My tale-of-caution post on yesterday's adoption of a patent reform bill in New Zealand has been one of my most popular writings this month -- popular in terms of the interest it has drawn, not in terms of the message that certain Internet crowds would have preferred. The stated goal of that post was not to curry favor with any audience, but to discuss the matter responsibly.
This follow-up post has two distinct parts. In the first part I'll explain the discrepancy between mainstream media reports that New Zealand has banned/outlawed/whatever software patents and the more nuanced and cautious opinions expressed by IP-focused writers, including a couple with whom I don't always agree. In the second part I'll try to explain with what kind of mindset one has to approach a piece of substantive patent law when trying to assess the effect it will have further down the road. What may appear crystal clear to a geek usually doesn't have its broadest meaning when patent offices and courts of law have to make sense of it. I think there are some easy ways to show why that New Zealand law isn't tantamount to the abolition of software patents.
Disconnect between mainstream media and IP-focused writers
Most media reports on the decision have been misguided, and have in turn misguided the readers of those articles. Interestingly, even an overtly (and sometimes shrill) anti-IP blog named Techdirt has been far more measured and closer to accuracy this time than all those mainstream IT media, saying that "New Zealand Sort Of Bans Software Patents" (emphasis mine) and that "while the intentions appear to be good, some of the language in the bill may open some loopholes, so we'll be interested to see how the bill is actually interpreted, and what happens within the actual tech sector". This more cautious take by Techdirt should give all those people pause who attributed, in certain discussion forums, my analysis, which was more than anything else an effort to prevent people from making false business decisions in reliance on political spin, to the fact that I have (voluntarily and proactively) disclosed consulting relationships with industry players Microsoft and Oracle, who presently support strong IP protection for software. It should also be food for fought for those who dismiss Intellectual Asset Management magazine's analysis as biased because its readers are mostly patent lawyers.
There's an undeniable discrepancy between the assessments of IP-focused blogs/websites on the one hand and mainstream media on the other hand. Now, whom should you trust if you don't have the time and background to research the whole issue for yourself? That's your choice, but you should at the very least be cautious when you see that IP-focused people throughout the political spectrum -- from radically anti-IP (Techdirt) to moderate (this blog, because I used to campaign against software patents but do understand the industry's perspective as well) to consistently pro-IP (IAM Magazine's Joff Wild and I disagree from time to time, such as on standard-essential patents) -- do not portray the New Zealand decision in the simplistic black-or-white manner in which mainstream IT media do. Even in case you assume that Techdirt's Mike Masnick (who said he was surprised when it turned out he had received funding, indirectly, from Google), IAM's Joff Wild and I are agenda-driven while mainstream media are not (or to a lesser degree), the presumed agendas of the three of us IP bloggers cover the entire spectrum.
Substantive patent law (the rules of patentability, including patent-eligibility) is a highly specialized subject. Patent law is way more specialized and, to the general public, far less accessible than most (if not all) other fields of law, and substantive patent law is a thin vertical slice, and by far the most abstract part, of patent law. In my anti-software-patent lobbying efforts I once met a Member of the European Parliament who's a law professor and he asked me questions about this for almost two hours because he wanted to understand the intricacies of software patentability rules. Keep this in mind when choosing which kind of sources to trust.
The Internet is huge, but you won't find a reporter for a totally independent mainstream publication who really understands this particular field.
Before we get to the question of competence: mainstream journalism isn't always as independent as you might think. Some publishers have stricter disclosure rules than others. Even where the rules are in place, you won't find 100% compliance across an entire industry. IT websites are typically advertising-funded. It's not advisable to assume that editorial decision-makers never care about the interests of the major advertisers who fund them. And journalists always depend on their readers. Populist headlines are quite useful for link-baiting, even if inaccurate. Just to be clear, most of the journalists I know seem perfectly honest, so most of the time the problem is just that they have to do several stories a day on a variety of subjects. They simply can't, even if they give it their best, have the same detailed understanding of a given topic as people who've been focusing on these complex issues and processes for years. Furthermore, they have space constraints and must keep things short and simple for an audience, but even experts frequently can't be accurate and short/simple at the same time.
I often get asked why my blog doesn't allow comments, and why comments are deactivated on almost all of my Google+ posts. (I do, of course, get responses on Twitter, and sometimes I answer there.) That's because there's too much half-knowledge out there that is too time-consuming to address. I'd have to debunk stupidity after stupidity to prevent readers from getting confused by pretenders -- people who use specialized terminology, invoke laws and cite to judicial decisions but actually don't know what they're talking about. I believe my time is better spent on other activities, and this blog is basically a by-product of, and promotional tool for, my work as a consultant.
