ANTI-SOFTWARE PATENT ACTIVISM INFECTS THE UK COURT OF APPEALS: It is being reported today that Sir Robin Jacob, a judge at the U.K.'s Court of Appeal who specializes in intellectual-property law, spoke about issues surrounding software patents at a seminar for the Society for Computers and Law on Thursday evening in London.
During the course of Lord Justice Jacob's presentation, he hauled-off on the U.S. patent system, and particularly the practice of patenting software. Here are some quotes:
I realize that the article only cites selected portions of Lord Jacob's comments, but this is pretty outrageous stuff. It is intellectually lazy, and his comments do nothing for progressing the ongoing discourse on patent policy (Lord Justice Jacob was reportedly "glad" that he hasn't had to consider any software patent cases in the appeal courts yet).- "Do we need patents for computer programs? Where is the evidence for it?"
- "The United States takes the view that anything made by man, under the sun, can be patented. And they have granted patents for business methods, mainly computer business methods. But as far as I can see, it would cover a new and improved method of stacking oranges on a barrel." [Blogger Note - see US Patent 4,955,529 - "Produce Dispenser"]
- "[People have] got to look at all IP rights critically and say, 'Do we need them?'"
- "It's been said that (searching for prior art) is all going to be sorted out and will be very easy in due course--I don't believe it . . . and some of the fuzzy patents that have emerged from the United States tell you that it's going to be very difficult to stop very ordinary things from being patented."
Regarding the "evidence" he demands from software patentees, the European Commission has funded a study on the legal, technical and economic effects of software patents on innovation, but the study is not due to be finished until 2007. However, there are preliminary studies that have been conducted in the US that show that software patents are important, especially to smaller companies. One study, conducted by Ronald J. Mann from the University of Texas School of law, supports this notion, and further rejects the idea of "patent thickets" being the bane of software innovation. Regardless, there is no econometric or bibliometric data in existence today that shows that software patents are impeding the advance of software.
Also, it is disturbing to me that a sitting appellate judge all-but-admits that he can't be bothered to look into the putative benefits of software patents and instead engages in some haughty analysis why patent law shouldn't accept new technologies into its domain (if one still considers software a "new" technology). His views are not tempered by any legal findings or analysis, but seem grounded in the metaphysical concept that computer programming is not "science."
Finally, it was inevitable, as it is with most anti-patent proclamations, that Lord Jacob erroneously applied concepts of obviousness (i.e., "very ordinary things") as a basis for determining patentable subject matter. It cannot be stressed strongly enough that "obviousness" is a legal concept that is completely independent of "patentable subject matter." An application that is found obvious won't issue as a patent because someone else has already disclosed the invention prior to the applicant. An application that is not deemed patentable subject matter won't issue as a patent because no one has the right to claim such an invention.
It has been the repeated practice of anti-patent commentators to simply cite examples of flaky patents and extrapolate the conclusion that patents, as a whole, should be banned. This is nonsense squared. As Nathan Myhrvold succinctly said, "the stock market has stocks of companies that are flaky and questionable, right? But does that mean we should just avoid all public security markets?"
Posts
I don't think Lord Justice Jacob is being intellectually lazy, he's just giving a speech, which means that he is allowed to summarize prior work without giving citations and footnotes. But the points he is alluding to are stated more rigorously elsewhere, such as by Mr. Ronald Mann.
The Mann article does two things: (1) it presents anecdotes about people whose business methods relied on patents. But if you are going to insist that anti-softpatent advocates not argue by anecdote, then you should bind yourself to not argue via anecdotes about that one guy whose business model relies on patents. This is especially true because there is such a wider variety of business methods in software than in traditional goods production. The Net is awash in software sold or produced via no-patent business methods. (2) Mann's article presents loads of bellyaching about software patents. From his paper: "Software patents are multiplying so rapidly that it is likely that many product startups that are developing ultimately will infringe patents held by large existing companies. ... Several of my interview subjects joked that they thought it likely---without any investigation or particular knowledge---that there would be something in IBM's portfolio that their product infringed..."
