well as a genius could readily see that having such an alternative would be a good thing for mainframe shops."

Innovation requires incentives for innovators as well as a functioning competitive environment

Still commenting on IBM's reaction to TurboHercules's March 2010 antitrust complaint:
Such an anti-trust accusation is not being driven by the interests of consumers and mainframe customers - who benefit from intellectual property laws and the innovation that they foster - but rather by entities that seek to use governmental intervention to advance their own commercial interests.
IBM, you can be sure that the European Commission is "driven by the interests of consumers and mainframe customers", which is why the EU competition authority even opened a second probe of IBM's behavior at its own initiative. Then the part on "intellectual property laws and the innovation that they foster" ignores that innovation takes two things: an economic incentive (and I agree that IP often plays a key role in that) and, equally importantly, undistorted competition.

On the last part about "entities that seek to use governmental intervention to advance their own commercial interests", TheRegister's mainframe expert Timothy Prickett Morgan accurately noted: "There isn't enough time in the day to list all the times Big Blue has benefited from the intervention of local, state, and federal governments around the globe."

And the final part of the March 2010 statement:
IBM is fully entitled to enforce our intellectual property rights and protect the investments that we have made in our technologies.
This again suggests that IPRs are an absolute thing, detached from all other considerations. They're not. There can be limits under competition law.

Let me make this very clear: I don't downplay the relevance of IPRs as a factor that results in investment (of time, money and energy). I've personally lived off IP for many years. I started writing articles for computer magazines when I was 15, computer books at age 16, then became involved with several commercial software projects (including three Blizzard games: Warcraft II, Diablo I, Starcraft I). I co-founded and managed a startup that depended on IPRs. MySQL was probably the most IPR-focused open source company, and I was involved with it as an adviser and shareholder. I defended some IPR-related strategic interests (broadcasting rights) of my favorite soccer club in an EU policy-making context. So I have a whole pro-IP biography, but I also value undistorted competition.

Let me quote TheRegister's Timothy Prickett Morgan again:
"As anyone who has watched the engineering done by Amdahl/Fujitsu and Hitachi [makers of mainframe products who effectively left the market] in the long and strange mainframe market knows full well, it was these companies that often innovated ahead of Big Blue [...]"

Patent validity and actual infringement are doubtful

Now that the European Commission launched its two parallel investigations of IBM's conduct, IBM radicalizes its patent holder rhetoric:
The accusations made against IBM by Turbo Hercules and T3 are being driven by some of IBM's largest competitors -- led by Microsoft -- who want to further cement the dominance of Wintel servers by attempting to mimic aspects of IBM mainframes without making the substantial investments IBM has made and continues to make. In doing so, they are violating IBM's intellectual property rights.
Those conspiracy theories are a distraction, especially in the IP context: should there be an infringement (which is at least doubtful), it would also be committed by the 11-year-old Hercules open source project. The TurboHercules product is Hercules as far as the software is concerned (alternatively available for GNU/Linux or Windows).

The part about "violating IBM's intellectual property rights" is very aggressive. Who knows whether the patents IBM believes are infringed are even valid? I talked in 2006 to a company that's pretty big in the smartphone business and they said a patent law firm had told them there's about a 75% or higher chance that a European patent someone asserts against an alleged "infringer" isn't even valid. So many patents get thrown out due to prior art, lack of inventive step, incomplete disclosure, or for other reasons.

Even if IBM asserts patents that survive an effort to bust them, there's still the question of whether they're actually infringed. Hercules is an emulator, so it isn't a CPU "clone": it has completely different inner workings. It takes the CPU instruction set as input in order to perform functions, just like a JavaScript interpreter takes commands in that programming language as input. What Hercules does internally may very well be so fundamentally different from what IBM's System z CPU does that the patents obtained on one don't read on the other.

The Supreme Court of the United States made a very appropriate statement in its 1966 ruling on the Brenner vs. Manson case: "A patent is not a hunting license." This meant to say that a patent isn't supposed to monopolize the right to solve a problem: it's supposed to relate to one particular solution. IBM's patents, if even valid, may protect its CPU. But they certainly can't prevent others from solving the same task -- the interpretation of a machine language instruction -- with different means.

The developers of Hercules haven't commmitted any wrongdoing by developing their solution independently. The investigation is all about whether IBM harms competition. I believe that's the case, and that's why I'm glad the competition probe was launched.

In the course of it, IBM may even learn that a patent is not a bullying license.

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.

Wednesday, July 28, 2010

In the IBM antitrust case, Microsoft matters only as a precedent

In a reaction to Monday’s announcement of two parallel antitrust probes of IBM’s conduct in the mainframe market by the European Commission, IBM blamed it all on Microsoft and its “satellite proxies”.

IBM also referred to accusations “being driven by some of IBM's largest competitors -- led by Microsoft”. Other than diversionary tactics and some typical patent holder rhetoric, IBM had nothing special to say.

In fact, since IBM’s public response to the complaint TurboHercules lodged in March, it’s been the same kind of message again and again. Now that the European Commission has launched two formal investigations, IBM’s mantra is less credible than ever.

Regulators have to protect the public interest

There’s an important difference between litigation and regulation.
  • If you sue someone, the court will have to take up your case unless it’s outrageously absurd at first sight. Access to justice for everyone.

  • By contrast, regulatory authorities follow up on only a minority of the complaints they receive. They don’t have the resources to deal with everything, but they have more flexibility to reject. Quite often, the European Commission explains the rejection of a complaint simply with a “lack of Community interest.”
T3 Technologies filed its complaint last year, TurboHercules four months ago. There will have been a fair amount of research by the Commission, as well as back-and-forth correspondence and meetings between the regulator and IBM (knowing the Commission’s modus operandi). There must already be some strong indications of wrongdoing on the table. Monday’s announcement wouldn’t have happened otherwise.

I can’t see how Microsoft would be in the position to use the Commission for any purpose. It’s the record holder in terms of the total of fines levied by the Commission on a single company (1.7 billion euros). Microsoft was pursued by the Commission over three different issues (two of which formed part of the same case) in recent years.

The European Commission started one of the two investigations at its own initiative

We’re talking about two separate, parallel cases brought by the Commission against IBM. The Commission’s press release explains this:
Both cases are related to IBM's conduct on the market for mainframe computers. The first case follows complaints by emulator software vendors T3 and Turbo Hercules, and focuses on IBM's alleged tying of mainframe hardware to its mainframe operating system. The second is an investigation begun on the Commission's own initiative of IBM's alleged discriminatory behaviour towards competing suppliers of mainframe maintenance services.
So the first case -- the “tying” case -- resulted from two complaints (the one by T3 and the one by TurboHercules). But the second case -- the “maintenance” case -- relates to an issue that the Commission found out about without even getting a complaint from anyone.

Diversion is not a defense -- especially if it doesn’t make sense

At the heart of both cases is IBM’s suspected abuse of a dominant market position (actually it’s a monopoly).

Whatever IBM does in the mainframe market, if it’s against the law, it’s against the law and hurts consumers. IBM’s diversionary tactics don’t change the facts.

This isn’t the first time for IBM to face a mainframe antitrust issue. There’s a tradition of probes of that kind going back to the 1950’s and the famous Consent Decree. The European Commission reached a settlement with IBM in 1984. And by the way, the US Department of Justice launched a preliminary investigation last October. The DoJ is similarly unsuspicious of being in Microsoft’s pocket as the EC. It nearly broke up Microsoft into two or three pieces.

The Wall Street Journal makes some good points

The Wall Street Journal asked whether what’s good for the goose is good for the gander. It recalled that IBM ”was a principal antagonist of Microsoft during the software giant’s epic wars with Brussels.”

So IBM accuses Microsoft now of something that it actually did itself: supporting antitrust action against a competitor. But the more important thing is this:
Now it’s time to print off a few more copies of that [Microsoft] ruling. It’ll be used against IBM for sure.
The WSJ explains the parallels between both cases very well. Highly recommended reading.

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.

Monday, July 26, 2010

European Commission launches antitrust investigation against IBM

The European Commission today announced the launch of two formal investigations into IBM's practices in the mainframe business, following complaints lodged by T3 Technologies last year and French open source startup TurboHercules in March.

By coincidence, this announcement was made just a few days after IBM launched its new generation of mainframe computers, an event that shows mainframes are still big business and far from obsolescence. There are estimates that the mainframe business (including software) generates about half of IBM's corporate-wide profits. The mainframe software market has an estimated size of $25 billion, about twice the size of the software market for Linux.

The Commission appears concerned about the tying of IBM's mainframe hardware products to its dominant mainframe operating system, z/OS. This is reminiscent of the Commission's previous objection to the tying of the Media Player to the Microsoft Windows operating system and the "browser case" that was settled last year and resulted in a browser choice dialog box for Windows.

In early April, I published a threat letter with which IBM tried to intimidate French open source startup TurboHercules SAS, whose founder started the Hercules open source mainframe emulator in 1999, with 106 patents and 67 patent applications. If you're interested in the correspondence between TurboHercules and IBM -- two letters from each company --, please look up this page.

There is a possibility of the Commission also formally investigating the complaint brought forward by NEON Enterprise Software, on which I reported here. The other complaints were filed earlier, and there's always some back-and-forth correspondence between a complainant and a defendant after a complaint. That process must still be going on with respect to NEON's very recent complaint, but I wouldn't be surprised if in a few months the Commission also picked up that case. Then there would be three parallel EU cases related to IBM's mainframe practices in light of the suspected abuse of IBM's dominant market position (a de facto monopoly, actually).

Moreover, the US Department of Justice announced in October that it investigated IBM's mainframe practices. Since then, there hasn't been any further announcement by the DoJ. It will be interesting to see if the DoJ makes a further announcement in the weeks or months ahead.

The open source aspect of the TurboHercules complaint and IBM's use of patents are the reasons for which I recently learned a lot about the situation in the mainframe market. I'm convinced that customers are locked in and milked shamelessly by IBM, and I hope that the outcome of the process will result in more customer choice, including the possibility to use the Hercules open source emulator to run legacy mainframe applications on affordable Intel-based servers.

For some time, IBM has been lobbying the EU as a self-proclaimed advocate of open source and open standards. I can't see how this antitrust probe will enhance IBM's credibility in that context.

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.

Saturday, July 24, 2010

New Zealand software patents:
abolition isn't certain

In recent months, many proponents of software patents and those favoring their abolition have not only been anxiously awaiting the Bilski decision by the Supreme Court of the United States but also locked horns over New Zealand's patent reform bill.

New Zealand is more or less antipodal to where I live, and I don't have any contacts down under. That makes it harder to comment, but by now I believe to have gathered enough information to be comfortable with expressing my view of the situation there.

I've come to the conclusion that it's too early to tell what the outcome will be, and I'll explain the different directions in which things might go, and why.

Believe me: substantive patent law is a very complicated matter, and it takes a lot of specialized knowledge -- not just common sense -- to figure things out properly. Hartmut Pilch, the founder and former president of the FFII, is a true expert in that field. We worked together closely. Initially, my NoSoftwarePatents campaign was independent from the FFII but coordinated many activities with Hartmut's organization. Later, I handed the campaign website to the FFII. Hartmut and I know how resilient software patents are, and we concur that the abolition of software patents in New Zealand is far from certain at this stage. I quote Hartmut on the New Zealand situation further below. In order for his insightful comments to be understandable, I have to provide some background.

Jubilant abolitionists and a minister being a typical politician

On 15 July, the two New Zealand organizations leading the fight against software patents cheered an announcement by New Zealand commerce minister Simon Power that a patent reform bill containing a general exclusion of software patents should pass without further amendments:
  • The New Zealand Open Source Society (NZOSS) wrote:
    "Minister Announces No Software Patents"

  • The New Zealand Computer Society (NZCS) announced:
    "It's official: Software will be unpatentable in NZ"
But what did the commerce minister actually say? His announcement is typically political, leaving different doors open. Its headline: "Minister announces way forward for software patents"

If I saw only that headline, especially the word "for" in front of "software patents", I would expect the text below to be a ringing endorsement of software patents. That headline certainly doesn't announce a state of no software patents at all. Much to the contrary, the first paragraph says:
"Commerce Minister Simon Power has instructed the Intellectual Property Office of New Zealand (IPONZ) to develop guidelines to allow inventions that contain embedded software to be patented."
Again, that doesn't sound like the door is closed. IPONZ is New Zealand's patent office, which after the parliamentary process will get to work out the details of what is and what isn't patentable there.

The process so far

New Zealand is overhauling its patent law. The law presently in force is the Patents Act of 1953. Within the New Zealand Parliament, the Commerce Committee took the lead on the new patent bill, which was presented by the country's government in July 2008. The Commerce Committee discussed controversial issues, such as software patents, with stakeholders. On 2 April 2010, the Commerce Committee amended the government bill and presented its first-reading report (HTML, PDF).

The key passage related to software patents is Article 15 para. 3A:
(3A) A computer program is not a patentable invention.
For the proponents of software patents, this straightforward exclusion went too far. According to the software patent wiki, they convinced New Zealand commerce minister Simon Power that instead of that exclusion the law should be modeled after the European Patent Convention (EPC). The EPC, too, excludes computer programs from the scope of patentable subject matter, but only "as such". Some background on the way the European Patent Office and European courts interpret an exclusion of software as such is contained in this earlier blog posting.

A New Zealand blogger (who set up New Zealand's first government web server back in 1995) called the reactions of the pro-patent camp "disingenuous."

Temporarily the abolitionist camp was worried that the legislative proposal could still be changed. Therefore, it's understandable that NZOSS and NZCS celebrated the commerce minister's announcement that he and the parliamentary committee agreed to pass the bill unchanged. It seems that the opposition to software patents that two leading New Zealand software companies -- Orion Healthcare and Jade Corporation -- mounted played an important role.

So what's next down under?

The New Zealand Herald interpreted Minister Power as saying that "[g]uidelines rather than a law change will be used to allow inventions that contain embedded software to be patented."