To be clear, independent journalism is important, and there are literally hundreds of journalists I've been in contact with just over the last few years, most of whom I totally respect and try to help with information when they request it. They do an important job, and their efforts and mine are complementary, except that there can be situations in which I disagree with mainstream media. For one example, let me refer you to Philip Elmer-Dewitt's article on how I was right and mainstream media were wrong on the March 1, 2013 Apple v. Samsung damages ruling. Philip notes that "[a]lmost lost in the flood of 'Apple setback' headlines was the lone voice of FOSS Patent's Florian Mueller, who pointed out that what was being described as a drastic cut in a $1.05 billion award could, in theory, turn out to be an increase". By now, everyone knows that I was right because Samsung acknowledged it. I was even being conservative: Samsung says Apple is claiming "vastly greater damages" in the retrial. If you insisted on the broadest definition of independence, you'd have had to rely on the mainstream press, which told you that Apple was going to get less money out of this case. But, with the greatest respect, independence does not compensate for incompetence. I always say what I believe in, and even if people suspect that I want to do clients a favor, there are examples when I get things right that others get wrong. A more recent example is the scope of a U.S. import ban Apple won against Samsung. Most of the initial media reports said that only certain older devices were banned. But the import ban would also apply to any newer (including future) devices that infringe in the same way. I published the letter the ITC sent to U.S. Customs because it leaves no doubt about the potential scope of the order.
Let me close this first section by conceding that there is a conundrum facing everyone trying to obtain information on these cases. On the one hand, mainstream media get certain things in these highly-specialized contexts completely wrong, and I've talked to various lawyers involved with the smartphone disputes (mostly when I saw them in German courts, where U.S. counsel is frequently also present) who told me they find my blog fundamentally more reliable on their cases than even the most reputable financial papers, which sometimes don't even understand who's actually won a lawsuit (if there's a mixed ruling). On the other hand, the kinds of people who specialize on these issues -- even anti-IP activists like the Techdirt author or a certain European anti-software patent organization -- will sooner or later have some financial relationship of some sort with some industry player, even if indirectly, such as through membership fees paid to industry associations or donations made to non-governmental organizations. These major players are like an elephant in a room: they're everywhere. You'd also be hard-pressed to find a professor in the relevant fields (law, economics, computer science) who's never ever represented or advised a client, never taken money for a study, never appeared in court as an expert witness. The holy grail of someone who's competent enough to have consulting and other opportunities with large organizations but stays totally independent -- and still finds enough time to stay up to date on all the details of an issue -- simply doesn't exist. That's the number one reason why specialized blogs like this one not only exist but thrive.
How patent examiners and judges interpret substantive patent law in practice
When laypeople read in the New Zealand bill that something isn't patentable if the invention lies solely in it being a computer program, they think that this can be interpreted in only one way: no more software patents. But they have to understand that legal professionals and software engineers are divided by a common language, to the extent that even the meaning of words like "invention" and "computer program" differ greatly between everyday use and legal use. If a given company told its software developers that they can no longer obtain patents on inventions matching that criterion, they'd stop filing any patent applications because it would look to them like a waste of time -- but if the same company sent the same instructions to its patent department, the lawyers would simply conclude that they have to draft around the exclusion, and that's what they would do from there on out.
What additionally complicates this is bias. More than 2,000 years ago, Julius Caesar wrote in his Commentary on the Gallic War (parts of which I had to translate from Latin in high school) that "fere libenter homines id quod volunt credunt" -- people believe quite readily that what they want to be the case is the case. Even if it's not. The Romans capitalized on this through certain stratagems, making their enemies believe that they were on a retreat, only to launch a massive assault once their enemies relaxed. It worked again and again.
In theory, patent examiners and judges should read substantive patent law without any bias. And many of them try hard to. And the one thing they certainly won't have is an anti-patent bias. New Zealand's patent office is not going to be interested in interpreting the law in a way that would require it to lay off half or more of its staff. It will have to apply the law, and it may receive instructions from the executive government. But it won't approach this new piece of legislation with a "good riddance" mindset.
The patent prosecution process is inevitably skewed because it's easier for a patent examiner to grant an application than to reject it. If you grant it, anyone against whom it's asserted later has to deal with the problem. As long as you reject it, the patentee's patent attorney will keep trying to persuade you to grant it, and will amend the claims and other parts of the documents until you're satisfied.
The foregoing is not a consideration for the courts who rule on the validity of a patent. Still, you can't expect any judicial activism of the abolitionist kind. Judges look at it from both angles. On the one hand, there must be limits to what is patentable. On the other hand, they know that patent law also means innovators are entitled to protection, provided that their inventions meet certain criteria.
When you have to balance boundaries with entitlement, New Zealand's new law doesn't provide nearly the degree of clarity that mainstream media would have you believe. A phrase like "lies solely in it being a computer program" is very interpretable; the examples of one patent-eligible and one patent-ineligible invention are worlds apart and not clear enough; and when you then see that lawmakers intended to keep "embedded" software within the scope of patent-eligible subject matter, but consider that the question of whether software is "embedded" can in most cases not be answered based on a technical solution for which someone seeks a patent (just like software developers who write "embedded" software may very well use code in it that they previously used in non-embedded software, or the other way round), then you know that all of the line drawing that needs to be done is now your job. This happens all the time, not just in connection with patents: politicians can't reach a consensus on a clear-cut, sharp set of rules, so they produce something vague and leave it to the courts to sort it out. And courts tend to find that an invention is not subject to an exclusion if the exclusion isn't sufficiently clear. That's the way it works, most of the time.