The gist of Mann's article is that the patent thicket is not stopping innovation (although I can give you counter-anecdotes)---people just don't do patent searches. You, Mr. Zura, are a software professional, because (the machine on which is loaded) this web site is a patentable invention that you designed; I'm guessing that you didn't do a patent search, and if you did, you didn't contact anybody about royalties. The fact that this web site is up indicates that your solution to the software patent problem is to just not take such patents seriously; Mann's interviewees generally seem to have the same attitude.
Instead, Mann found that the attitude of for-profit software practitioners is to just wait for a lawyer to call them up, and as per the quotes above, they're pretty sure that at some point that will happen. It's surprising how, despite Mann's efforts to write a paper that says software patents are wonderful and perfect, consistent complaints about the inevitability of lawsuits and unproductive hassle keep cropping up. [I commend Mr. Mann for not supressing interviewee statements that don't fit his thesis, by the way.]
I don't think there's anything in the paper that supports the statement 'patents foster innovation', a few anecdotes of the successful use of software patents for venture capital, and lingering evidence throughout that there are problems with applying patents to a decentralized market like software.
Finally, Lord Justice Jacob is absolutely correct when he says "[People have] got to look at all IP rights critically and say, 'Do we need them?'"---I can't believe that you would disagree. To call somebody an activist because he proposes that we look critically at the law is, to put it politely, silly. Yes, the CAFC overturned 200 years of common law in the mid-90s, but their ruling said not a word about whether this was necessary to promote the progress of Science and the Useful arts, and cited no evidence that this expansion of patentable subject matter has any economic foundations. You seem to be proposing that we continue to not question whether this expansion of patent law has economic foundation.
Your claim, "Regardless, there is no econometric or bibliometric data in existence today that shows that software patents are impeding the advance of software" is likely correct, especially now that we've agreed to exclude anything that smells of anecdote. But patents are not about failing to obstruct, they're about promoting the progress, and there is no econometric or bibliometric data that says that that is happening, either. In terms of more theoretical considerations, there is abundant evidence that the industries that have suddenly been added to patentable scope are indeed substantively different, and there are fundamental issues with applying patents to them. If I may cite my own work as an example, I have an essay on the problem of applying patents to a decentralized market that was reprinted here.
There are actually some substantive economic studies that strongly suggest that in areas such as software development and general "business methods" (the definition of which tends to vary depending on the audience), the value of patents to future innovation in the economy declines significantly when the market in question has a low investment threshold or barrier to entry. In an age when anyone can create innovative software with little investment, strong software patents may tend to stifle, rather than foster, innovation, because the recipient of a software patent can receive a windfall of coverage based on a small investment and a small dedication of information and disclosure to the public. Compare this to pharmaceuticals, as a counter-example, where research investments can go into seven or eight figures. Thus, for software, patents can create substantial "dead weight loss" for the economy -- they make money for the firm holding the patent, but overall the economy loses money. Moreover, you will see (and are seeing, in the U.S.) widespread escape from the territorial reach of U.S. patent laws for companies attempting to create innovative software products, and as a result even more loss to the U.S. economy.
Posner & Landis, in their stunning (and underdiscussed) work on the Economics of Intellectual Property, actually delve into this issue in detail. Moreover, there are economic studies from MIT and Cato that you can find via Google that explain, in pure economic terms, how pure software and business method patents serve as a substantial harm to the long term health of the U.S. economy, and, especially for "pure" business method patents, tend to impede exactly the type of free-market competition that the U.S. economy was built upon.
Good points. With regard to Lord Jacob, I think it is understood that no one is demanding that his comments, taken in the context of a speech before a practice group, require formal citations. However, as a sitting judge, I think it was inappropriate of him to flippantly dismiss matters such as software patents that are likely to appear before his court. I'm not sure what constraints there are in the UK for judicial commentary, but in the US, judges are highly discouraged from publicly advocating positions on issues that they would be deciding in court (e.g. "all abortion should be banned because it is an evil thing"). Regardless of what a judge may personally think, it is expected that they will base their decisions on an impartial application of law to a specific set of facts.