In other words, the law will contain an exclusion of software patents, but the national patent office will draw up guidelines that will allow patents on inventions containing embedded software.

The law is broad and general. Many may believe that it would be sufficient to simply say "a computer program is not a patentable invention" in a law. Wrong. The biggest issue in substantive patent law is line-drawing. We as programmers may have a clear idea of what a "computer program" is, and what it is not. But when a patent application gets filed, the word "computer program" may not appear in it at all, even though computer programs could infringe a patent that would be granted on such an application. The patent claim (the scope of the patent) can be an "apparatus" or a "method". It doesn't have to say "computer program" at all.

Embedded software

So what is "embedded software" and inhowfar is it different from a computer program? That will now be the subject of much discussion in New Zealand, I guess. The commerce committee's bill says in its footnote 4:
Embedded software is computer software which plays an integral role in the electronics it is supplied with (e.g. cars, pacemakers, telephones, and washing machines).
Note the word "telephones"! Today's smartphones are pretty powerful and functional computers. The software that plays an integral role in running them isn't too different (in terms of operating systems, programming languages etc.) from the software running on a server or a desktop PC.

So the New Zealand patent office (IPONZ) will have to come up with guidelines that enable patent examiners to make a distinction between "computer program" and "embedded software", with the latter even including the software powering smartphones.

What does this mean in practice? Let me give an example: if someone filed a patent application for an "apparatus for the storage and retrieval of contact data", the patent office might grant the patent because this would be a typical smartphone application (every smartphone comes with an address book). But the patent claims could ultimately also read on an address book program for a personal computer.

Contributory infringement

Some may think it doesn't matter to programmers and software distributors because they can never infringe such a patent. Actually, they can: the legal theory is called "contributory" or "indirect" infringement. The idea is that the infringement as a whole requires hardware and software. But if the software plays an essential role in it, then publishing and selling such software would be a contributory (indirect) infringement of the aforementioned device patent. In practical terms, a contributory infringement comes with pretty much the same negative consequences as a direct one: the patent holder can seek an injunction and indemnification even against an indirect infringer.

So those guidelines are now critical. The law is the law, but it's not detailed enough for patent examiners to work with. The guidelines will determine the daily work of the patent office there. IPONZ will have to be careful in designing those guidelines because ultimately a court could determine that certain patents go against the law. The guidelines won't have the status of a law. A court will review the decisions taken by IPONZ (if applications or third parties appeal those decisions) and will then decide, case by case, whether the patent office interpreted the law correctly. Over time that would provide more clarity.

This means that it may actually take a number of years before it's clear whether or not New Zealand allows software patents. The IPONZ guidelines will certainly give an indication as to what that authority plans to do. Still it will remain to be seen exactly how the guidelines are applied, and what the courts ultimately say.

Possible results

I have to repeat it: that law isn't clear. Not even with the straightforward exclusion of software patents. It says that a computer program isn't a patentable invention. It doesn't say that methods that could be implemented in or by a computer program -- as well as in other forms, such as hardware -- aren't patentable. For the reasons I explained above, those are different things.

If you define a software patent as a patent on software (and only that), the law will probably take care of that problem. But a more pragmatic definition is that a software patent is a patent that can be infringed by software developers, publishers and users. The New Zealand bill doesn't take care of that question. This will depend on the guidelines and, ultimately, the courts.

How could they draw the line? The problem is that this can't be fine-tuned like the fishing quotas that two countries negotiate. It's a pretty polarized situation where there's either a far-reaching exclusion that abolishes large parts of the entire patent system, or you get, like in Europe, an exclusion on the surface that ultimately results in patents that read on software, even if those patent applications may be phrased in a way that doesn't suggest so to a layman.

Both approaches are possible outcomes when the objective is to distinguish "computer programs" from "embedded software":
  • The abolitionist approach would be to accept patents on devices containing embedded software only if the technical contribution of the invention relates to an advancement in an applied natural science. For an example, a car brake controlled by a computer may use an algorithm to computate the most efficient use of the available braking power based on road conditions. Under a restrictive regime concerning software patents, the device as a whole would be patented only if the computer-controlled car brake uses what experts call "controllable forces of nature" in an innovative way resulting in a shorter braking distance. A better memory management algorithm within the car brake's controlling software would not be considered patentable.

  • The liberal approach would also consider software running on a standard computer to be "embedded." In other words, as long as the software is outside the computer on some medium, it's considered software -- once you actually run it on a computer, the device as a whole could under certain circumstances be considered a technical invention containing embedded software. Also, this raises the question of whether operating system software is embedded software by definition (especially when considering that smartphone and PC operating systems are slightly different variants of general-purpose operating systems).
The jury is still out

At the start of this I mentioned Hartmut Pilch, the founder of the European anti-software-patent movement. When he saw a mutual friend of ours declaring (on his Facebook wall) victory based on the first reports from New Zealand, Hartmut wrote the following comments (which he authorized me to quote, with a clarification he later inserted):
"They will have to watch this very closely and we too. The backlash of the patent mass-producer lobby will be enormous and the opportunities for [IPONZ] and courts to misunderstand or otherwise dodge the lawmaker's instructions manifold.
One obvious way for the patent lawyers to twist the new law will be to interpret it as a call from the lawmaker to follow European i.e. EPO examples, with the added (false but effective) argument that TRIPs doesn't allow anything else."
TRIPs is the international Agreement on the Trade-Related Aspects of Intellectual Property Rights, and it requires patents to be "available for any inventions, whether products or processes, in all fields of technology, provided that they are new, involve an inventive step and are capable of industrial application."

As you can see, the devil is in the details. I'll continue to watch the process and will comment on it again sooner or later.

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.

Wednesday, July 21, 2010

OpenForum Europe: hypocrites lobby the EU but don't get their own houses in order

The worst thing that can happen to a good cause is to be used as a pretext by blatant hypocrites. The adoption of Free and Open Source Software by governments is definitely a good cause. So is interoperability. But OpenForum Europe and its members -- IBM, Google, Oracle and Red Hat -- should get their own houses in order rather than pressure European politicians and spread dishonest propaganda.

Monday's edition of the New York Times just reported on the tireless efforts of that gang to lobby the EU over a set of guidelines called the European Interoperability Framework (EIF).

They claim that it's all about open source and interoperability. In reality, OpenForum Europe and its members simply pursue their commercial interests. They call themselves "open", but actually they aren't. They demand interoperability when others should open up. They deny it when their own interests are at stake.

Let's go over that bunch of hypocrites one by one.

OpenForum Europe

A notorious fake representative of open source for many years. Its chief executive lobbied European politicians for software patents, falsely claiming to speak on behalf of the open source community while actually just serving his master: IBM. Bruce Perens, the author of the Open Source Definition, wrote an op-ed for The Register to protest against such conduct.

IBM: International Bullying Machines

The biggest patent bully preaches interoperability but practices the exact opposite in its core business.

The mainframe business still generates about half of IBM's profits because customers are locked in and squeezed out. Innovative solutions that provide interoperability, such as the open-source Hercules mainframe emulator, could loosen IBM's stranglehold on the market. So IBM goes for their throat, particularly by using patent warfare.