The NZ patent reform bill is not clear. I'll tell you what would be clear, or relatively speaking, clearer:
- If New Zealand had adopted Richard Stallman's proposal of a broad exclusion from enforcement, then it would simply be impossible to sue software developers and users over patents, regardless of what those patents cover. (This would, of course, depend on whether the exclusion from enforcement is phrased clearly by lawmakers, but I just assumed so for the sake of the argument.)
The NZ law would at least be a whole lot clearer if the examples of a patent-eligible and a patent-ineligible kind of invention were closer to each other. The example of an auto-complete application using a database and forms doesn't mean much: based on that description alone, the rather permissive European Patent Office wouldn't grant you a patent, and even the most permissive USPTO most probably wouldn't.
The auto-complete example misses the point. The tricky area in which to draw the line involves software that makes a technical contribution, such as a file system that stores data more efficiently, a database that runs certain types of searches more quickly, security software that detects attempts to hack a system, video codecs -- and those inventions can be implemented in hardware as well, so an exclusion that merely refers to computer programs (and uses an auto-complete program as an example) doesn't really capture them.
Video codecs, which I just mentioned, are a particularly good example: the NZ patent reform bill doesn't say that graphics adapters are no longer patentable. Now, if someone patents a technique that can be used in a graphics adapter, but someone else implements it in software, then he may get a software patent in the sense of a patent that software developers and users may be found to infringe. The patent office can't reject the patent application because the law doesn't say that innovations relating to graphics adapters are no longer patent-eligible (this also was never the intention of any UK court, by the way). But at the enforcement stage, there's no exclusion in NZ of the kind that Richard Stallman advocates. So the patent can be granted, and an infringer can ultimately be held liable, with all that it entails.
I'm confident that these explanations have already furthered many people's understanding of the issue. It's a difficult subject, and I encourage everyone with an interest in understanding this in more detail to study the European situation. The U.S. approach (which the Supreme Court described as inclusive in its Bilski opinion) is that the scope of patent-eligible subject matter has boundaries, but no exclusions of the kind that Europe has had for several decades now, and that people overrate in connection with the New Zealand patent reform bill. Of course, most of the people who reported and commented on the New Zealand decision have never even read a single European court decision on patent-eligible subject matter. (By the way, just this month I've requested the Federal Patent Court to let me inspect the record of roughly two dozen pending nullity -- i.e., invalidation -- cases.)
Almost a decade ago, the Council of the European Union adopted a proposal on software patents that I was fighting hard against. If you believe that the New Zealand law abolishes software patents, how would you have liked this article from the European proposal, for example?
"Article 4.A
A computer program as such cannot constitute a patentable invention.
A computer-implemented invention shall not be regarded as making a technical contribution merely because it involves the use of a computer, network or other programmable apparatus. Accordingly, inventions involving computer programs, whether expressed as source code, as object code or in any other form, which implement business, mathematical or other methods and do not produce any technical effects beyond the normal physical interactions between a program and the computer, network or other programmable apparatus in which it is run shall not be patentable."
Doesn't that sound even better than the NZ law? And it gets even better, seemingly, if I show you how the European proposal defined "technical contribution", a term that also comes up in the article I quoted above:
"Article 2b
(b) 'technical contribution' means a contribution to the state of the art in a field of technology which is new and not obvious to a person skilled in the art. The technical contribution shall be assessed by consideration of the difference between the state of the art and the scope of the patent claim considered as a whole, which must comprise technical features, irrespective of whether or not these are accompanied by non-technical features."
Well, even the above language, which appears at first sight to constitute a clearer exclusion of software inventions from patent law, would have allowed for most software patents to be granted. You don't have to take my word for it. I'll just refer you to the analysis published at the time by the Foundation for a Free Information Infrastructure (FFII), the NGO that started the European anti-software-patent movement. Here's how the FFII commented on Article 4.A of the European proposal:
"This article is based on a JURI amendment (= amendment from the people who also run the Council's working party) that made it through the EP Plenary Session only because it is meaningless and therefore harmless. The Council's additions however make this amendment harmful. They use the wording of Art 52 EPC and insinuate that a 'program as such' is only the 'expression' side of this program, i.e. narrow copyright-style claims that nobody would write into a patent anyway. This interpretation does not conform to the usual requirements of interpretation of laws and is not currently in use at the patent courts. It deprives Art 52 EPC of all limiting meaning."
You may now be confused, but hopefully at a higher level. Don't take exclusions from the scope of patent-eligible subject matter at their face value. It's a lot more complex than that.
If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.
Share with other professionals via LinkedIn:
