Tuesday, June 29, 2010

Who lost Bilski vs. Kappos besides Bilski & Warsaw? Ten answers

Yesterday's Supreme Court ruling in re Bilski was unfortunately a clear victory for those favoring an expansive patent system and the patent inflation it entails. The two "inventors", Bernard L. Bilski and Rand Warsaw, saw their patent application thrown out, and that makes them the losers of the court proceeding, but few people will care about them now. Steven Vaughan-Nichols accurately wrote that Bilski loses, but the patent madness continues.

There are many losers of this outcome, and here's my top ten (besides Bilski and Warsaw -- may they live long and prosper without the patent they applied for):
  1. The free software and open source communities
  2. Software patent abolitionists
  3. Small and medium-sized companies who can't or don't want to play the patent game
  4. The proponents of bogus treatments: Linux Foundation, Open Invention Network etc.
  5. The Patent Absurdity movie
  6. Red Hat
  7. Google's foray into new markets (Android, WebM)
  8. Salesforce.com (Marc Benioff)
  9. The "captive court" theory
  10. IBM's open source credibility
Let's go over them one by one.


1. The free software and open source communities

The notion of Free Software is fundamentally incompatible with software patents. CiarĂ¡n O'Riordan, the director of the EndSoftPatents.org campaign, made a statement at a European Commission hearing four years ago where he accurately said that software patents and free software don't mix whether you cut the price of a patent in half or double it.

But the other part of FOSS, the open source community, is equally affected. While it doesn't emphasize the concept of freedom as much as Richard Stallman and his followers, I know many open source advocates who are no less opposed to software patents than RMS is.

I venture to guess that the Bilski ruling will represent an obstacle to GPLv3 adoption. I wish the whole world could accept the patent clause in GPLv3, which is meant to counter patent licensing deals by FOSS companies and other entities, but under the circumstances it will be very hard to convince businesses and other contributors to FOSS development that this our-way-or-the-highway approach works in the world we (currently) live in.

I am forced to say this to my dismay. I know that some may try to hold this against me but I spell it out like it is.

2. Software patent abolitionists

As the founder and former director of the NoSoftwarePatents campaign, I'm part of this movement. This isn't a concession speech on behalf of the cause. But we are at a juncture where we have to face the recent lack of progress in the political arena as well as in courts. No one is ever beaten unless he gives up the fight -- but if a certain approach doesn't work over an extended period of time and on different continents, then it may be time for a better mousetrap, ideally one that can pave the way toward abolition further down the road.

The majority of the justices took a clear pro-patent position. A minority of the court -- which is also among the losers, especially Justice Stevens -- supported certain pieces of well-reasoned criticism of the institutionalized excess of the patent system.

The SCOTUS determined that courts will have to support software and business method patents unless lawmakers intervene. However, the massive support that big industry lends the system and the parameters that globalization brings with it favor the other camp.

Pieter Hintjens, a former president of the FFII, once predicted that the abolitionists of software patents (if not all patents) would succeed more quickly than the opponents of nuclear energy. The anti-nuclear movement has achieved something. It took that movement decades to score a few partial victories, but then they didn't have BlackBerries, Wikis, Twitter and Facebook to organize themselves in the beginning. The problem with abolishing software patents is that it's almost binary: you either do away with very large parts of the entire patent system (which may be the right thing to do, but it's quite a challenge) or you have to content yourself with minimal restrictions. That's tough.

3. Small and medium-sized companies who can't or don't want to play the patent game

While big industry is united behind the patent system, not just in the IT industry but beyond, small and medium-sized enterprises (SMEs) are different.

Most patent trolls are SMEs, and they obviously cheer the Bilski decision. There are also some companies who are indeed practicing entities but their products or services are very limited. Think of someone selling a copy protection mechanism on which he has a patent or two: he likely won't need anyone else's patents to build his product, so the system works for him.

The SMEs who suffer are those who build larger programs. In a few hundred thousand lines of program code, any single line could theoretically infringe someone's software patent. Practically, you could probably find thousands of patents that might read on such a program. If you don't have the legal department and other resources to fend off infringement claims, and if you just don't have enough patents yourself to cross-license with large players on attractive terms (ideally without money changing hands), then the Bilski ruling has just made it even harder for you to innovate.

