At last, it seems, something is to done about the dyfunctional way America's patent system operates. Two recent developments suggest calls for patent reform are finally being heard at the highest levels. First, in 2013, defying expectations, the House of Representatives passed (by an overwhelming majority) the Innovation Act, a bill aimed squarely at neutralising so-called patent trolls. These are individuals or companies who buy up lots of patents and then use them to extract payments from unsuspecting victims. Second, the US Supreme Court agreed to rule on what is the most contentious issue of all：which inventions are actually eligible for patent protection.
Frivolous lawsuits filed by trolls cost American companies $29 billion in 2011 alone. Troll (known in the legal world as ゛patent assertion entities゛or ゛non-practising entities゛) do not make anything, but send out thousands of ゛demand letters゛to companies that allegedly infringe what are often vague and overly broad business-process and software patents. In recent years their targets have spread from high-tech firms to universities, retailers, hospitals, charities and even consumers. The trolls demand a settlement fee, which many of the victims pay rather than face the punitive cost of litigation - $2m or more these days.
Concern that the flood of frivolous patent suits in America is hobbling innovation and competition has caught Congress's attention. Startups threatened by such actions frequently have to withdraw from the business or go bust. Bigger companies that choose to fight spend tens of millions of dollars on litigation that could otherwise go on research and development.
In a sense, the legal bullying by trolls is a symptom of a wider complaint：the poor quality of many patents granted by the US Patent and Trademark Office (USPTO) in the past decade - especially those covering computer software and business transactions (themselves often based on software algorithms). One such patent covers simply upgrading computer software over the internet. That is admittedly useful, but hardly novel or non-obvious - the three fundamental requirements for eligibility adopted by patent jurisdictions everywhere.
Numerous corporations have started using their portfolios of poorly defined software patents to prevent rivals from entering the market. Witness the series of lengthy legal disputes between Apple and Samsung. Too often, it seems, software patents operate not as an incentive for innovation, but as a barrier to entry and a tax on new-product development. This is not what patent law was intended to do. Although the USPTO has been mainly at fault, America's courts have not helped.
One of the worst offenders has been the US Court of Appeals for the Federal Circuit, the appellate court that rules on patent disputes, among countless other things. Unlike its specialised counterparts in Europe and Japan, the Federal Circuit, with its grab-bag remit, has never acquired expertise in patent jurisprudence. As a consequence, it has issued some bizarre software rulings.
A cornerstone of any patent system is that laws of nature, natural phenomena and abstract ideas cannot be patented. Were that the case, all manner of monopolies could spring up based on common ideas - such as boiling water to make tea - that could feasibly be used to prevent others from doing the same, or at least require them to pay a licence fee. Historically, the courts have viewed software inventions as far too abstract to qualify for a patent.
That changed in 1998, when the Federal Circuit found that a business method (in an action known as the State Street Bank case) that ゛produced useful, concrete and tangible results゛was eligible for patent protection. A flood of equally dubious patents quickly followed.
A decade later, in the Bilski case, the Supreme Court revisited the issue of whether a method for doing business could be patented. It found the ゛invention゛(a way of hedging commodity risks) too abstract to patentable, but failed to provide a clear test of what constituted an abstract idea for judges in lower courts to apply.
In a much-watched patent dipute dating back to 2007 between Alice Corporation and CLS Bank, involving a way to avoid settlement risks when closing financial transactions, a lower court found the idea too abstract to be patented. But on appeal, two federal judges disagreed, arguing that the implementing the invention on a computer rendered it non-abstract - and thus patentable.
A ten-member panel of the Federal Circuit subsequently agreed to rehear the case. Unfortunately, their opinion, issued in May 2013, has left the situation more confusing than ever. Between them, the ten judges issued no fewer than seven different opinions. A majority agreed that the patent in question was ineligible. But one dissenter on the panel, Judge Kimberley Moore, warned that if the patent were ruled invalid, it would cause the death of hundreds of thousands of similar patents for business methods, financial systems and software, and would ゛decimate the electronics and software industries゛.
As the Supreme Court now intends to review Alice v CLS Bank, it will finally confront the most fundamental of issues in patent law today： whether or not software patents are impermissibly abstract. A ruling is expected by July 2014.
Would it matter if software patents were judged too abstract to warrant patent protection? Despite Judge Moore's misgivings, patent isstance is a poor measure of innovation. Patenting is strictly a metric of invention. Innovation is such a vastly different endeavour - in terms of investment, time and the human resources required - as to be virtually unrelated to invention.
Indeed, many innovators have argued that electronics and software industries would flourish if companies trying to bring new technology (software innovations included) to market did not have to worry about being sued for infringing thousands of absurd patents at every turn. A perfectly adequate means of protecting and rewarding software developers for their ingenuity has existed for over 300 years. It is called copyright.
