The issue of whether software should be patentable has been settled in New Zealand, and may soon be settled in the US. The Washington Post reports:
In a trio of Supreme Court decisions between 1972 and 1981, the court held that mathematical algorithms were not eligible for patent protection. Since computer software is little more than mathematical algorithms encoded in machine-readable form, most of the software industry assumed this meant you couldn’t patent software. But then, in the 1990s, a patent-friendly appeals court handed down a series of decisionsthat opened the door to patents on software. That triggered a wave of patenting that has drowned the technology industry in litigation.
In principle, the Supreme Court’s old precedents ruling out patents on mathematical algorithms are still good law. And today, the Supreme Court announced that it would hear an appeal that, for the first time in 30 years, will directly address the patentability of software.
The court that has to date decided the law is the little known United States Court of Appeals for the Federal Circuit. It was only created in 1982. Unlike the other appeal courts, its rulings are binding on the entire United States, not just one circuit. Only the Supreme Court can over-rule it.
Now the Supreme Court will have an opportunity to weigh in on the case. And while the high court could issue a narrow ruling based on the details of the patents in this case, it could also take the opportunity to fix the software patent mess more broadly. All it would need to do is to reiterate its earlier position that patents claiming mathematical processes — a.k.a. computer software — isn’t eligible for patent protection unless it’s tied to a specific machine or physical process.
Which is basically now the law in New Zealand.
The high court will be reluctant to do this because it would be disruptive. Reiterating that mathematical algorithms can’t be patented would call into question thousands of patents held by major software companies. And these companies could complain, with some justification, that the Supreme Court’s failure to rule on the issue for more than 30 years was a tacit acceptance of rulings by the Federal Circuit.
Still, the federal circuit cannot overrule Supreme Court precedents. And the federal circuit’s experiment with software patents has been a disaster. As the patent scholar James Bessen has argued, the patent troll crisis is really a software patent crisis. Software patents are far more likely to be involved in litigation than other types of patent. The result: According to Bessen’s calculations, troll-related litigation cost the U.S. economy $29 billion in 2011 alone. Reiterating that “pure” software can’t be patented wouldn’t just be good law — it would also save the nation billions of dollars in litigation costs.
The case will be watched very closely.