In light of that, I think SMEs -- like the FOSS ecosystem, which includes many SMEs -- should increasingly make the distinction between more and less harmful ways in which others use their software patents. I wrote about that recently. Trolls are a problem, but strategic holders with an exclusionary agenda are even worse.

4. The proponents of bogus treatments: Linux Foundation, Open Invention Network etc.

I have been vehemently opposing for a long time any attempts to lull the FOSS community (and others, especially political decision-makers) into a false sense of security concerning the risk that software patents represent.

I criticized IBM's "pledge" of 500 patents back in January 2005, on the very day it was announced. Five years later, I published evidence for IBM's betrayal of the pledge.

In a November 2005 slashdot op-ed, I explained why patent "pledges" and "pools" are fundamentally flawed approaches. That was the day on which the Linux Foundation (then named OSDL) started its Patent Commons, which I guess has never helped any FOSS developer the slightest because programmers don't look up a list of "pledged" patents when they go about their work. They want to go ahead and write code and deal with patents subsequently.

In recent weeks, I have discussed the Open Invention Network (OIN). I described it in this "OIN demystified" posting. Last week I discussed its less than informative press release on its new Associate Member program. But I actually wish the OIN could do a better job and give itself a trustworthy legal structure. That's why I suggested four alternative ways to address the biggest problem I have with the OIN, which is its ever-changing and arbitrary scope of licensing.

After the disappointing Bilski ruling, I encourage everyone else in the community to ask yourselves three questions:
  1. Haven't all those smokescreens been a major distraction from the planning and pursuit of better initatives? That wasn't only my concern. Richard Stallman and Bruce Perens also made such statements back in 2005.

  2. How can the existing initiatives be improved? For the OIN, I tossed out suggestions, and I'll think more about this.

  3. Do new initiatives such as the Defensive Patent License (DPL), possibly coupled with active patenting by the community, have new elements to offer that can make a major difference?

5. The Patent Absurdity movie

Two months ago I criticized the Patent Absurdity movie for several reasons.

Obviously, Patent Absurdity wasn't meant to be a pleading for the SCOTUS to consider, although some hoped it could indirectly influence the outcome. But given that the decision didn't even mention software patents directly (although it is a very strong indirect endorsement of them), the movie's close connection with the Bilski case makes it even less useful.

The idea was a good one. It needs a better implementation. One of the technology policy movies that I really like is Epic 2014. Video can be quite powerful, and I hope that a successor to Patent Absurdity will make a more compelling case and will be less interview-centric, especially since at least a couple of the persons interviewed aren't perfectly credible opponents of software patents in my view.

6. Red Hat

Unlike Novell (which has a licensing deal with Microsoft in place) and Canonical (which became the first GNU/Linux distributor to sign a deal with MPEG LA), Red Hat has so far refused patent licensing as "a tax on innovation". Red Hat's CEO Jim Whitehurst recently blasted software patents, but he may have to bite the bullet.

Red Hat has also contributed to the mess. Red Hat's own patent promise is weak, and many in the community know it. The DPL may be an opportunity for Red Hat to prove it's truly defensive. What's far worse is that Red Hat partners, commercially and politically, with IBM and other proponents of software patents on different initiatives -- pretty much all of the ones I mentioned on my list of bogus treatments (section 4 of this posting). A lot of that was driven by a former Red Hat lawyer who has meanwhile left. He even lobbied alongside some pro-patent companies to keep the EU software patent directive alive when we had already defeated it in July 2005.

It would be best if Red Hat could clearly dissociate itself from activities that contribute to the mess or help sustain it. Red Hat should act in accordance with the recent declarations of its CEO, which are so far only lip service.

It would be unfair to call Red Hat's business model -- they are the largest Linux company but only contribute about 10% of Linux development -- purely "parasitic", but describing it as "symbiotic" is a euphemism. At any rate, Red Hat may have to invest more in innovation and take out more patents, not only in absolute terms but also relative to sales.

7. Google's foray into new markets (Android, WebM)

Like many other patent holders, Google wants to have its cake and eat it: they love their own patents (especially in the search engine business) and don't like everybody else's. That's not realistic, and I think Google may increasingly realize it. Google should have been opposing all software patents for a long time, instead of telling others in the industry how important their search engine patents are for their core business and how proud they are of those.