IBM's aggression against TurboHercules, a French open-source company started by the founder of the Hercules project, is an attack on the very concept of interoperability.

The debate on which the New York Times reported is largely about whether patents related to interoperability should be licensed on a royalty-free or a fair, reasonable and non-discriminatory ("FRAND") basis.

IBM and its OpenForum Europe allies claim that FRAND, which ensures that overcharging can't occur, isn't good enough. They say it has to be royalty-free. But on the mainframe side, IBM doesn't even offer FRAND. Let alone royalty-free. They don't offer anything. They want to shut out competition altogether, which is the most harmful way to use patents.

That discrepancy between using patents as a weapon of total destruction and saying other companies' patents should be made available on a royalty-free basis is inexplicable. If they preach royalty-free, they should offer it. Or they should preach and practice FRAND. Either way they'd be consistent. But demanding one extreme and pursuing the other is hypocrisy at its worst.

When asked about this contradiction, IBM executives refuse to answer.


Last time I checked, all of Oracle's money-making products were closed-source. Until that changes, I can't see how such a company can credibly advocate open source interests in Brussels, or elsewhere.

Oracle acquired several open source technologies as part of Sun Microsystems. The deal was closed in late January, and the open source community is upset about Oracle's stewardship of several of those projects. ZDNet's open source blog summarized the situation concerning Java, OpenSolaris and OpenOffice, concluding that "if open source is all about ending vendor lock-in, Larry Ellison is its worst nightmare. And since acquiring its crown jewels, I would argue, that nightmare has slowly come true."

Just last week, the OpenSolaris board launched what CNET calls an "OpenRevolt against Oracle."

The most appalling example of Oracle's hypocrisy about interoperability is this: Sun used to provide a free-of-charge tool to open ODF (Open Document Format) files with Microsoft Office. Oracle decided that this kind of interoperability tool should be monetized. It now costs $90. It wasn't open source before, but it was available for free, and if Oracle is serious about promoting royalty-free standards such as the ODF, then it should encourage the widespread use of such a tool.

They claim interoperability must be royalty-free. But they put it behind a paywall.


Google isn't a software vendor, at least not in the traditional sense of the word. To the extent they make software available to the general public, they do so on open source terms.

But how open is Google where it generates the bulk of its revenues -- meaning its search engine? Not so much, it seems. Open source blogger Dana Blankenhorn made an interesting proposal on ZDNet: an open standard for search engines. He considered this a compromise proposal in light of Google's fight to keep search a secret.

Google should address the "open standards" issue in connection with Internet search before lobbying the EU. And while at it, Google might as well ask itself if its fight against Scroogle, an independent not-for-profit website that delivers Google search results while protecting the privacy of users, fits in with its lobbying for open standards and open interfaces.

Red Hat

Red Hat is much smaller than its OpenForum Europe allies. IBM is the key driving force, and Oracle and Google are much more powerful. Compared to them, Red Hat is just another "hanger-on" that will follow IBM anywhere.

Red Hat supports all of IBM's patent initiatives, including the Open Invention Network, which is the opposite of "open". And a Red Hat manager who spends a large part of his time on EU lobbying defended (on Twitter) IBM's aggression against TurboHercules and interoperability.

OpenForum Europe and its members should support the EU's more important interoperability initiative

If OpenForum Europe and its members really cared about interoperability, there would actually be a much bigger opportunity to make headway for the cause. The European Commission is preparing an initiative to ensure that all "significant market players" will open up their products and services. So why don't OpenForum Europe, IBM, Oracle, Google and Red Hat come out loud in support of that plan? Why do they waste their time and that of many other people on a set of public procurement guidelines when there's actually an opportunity for something with really wide-ranging and highly positive effects?

The answer is simple: they aren't sincere about interoperability. That's a fact they prove every day.

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.

Monday, July 19, 2010

Patent expiration benefits Free and Open Source Software

FreeType, a FOSS project developing a font rendering engine, cheers on its webpage the expiration of three Apple patents related to the rendering of TrueType fonts.

While those patents were valid, the project disabled one technical component called the "bytecode interpreter" and called on its users not to use it even though the program code was in place. Now that functionality is enabled by default. Yesterday FreeType just released its latest version, 2.4.1.

Two of the three Apple patents in question were filed for in May 1989. The third one was applied for in 1992, and I presume Apple decided not to renew it because there wasn't much (if any) commercial value left after the other two TrueType patents were past the end of the 20-year maximum term of validity of US software patents.

[Update] Through Slashdot I became aware of a juxtaposition of two screenshots that shows what difference in quality the "bytecode" algorithm makes. It's pretty visible. [/Update]

It was interesting to read this because two days ago I commented on Richard Stallman's concerns over patents that could affect Mono and DotGNU and in that context I said that if RMS wants a totally patent-unencumbered development platform, it will have to be more than 20 years old. Plus it would have had to be open-sourced back then because without publication something cannot serve as prior art (meaning as a previous invention in order to invalidate patents filed later). Even the slightest change to a code base during that time span could result in some new patent infringement. But the part that's more than 20 years old is (provided that it was published back then) safe.

The FreeType example shows that sometimes useful things do become available for free thanks to patent expiration. That doesn't solve all problems, but every useful thing is potentially good news for someone. In this case, for the FreeType folks, who probably had been awaiting that moment for many years.

Several years ago, the nastiest patent related to the Graphics Interchange Format (GIF) also expired. Unisys had acted pretty much like a patent troll because its core business was in bad shape and the GIF patent became a key asset to the once-great company.

That GIF patent is the only important software patent expiration I had been aware of before reading about TrueType.

In political debates over whether or not software should be patentable, I criticized the 20-year term as being way too long for software. In a speech I gave at a demonstration against software patents in Munich in April 2004, I said that the patents reaching the end of the 20 years at that time were basically from the heyday of the Commodore 64. That was about right, although the IBM PC was already a few years old in 1984, and the Commodore Amiga was launched a year later. The TrueType stuff is now at least from the era of the breakthrough of Windows.

It will be interesting to see which FOSS projects will benefit from the expiration of the next relevant software patents.

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.

Saturday, July 17, 2010

Richard Stallman's Mono and DotGNU patent concerns

Glyn Moody, a true FOSS expert among journalists and author of the OpenDotDotDot blog, has interviewed Richard Stallman, the founder of the software freedom movement, by email to discuss RMS's concerns over software patents in connection with free implementations of the .NET programming interface (Mono and DotGNU).

Glyn's article was published by Computerworld UK.

To sum up the key points, RMS stated the following in the interview:
  1. Due to patents held by Microsoft on its .NET technology, the availability of two .NET implementations (Mono and DotGNU) under free software licenses doesn't mean that free software developers should, according to RMS, write code for the platform. His call on the free software community: "You shouldn't write software to use .NET. No exceptions." RMS says that Microsoft could one day use patents against free .NET implementations.

  2. RMS thinks that the C# ("C Sharp") programming language should also be avoided. He makes a distinction between that one and .NET because C# was "standardized by a standards committee" and Microsoft made "a stronger commitment" concerning patents than for alternative implementations of other elements of .NET.