Second, you can either oppose software from a legal standpoint (i.e., the caselaw is wrong), or from a philosophical standpoint (i.e., computer programming is not science and does not deserve patent protection). Judging from his comments, he appears to oppose software patents for philosophical reasons.
Thus the ultimate question is "is computer programming 'science?'" If not, why not? If so, why should inventions created by programmers be denied access to patents unlike all other areas of science? And if software patents are to be opposed for their exclusionary effect, why not ban patents altogether?
As I have repeatedly stated, the US had, for over 2 decades, a vigorous debate over the degree to which patents would be restricted for software. The problem is that once you start restricting method claims for software, you might as well start restricting method claims for all other areas of technology - after all, method steps are merely reciting a sequence of physical truths, be they electrical, mechanical or chemical.
Finally, the legal bottlenecks described by programmers with regard to software patents are no different from any other technology. Take, for instance, nanotechnology. In July of 2005, IEEE pulished a survey of nanotechnology patents and found that 3 areas of nanotechnology (quantum dots, nanowires and dendrimers) were controlled by 4 companies with their patents. In the area of carbon nanotubes, it was found that many companies had hundreds of patents that often overlapped one another. This is a classic licensing nightmare. Should nanotech patents be banned because of this problem?
The opponents of software patents are self-serving hypocrites or simply denying the simple fact that anything can be implemented as either software or harware or a combination of both...
A classic Turing equivalence - the First Law of Computing if you will...
Interesting. Some seem to equate wearily dismissive and lazy.
However a general difference seems to exist between those who can extrapolate to a general statement, which most people can get their heads around, and those who wish to either equate or confuse 'all software patenting' with the fluff that is granted at present in the US.
Judge Jacobs seems to be stating in quite easy to understand terms (Clapham Omnibus Man) that business method patents with only the most tenuous link to some everyday implementation, and where the enthousiasm lies solely in the marketing ploy or administrative steps, are of at best a nuisance value to society. This view can be traced throughout the case law of various European countries and also in the drafting notes behind the European Patent Convention. If this is to be changed then one would do well to read Adam Smith's 'Wealth of Nations' where he warns of listening to the man of commerce who asks for a change in law - he cautions that such a request is rarely for the good of Society but may be assumed to be made for the purest of reasons, namely greed.
Come on chaps, do you really need Turing to hide behind? The issue is simple.
If the application includes software which does something that a programmer/analyst furrows his or her brow at, then says 'ah - cool' then fine. Society is enriched by the disclosure and other inventors may seek to provide alternatives or improvements.
If there is no such reaction there are two possibilities; it's excluded from patentability (the extreme case where there is no substance to consider for 'ah - cool') or it is simply obvious (which is why we have patent examiners with technical qualifications, who may or may not show documentary evidence dependent on the level of obvious). Even the EU occasionally understands this.
Non-obvious business ideas? If you want to do that in the US, go ahead. Just consider for a moment that, as in other areas, it may be the US which is out of step, not the rest of the world. Such blasphemy! I don't say to give protection is stupid; let the mess at the USPTO and the fuzzy patents ridiculed in your press and dealt with in your courts speak for itself.
Case Law before the European Patent Office is clearly going in this direction and is relatively unchallenged in the appropriateness of its decisions other than by some extreme self interest groups and personalities. The jurisprudence is relatively (not perfectly, but hey) consistent over time and it is worthy of note that both Judge Jacobs and the EPO, especially the Boards of Appeal, appear somewhat more jealous of their independence of decision than other instances around the world.
Although slight differences in logical methodology exist between UK and EPO decisions they are delightfully aligned in outcome and share a common desire to call a spade a spade.
Good software patents fine, trivial, silly patents, some of which include a mere nod towards the implementing software/hardware not fine.
Post a Comment