What I really appreciated was a statement by Google's chief lawyer, David Drummond, that Google considers the use of patents against open source a bad idea and won't ever do it (quoted in this blog posting).

I'd also love to see a patent-unencumbered codec. However, WebM does raise the legitimate questions of patent clearance and indemnification -- especially after Bilski. But if there aren't satisfactory answers, then I tend to believe that MPEG LA isn't the biggest patent-related problem of all the ones that open source faces. Its licensing terms don't appear outrageous.

Android also faces patent issues. The Bilski ruling doesn't support HTC and other vendors of Android-based phones in their dealings with Apple.

Google's exclusionary use of patents in its core business (search engines) and its aforementioned support for the patent system may now come back to haunt Google as it forays into new markets in which the incumbents have already set up major patent thickets. I wouldn't be surprised to see Google rethink its stance on the desirability of patents in the coming years. In fact, I hope so.

8. Salesforce.com (Marc Benioff)

Some may call Salesforce.com founder and CEO Marc Benioff a "bigmouth" but he certainly does have guts. Instead of doing a licensing deal with Microsoft (which many others such as Amazon.com previously did), he decided to stand up and fight. The NoSoftwarePatents movement would love him to prevail in court.

But if he reads the Bilski ruling, he should pay particular attention to the court's reference to "technologies for conducting a business more efficiently". That's exactly the category of patents -- software-implemented business methods -- that represents the greatest threat to Salesforce.com. The Bilski patent application per se wasn't such a patent. The SCOTUS opinion, however, makes it clear that software-implemented business methods should be patentable in general.

9. The "captive court" theory

In the current debate over EU patent reform but also in other contexts, the "captive court" theory -- which came up in a SCOTUS ruling years ago -- was the most important element of the criticism that our movement voiced. I also mentioned it on multiple occasions, such as in my recent presentation at LinuxTag.

The theory is that courts that are patent-focused are essentially part of the patent system and more likely to favor the interests of patent holders as well as a broad scope of patentable subject matter, including software patents. The assumption is that people whose own career and influence is very much linked to the strength of the patent system are more favorable to such concepts as software patents. Even if it's not a matter of grabbing more power (to many of those people it probably isn't), everyone believes in the good that his profession can do and that's a kind of bias.

I still think the concern isn't wrong, but what can no longer be claimed now is that independent courts are willing to abolish software patents. Recently the German Bundesgerichtshof (Federal Court of Justice) declared software patents perfectly legal, in a ruling on an XML/HTML document generator. The BGH is the German equivalent of the SCOTUS, and the ruling is more specifically focused on software patents but has a Bilski-like overall effect.

10. IBM's open source credibility

They say that the first victim of war is the truth, and sometimes that's also the case in legal disputes. In this case, I don't think the truth is top of the list but there was one incident in the process that was really outrageous: IBM claimed that software patents liberated programmers and fueled the explosive growth of open source software development. That claim is unfathomable and insulting to the FOSS movement.

That amicus curiae brief was filed last year and previously written about by Glyn Moody, slashdot, and the software patent wiki.

On the occasion of the Bilski decision I felt it was necessary to remind people of it (and to inform those who weren't previously aware). IBM has a big credibility problem because it tries to have its cake and eat it: one the one hand, IBM claims to support open source and open standards, and on the other hand, IBM is the world's largest patent bully.

I don't deny that IBM has made important and useful contributions to open source (even to free software, although that is not IBM's ideology). However, if push comes to shove, its patents are dearer to IBM's heart than FOSS. IBM is also the primary sponsor of the bogus treatment schemes I listed in section 4. Through donations to various organizations such as the Linux Foundation, IBM has systematically silenced many critics of software patents (even Linus Torvalds welcomed one of IBM's bogus treatments for patents, which was very disappointing because he is genuinely against software patents).

As long as IBM doesn't decide on a fundamental departure from its patent-mongering ways, it will be an open hypocrite as opposed to a sincere open source advocate. The Bilski situation is another example of the credibility problem that IBM has more than any other company in the IT industry. It's high time it did something about it.

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.