  3. As a requirement for RMS to encourage the development of free software on free implementations of .NET, Micosoft would have to "make an ironclad commitment that its present and future patents will never be used against implementations of DotNET."
The exchange between Glyn and Richard has raised important questions and resulted in interesting answers. I agree with RMS on the incompatibility of software patents with the notion of free software and on many other patent issues, but I don't think the advice he gives to the developer community makes any sense at all in this particular case.

RMS's utopian advice runs counter to commercial logic and fails to advance the cause of software freedom

Adopting his advice would be an utterly stupid decision for any developer from a business point of view, which is actually normal because RMS's agenda is all about software freedom and not at all about commercial success. But even from a non-commercial free software point of view I think this kind of advice doesn't make sense in the world in which we actually live.

The only alternatives in terms of programming languages and platforms that could perhaps be supported under RMS's premises would have had to be open-sourced in exactly the same form more than 20 years ago (without even the smallest modification made ever since) and then the new software one writes on top of it today wouldn't be guaranteed to be truly free software for another 20 years. 20 years is the potential life expectancy of a software patent, and I'll explain the logic of this further below.

So unless someone wants to waste 20 years or even an entire professional life of 40 years just for the sake of an ideology, it's better to reject such utopian advice and take a realistic perspective on patent-related risks.

There's no particular reason not to develop software for .NET (as compared to any other platform on this planet), and free implementations of .NET such as Mono and DotGNU aren't really less free than a free Java application server or a PHP interpreter for Apache.

RMS focuses on the lesser risk and ignores the greater one

The fundamental mistake made by RMS in the aforementioned interview is that he narrows the whole patent-related risk down to only one company (Microsoft), which actually has a stronger commercial interest than any other in the world to make the .NET platform popular and to ensure developers succeed with the applications they build on top of it. And we all know how much competition there is between platform companies for the hearts and minds of developers. More importantly, Richard completely ignores the fact that hostilities against Mono and DotGNU could also come from other patent holders.

Even if one sides with RMS concerning what the greater risk is and rejects my business logic for the platform company itself being most likely to want the best for its application developers, no one can reasonably deny that there's a huge number of patent holders other than Microsoft whom RMS fails to take into consideration. That mistake is sufficient all by itself -- regardless of how to assess the different risks -- to prove RMS's advice concerning C#, .NET, Mono and DotGNU wrong in the sense that there are also patent risks concerning any other programming language or platform out there.

Yes, free software is incompatible with patents, but software patents exist on pretty much every kind of software technology and therefore I don't think one can make the case that free .NET implementations or software written to run on them is inherently less free than other software available under the same licenses.

Time heals the wounded and invalidates patents

In the patent minefield that exists, there's no such thing as a reliably patent-unencumbered programming language or API (application programming interface) except for a hypothetical scenario of no practical relevance. That scenario is one in which all patents that may read on the platform have either expired or can be easily invalidated.

Since software patents (at least in the jurisdictions I know) have a maximum term of validity of 20 years and patented ideas must be new by the time of the application (or they can be invalidated later on the basis of "prior art"), one can argue that if software was published (as open source) more than 20 years ago, all patents will either have expired or the published source code (which should be time-stamped to prove its vintage year) could be used as prior art to take down younger patents on the same technology.

Therefore, free software developers would have to use free platforms that are more than 20 years old (and were published as free software back then, not just later). The applications they write couldn't be guaranteed to be patent-unencumbered for another 20 years after the publication of their source code.

In other words, the price to be paid for a guarantee of being patent-unencumbered is to be decades behind the evolution of technology, and to be extremely patient relative to the duration of a human professional life.

The pragmatic alternative is to regard free software as a great idea and a wonderful vision, but to understand that patents make all software potentially non-free, regardless of whether the patents in question are held by Microsoft or anyone else.

The solution proposed by RMS (an "ironclad commitment") would be desirable but insufficient

At the end of the interview, RMS made the proposal I mentioned in item 3 of my summary of his position at the beginning of this posting. He said that Microsoft should make an "ironclad commitment" not to use current or future patents against free implementations of the .NET API.

I, for my part, would very much welcome such a commitment. But I disagree that it would make all the difference that RMS suggests it would make. It wouldn't solve the problem of third-party patents. Every other current or future software patent holder in the world would also have to make that promise in order for RMS's vision to materialize.

The second part also applies to platforms for which the original developer makes an "ironclad" patent promise. Even a free programming language like PHP can infringe and almost certainly will infringe on some third-party patents out there, unless you take a programming platform that was open-sourced more than 20 years back (as I explained further above).

So I strongly recommend to focus on how patent holders actually use their rights. In that respect, I will comment on Microsoft in greater detail in some other posting, but I can already say at this stage that there simply isn't any evidence of Microsoft using patents in a way that would drive companies out of business or jeopardize the existence of FOSS projects.

Don't cut off your nose to spite your face

RMS also refers to Eben Moglen's assessment that Microsoft's Open Specification Promise "is not something we can rely on." I can see why Richard and Eben say so. But I can also see reasons for which one could say the same about (to name but a few examples) Red Hat's patent policy, the promises Oracle made concerning the acquisition of MySQL (without wanting to comment on what Oracle is doing now), Google's vague assurances concerning WebM, the Open Invention Network's arbitrary scope of protection, or IBM's broken patent pledge.

Concerning IBM's pledge, I remember that RMS also commented on it unfavorably back in 2005 when it was made (not as aggressively as I did, but it was clear that Richard also rejected that approach). So he's aware of the fact that vendors don't make those public commitments in an "ironclad" form. That's a general problem and it's not particular to .NET, C#, Mono and DotGNU. Nor are the other concerns voiced by RMS specific to those technologies.

That's why I think a decision to write software for .NET, or to implement .NET interfaces in free software, isn't a statement against freedom any more than using any other current platform: Java, PHP, you name it. But acting in accordance with RMS's advice would be a self-imposed restriction of freedom, for no good reason.

With the greatest respect (which he deserves), he sometimes proposes to cut off one's nose to spite one's face.

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.

Saturday, July 10, 2010

The silver lining in the Bilski decision isn't where most people believe

About two weeks ago the Supreme Court of the United States (SCOTUS) handed down its opinion in re Bilski, a business method patent case. The patent application was rejected, but in a way that didn't draw any kind of line that would affect patents on software technology.

I commented on it within about an hour of its publication, concluding that the decision didn't invalidate even one software patent (the Bilski application itself wasn't a software patent application) and that only a decision to grant a patent on the Bilski application could have been any less restrictive. On the following day I listed the top ten losers.

Meanwhile discussion has continued and I've read a number of other opinions. Some of those were very realistic, such as Steven Vaughan-Nichols's analysis. Others took a more optimistic perspective and argued that the narrow scope of the ruling left the door open to more restrictive decisions in the future.

As the saying goes, every cloud has a silver lining. So where is it in the Supreme Court's Bilski opinion? There is one, but it's not where others seem to think it is. I'll start with where I believe many others are on the wrong track.

The "abstract idea" approach is a losing strategy

The conclusion from the Bilski ruling that patents on software technologies might one day be invalidated on the basis of being abstract ideas -- which is how the non-software Bilski application got rejected by the SCOTUS -- is
  • a gross misinterpretation of the ruling, blatantly ignoring the court's unambiguous endorsement of patents on software technologies,

  • an ideological argument that bears no legal or political weight with a majority of reasonable decision-makers,

  • and, therefore, destined to remain unproductive at best and counterproductive at worst.
I know that some people who subscribe to that ideology won't like to read what I just wrote. But I'm not writing this for the ones who'd rather lose for the sake of an ideology. I'm writing this for those who understand that one can very well adhere to an ideology and simultaneously recognize that it takes something other than ideology to win.

Ideological blindness is the number one reason to which I attribute the fact that software patent abolitionism hasn't made any real headway (other than some defensive success).

There are many different angles from which one can come to the conclusion that software should be a largely or entirely patent-free field. Often when I talk to people who have that belief, it turns out that each person believes his reasoning for why software patents are undesirable is the truth and the winning argument. There are activists who think like it; there are also executives of smaller companies whose narrow perspective prevents them from recognizing that politics bears some -- but only limited -- resemblance with marketing.

Let's better face this fact: there isn't a single killer argument against software patents that will convince a non-programmer if that same counterpart has also heard the pro-patent argument. If you can ever convince a majority of decision-makers, you'll have to do it indirectly. The direct approach has been tried by many people for many years -- to no avail (except, as I mentioned before, in a defensive situation).

The Bilski case was likened to two past cases and deemed different from a third past case

A lot of FOSS advocates basically argue that since the SCOTUS didn't explicitly say that software must be patentable, there's always a chance to go back with another case. That's just wrong. It's a typical exhortation to hold out (or, more precisely, to cling to a flawed strategy).

Don't let others fool you just because they don't want to adjust to reality. Here's a non-legalese explanation of what the SCOTUS really said.

The SCOTUS clearly stated that it did not want to issue a wide-ranging ruling with unintended consequences on other areas of patentable subject matter than the Bilski type of non-software business methods. And the SCOTUS determined that it wasn't really forced to overshoot: there already were precedents for similar concepts that were found unpatentable on the grounds of representing "abstract ideas."

The dreamers who think that software patents could be abolished on that same basis base their hopes on the fact that the SCOTUS didn't specify a set of rules that would define what an unpatentable "abstract idea" is. Experts would say: the court didn't establish a legal test (or a set of legal tests) that can be used to make that determination.

But the SCOTUS gave a couple of examples, and in the usual case-law style, those are cases put before it in the past. The court found that -- without even attempting to put it onto an objective basis -- the Bilski application was of a very similar nature as the ideas held unpatentable in two past cases (Benson and Flook). The court furthermore determined that the Bilski application didn't have enough in common with the patent considered valid in the Diehr case.

Without digressing into the details of those cases, let me just say that Benson and Flook related to general ideas without a very specific application and implementation. In my opinion, the Diehr patent shouldn't have been granted either, but there's no denying the fact that it was much more specifically tied to a technical purpose than Benson and Flook -- and than Bilski, of course.

There was a lot of disappointment among patent abolitionists that the SCOTUS didn't seize the opportunitay presented by the Bilski case to do some more specific line-drawing. While no one wanted to insult the court directly, the criticism suggested a lack of courage. I don't think that's fair. I believe the SCOTUS was right to find that the Bilski case per se presented nothing that hadn't been answered by it before. It was more of the same, and that's why it was the waste of time and money that the Software Freedom Law Center said it became. The case just wasn't suitable to what some people -- such as the SFLC -- would have liked to achieve. So don't blame the court.

The SCOTUS didn't draw a clear line but gave plenty of hints

Obviously a ruling based exclusively on similarities to past cases (without elaborating on inhowfar there were common elements) is less clear than a set of rules. The court almost implied that "if something is an abstract idea, we'll see it anyway." In the meantime, people should just look at the examples and draw inferences from those.

But the SCOTUS made some clear statements in its reasoning as far as software patents are concerned. Note that in the following I'm referring to some passages of the reasoning that were written by Justice Kennedy, who also presented the majority opinion, but those particular passages were not supported by Justice Scalia.

On page 9 of the decision, a more restrictive approach was rejected because it "would create uncertainty as to the patentability of software, advanced diagnostic medicine techniques, and inventions based on linear programming, data compression, and the manipulation of digital signals." While that is based on a reference to position papers (amicus briefs) submitted by pro-software-patent organizations, the way the SCOTUS refers to those concerns leaves no doubt that the justices who supported the passage agreed that the Bilski decision shouldn't cause collateral damage in those areas.

Now look at that list again: if even "data compression" should be patentable in principle, there's just no way that software would be considered too abstract an idea. Data compression is the kind of software patent that is closest to pure mathematics. One may argue -- and I personally believe -- that it is essentially pure mathematics and the argument of proponents of patentability that it's "applied mathematics" doesn't convince me at all. I can see "applied mathematics" in play if a car brake is computer-controlled to maximize its efficiency. I can't dismiss the idea that computer graphics can involve "applied mathematics" (I may not want patents on graphics algorithms for other reasons). But with "data compression" I just consider it incredible that some people would (and actually do) claim that those are "applied" as opposed to pure mathematics.

So if there's such a widespread belief that data compression should remain patentable in the Information Age (and that's what it does unless one wants to just interpret the ruling in completely unreasonable ways), then this suggests to me that the entirety of patents on software technologies is safely outside of whatever the SCOTUS would consider an "abstract idea."

The SCOTUS makes it very clear that as new technologies evolve, the patent system was intended (by the Founding Fathers) to expand accordingly, unless there's legislative intervention to restrict it. In this regard, a majority of the court also referred to "technologies for conducting a business more efficiently" (which I mentioned in connection with what Bilski means for

That's just one of several examples -- but in my opinion the best one -- of where the SCOTUS makes it clear that at least some business methods must be patentable.

So if even software-implemented business methods are patentable, there's just no way that future SCOTUS rulings would hold typical software patents to be "abstract ideas" and therefore unpatentable.

Ideologues will say that software is a product of authorship rather than of engineering. I understand some of the reasoning and I support it, but many critics of software patents are just unrealistic in terms of how they make that point. Claiming that software development is closer to composing music than to electrical engineering is crazy. I've been in the software industry for 25 years now and I've always referred to software as "technology" and to professional programmers as "software engineers", even though I can also see what programming has in common with writing. Having authored twelve computer books, I believe I can -- and I do -- appreciate that.

So programming has common elements with both engineering and authoring: that doesn't mean I can deny the engineering part of it just because I don't want to deal with patents in my field. I can have other reasons, but that one isn't a useful argument.

The idea that every software patent is just an abstract idea is an abstract idea in and of itself. And it won't get us nowhere.

The actual silver lining: the SCOTUS' remark on striking the balance

Radicals are always more receptive to fundamentalism than to an argument based on striking a reasonable balance. But the latter is what works best to convince rational decision-makers.

Near the top of page 10, the Bilski decision contains a wonderful passage that is infinitely more helpful with a view to the future than the whole "abstract idea" thing:
This Age puts the possibility of innovation in the hands of more people and raises new difficulties for the patent law. With ever more people trying to innovate and thus seeking patent protections for their inventions, the patent law faces a great challenge instriking the balance between protecting inventors and not granting monopolies over procedures that others would discover by independent, creative application of general principles.
This passage is another reason for which I think a lot of critics of the decision are biased. What I just quoted shows that the justices supporting that passage understood very well that there may be a problem with software patents. However, a majority of the SCOTUS didn't consider the Bilski case the right occasion on which to address it, and it may not even regard any future case as an opportunity to determine "where that balance ought to be struck."

The quoted passage basically says: In the past there was a much smaller number of people who came up with potentially patentable ideas as part of their work. There were a few scientists in laboratories (not literally, but that's roughly the idea). These days there are tens or hundreds of millions of people who have a computer at home or at work and know how to program it, and maybe the traditional approach taken under patent law doesn't work well in such a situation and results in too many patents and -- a highly important aspect -- too many incidents of inadvertent infringement through independent creation.

While that doesn't sum up all of the reasons for which I dislike software patents, it addresses the core part of it. I mentioned in other contexts that my fundamental problem with software patents is the risk of inadvertent infringement. With copyright, that risk exists in a theoretical form but not in a practical one. With patents, it's a serious issue, especially in the field of software.

I believe that the oppponents of software patents should focus on that part of the Bilski opinion and try to build a case on that basis. Maybe there shouldn't be just another legal case because the SCOTUS also stated on several occasions that the courts "should not read into the patent laws limitations and conditions which the legislature has not expressed." But at the very least the quoted passage from the decision gives some guidance in terms of how the case should be presented to lawmakers.

I know that many in this movement won't want to go down that avenue for the fear that the outcome would be some patent quality initiative as opposed to abolition. And if such a patent quality initiative didn't live up to expectations, it wouldn't change anything. I understand. I share the concern. But I don't see any other silver lining in the Bilski decision (as far as the majority position is concerned). The argument that the number of innovators is huge and that too many patents result in too much inadvertent infringement is one that non-programmers can understand. Unlike the "abstract idea" that won't ever have any material impact.

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.

Thursday, July 1, 2010

{Interoperability} Significant market players to face EU interoperability rules

Amid all the brouhaha over the European Commission's approach to open standards, there's been hardly any attention for an exciting initiative that could greatly advance the cause of interoperability (the ability to make different IT products work together efficiently, such as through application programming interfaces and the exchange of data).

The aforementioned new initiative aims to create a legal requirement for interoperability that would affect all "significant market players", not only the ones who fall under the scope of antitrust law.

An oversimplified description of what a fundamental change this would mean is that the kinds of interoperability requirements the European Commission previously imposed on Microsoft with respect to Windows could then also affect others, such as Apple, Nokia and RIM with respect to their smartphones, or Adobe with respect to Flash, PDF and Photoshop. And many others.

"Significant" is the key word. Antitrust law can be used to fight abuse of a "dominant" market position. Dominating a market implies a sizeable gap between a market leader and the rest. That legal test puts many powerful companies beyond reach for antitrust proceedings, but a wider circle of non-dominant companies can clearly be considered significant and it's time to do something about them.

This wouldn't mean an exploding number of antitrust cases. On the contrary, a major design goal is to achieve interoperability without having to go through lengthy antitrust proceedings. At the end of the process there would be a new European law, specific to the subject of interoperability between IT products. It would set out the rules for all significant players in that market.

The legislative process hasn't begun yet. The European Commission is now going to explore the feasibility of this plan, and if there's green light, then the actual lawmaking process will likely begin in 2012. This will take time, but it can have such a profound and highly positive impact that it's worth it.

Free software and open source can gain from this in two ways. One, software that is available under a FOSS license will probably meet all of the criteria set out by the possible new law. Two, a number of proprietary software vendors beyond the reach of antitrust law would be required to make interfaces and data formats available to all competitors, including FOSS-based competitors, on a fair, reasonable and non-discriminatory basis. Short of abolishing software patents, it's hard to imagine a FOSS-friendlier legislative initiative.

I recently heard the European Commission's Vice President for the Digital Agenda, Neelie Kroes, talk about this idea at a Brussels event. It became clear that she's very enthusiastic about this, and rightly so. She said in a recent interview: "Any kind of IT product should be able to communicate with any type of service in the future." This could be great stuff indeed.

You now have the basic idea, and I will report on this initiative when there are new developments. This one is just the first posting in a four-part series on the subject. Click here for the second part, which discusses the regulatory gap that currently exists because many major companies are not dominant in a legal sense. And by the way, you can follow me on Twitter @FOSSpatents.

{Interoperability} Market dominance vs. significance: closing a regulatory gap

This posting is the second one in a four-part series on a legislative initiative for interoperability currently being evaluated by the EU. Click here for the first part of the series (a brief overview of what this is all about).

EU competition law has four main areas: cartels, mergers, state aid, and cases against the abuse of a dominant market position.

The fourth area is the one to leverage if you aim to restrict the way a single powerful company (that doesn't form a cartel with others) uses its patents. That part of the law can only solve a problem if a company (i) dominates its market AND (ii) behaves in a way that is considered anticompetitive (such as refusing to disclose technical information necessary to interoperate with its dominant products, or wielding its patent arsenal to shut down reverse engineering of the same).

In connection with interoperability, the Microsoft case has established some helpful principles. In fact, the EU leads the world by example as far as interoperability is concerned, in no small part thanks to Mrs. Kroes's work as competition commissioner in recent years. The most recent example of how companies with an interoperability concern rest their hopes on the EU is a complaint by a US company named Versata against SAP.

However, if a company is not dominant in a competition law sense, then there's simply no case, neither in the EU nor in any other jurisdiction I know (such as the US) on the grounds of an abuse of a market position.

Even many big players can claim not to be dominant

Many companies can escape that part of the law because the legal test for market dominance is a very high hurdle. If a company has a quasi-monopoly and dwarfs its competitors, then it's certainly dominant in the given market. But if it's "only" a clear number one, there could still be enough competition in the market that dominance must be denied.

Let me give you an example for how high a hurdle it is: in my personal opinion, Oracle dominates the market for database software. It has roughly a 50% market share based on revenues, and it acquired MySQL, which is by far and away the most popular open source database. However, if a court of law had to decide whether Oracle is dominant in a legal sense, the counterargument would be that IBM's DB2 and Microsoft SQL Server are competitive forces to be taken into account.

Whether or not a company is dominant heavily depends on how the relevant market is defined: geographically and in terms of product characteristics. In the total worldwide market for mobile phones, Apple would probably not be considered dominant because Nokia still sells more units, the collective volume of Android-based phones is quite high, and RIM (BlackBerry) is also strong. But in a more narrowly defined subset of the market, Apple's market share could be considered to be much higher. Also, Apple could be considered dominant as an online music distributor or as a distributor of iPhone/iPad applications.

The outcome of the dominance test is always binary: there is a case, or there isn't. There can be intervention, or there can't. As a result, there's a huge regulatory gap.

While a few companies are considered dominant in certain markets, such as Microsoft for client PC operating systems or IBM for mainframes, there are many others who are also extremely powerful and have their customers locked in, but under antitrust rules they can't be pursued no matter what they do.

The EU can't and won't try to expand the scope of antitrust law as a whole. But the European Commission has apparently recognized that it shouldn't be required to bear the burden of proof that a company is dominant only to ensure interoperability. There should be a general obligation affecting not only dominant players but also the much wider circle of "significant market players", many of whom could also use their intellectual property rights (especially patents) to limit choice and stifle innovation.

How to define significance, such as in terms of percentage of market share, is one of many things the Commission is now presumably pondering.

For the next (third) posting in this four-part series on legislative initiative for interoperability currently being evaluated by the EU, please click here.

{Interoperability} Procedural framework: an action item in the Digital Agenda

This posting is the third one in a four-part series on a legislative initiative for interoperability currently being evaluated by the EU. Click here for the first part of the series (a brief overview of what this is all about) or here for the previous part, which discusses a regulatory gap that currently exists.

In terms of a legislative process, this one isn't even in its infancy. It's in a prenatal state. I believe there's a pretty good chance that the underlying idea will result in a new law within a couple of years. But if, when and in which form will depend on the process.

Legislative proposals at the EU level don't come as surprises. There's always some deliberation prior to kick-off, and while a lot of meetings are private, the overall direction in which an initiative is heading is written and talked about in public.

The idea of possible legislation on interoperability requirements for significant market players is part of a comprehensive work program called Digital Agenda for Europe. The official working document (HTML, PDF, other languages) was published in the second half of May. It identifies eight "action areas" and 16 "key actions" as well as many items that are called "other actions".

Subsection 2.2.3 of the Digital Agenda talks about ways to enhance interoperability and makes the following statement that I'll quote and explain:
Since not all pervasive technologies are based on standards the benefits of interoperability risk being lost in such areas. The Commission will examine the feasibility of measures that could lead significant market players to license interoperability information while at the same time promoting innovation and competition.
The meaning of this is that in addition to official standards controlled by consortia, there can also be de facto standards (such as data formats or interfaces that are as widely used as official standards) belonging to individual companies. The European Commission would like to ensure that such companies don't monopolize their data formats and interfaces.

The word "license" makes it clear that intellectual property rights are involved. In the interoperability context, patents are particularly relevant, and they are information documents (hence the word "patent"). But the Commission's wording leaves room for additional options.

The bold-face passage is very broad and vague. To "lead significant market players to [...]" could per se mean anything from politely asking to soft pressure to the creation of legal obligations. Below that paragraph, there's a list of action items. The last one of them is:
Examine the feasibility of measures that could lead significant market players to license interoperability information to report by 2012.
This is still very broad. It adds specificity in the sense that the Commission wants to know by 2012 which options it has. But a speech provided clarification.

A legislative initiative is the preferred course of action

A recent speech by the European Commission's Vice President for the Digital Agenda, Neelie Kroes, made it perfectly clear that the measure she wants to take is to create a new piece of legislation:
Whereas in ex-post investigations we have all sorts of case-specific evidence and economic analysis on which to base our decisions, we are forced to look at more general data and arguments when assessing the impact of ex-ante legislation. Just to be clear, while it is still early days, it is certainly possible that I will go for a legislative proposal.
Not only was the preferred way forward clarified but also that the envisaged measure will be far-reaching:
This could have a profound impact on the industry concerned so it is not a decision taken lightly. Many of you work for companies that could be concerned by such a measure. I invite you all to let me have your views.
This invitation for stakeholders to communicate their positions to the Commission suggests that a formal consultation process will take place sooner or later. That's the usual approach taken by the Commission.

At this stage, many key aspects of the future proposal have not yet been determined. Here's another quote from Mrs. Kroes speech that underscores the need for consultation and deliberation:
We are thinking very hard about how this could be achieved. Any such initiative would probably be limited to certain types of IT products. And it would likely involve some form of pricing constraints.
I have seen other legislative initiatives where it was fairly predictable at a comparable stage what the Commission had in mind to do. In this case, it appears that there actually are a lot of questions, including some fundamental ones, that have not yet been answered. That's why those of us who like the basic idea of this should make our contributions to the thought process sooner rather than later.

For the next (and final) posting in this four-part series on legislative initiative for interoperability currently being evaluated by the EU, please click here.

{Interoperability} FOSS-related opportunities and priorities

This posting is the fourth (and final one) in a four-part series on a legislative initiative for interoperability currently being evaluated by the EU. Click here for the first part of the series (a brief overview of what this is all about) or here for the previous part, which discusses procedures.

Like I wrote in the first part of this series of postings, I regard the initiative to impose interoperability requirements on significant market players as a first-rate opportunity for free software and open source.

Beneficial with or without software patents

There's no question that the number one item on the political wishlists of most community members is -- and will continue to be -- the abolition of software patents. I'm also aware that many in the community would prefer for interoperability-related patents to be available on a royalty-free basis. The last quote above from Mrs. Kroes's speech indicates "pricing constraints" as a likely option, which makes it pretty clear that patent holders won't be required to grant licenses on a royalty-free basis.

But even if some of us fear the initiative might not go far enough, we should at least support the parts we like and get as much mileage out of it as possible.

The FOSS community should embrace and support this interoperability initiative. There really is the chance to make some important headway. Everyone who opposes software patents (and patent royalties) altogether can continue to advocate that position. Even if we achieved the abolition of software patents one day against the odds, this interoperability initiative would still have value because it will very likely deal with more than just patents. For an example, undocumented interfaces are a problem with or without patents, but the future interoperability law could solve it.

Looking at Mrs. Kroes's track record, I'm sure she will make the most open-source-friendly proposal she can under the legal parameters and political circumstances that exist. Even if she may prefer royalty-free interoperability, it's not only politically but also legally impossible for a government to expropriate right holders without adequate compensation.

Politics is the art of the possible

The potential benefits of a European IT interoperability law are huge. Let's try to achieve as much as feasible. Politics is the art of the possible, and progress has to be made one step at a time. I don't see any other legislative idea in Europe (and this one would certainly have repercussions around the globe) that offers such an attractive combination of being potentially helpful and politically achievable in the near to mid term.

I believe SMEs (small and medium-sized enterprises) could be important allies to make this happen. We should work with them to give Mrs. Kroes the input and political support that will be needed to overcome whatever resistance some may try to mount (including some who demanded interoperability in the past, when other companies' intellectual property was concerned, but don't want to provide it with their own products and will therefore try to get the bill diluted if not derailed).

We can still be against software patents in general and explore ever more ways to achieve that goal. But that shouldn't preclude us from seizing what looks like a wonderful opportunity on the interoperability front.

If you are also excited about this initiative, please stay in touch by bookmarking this blog, emailing me via the contact form, or following me on Twitter @FOSSpatents.