Few debates over intellectual property protections have whipped up as much feeling as the recent battle over the European software patents directive.
When, in early July, the European parliament voted by 648 to 14 to reject the so-called Computer Implemented Inventions Directive, most open-source software companies cheered.
Although the intention behind the directive was to harmonise the approaches to patenting software-related inventions taken in different member states, some pressure groups feared it could represent a slippery slope towards making all software patentable – as, broadly-speaking, occurs in the US.
As a result, the directive developed into a vociferous battleground between the open source movement and those who claimed that lack of effective protection for “computer-implemented inventions” would stymie research and development across Europe.
After several years of argument and lobbying, it all proved too much for European parliamentarians. The directive failed to make headway and lawyers on both sides of the debate breathed a sigh of relief. “No directive is better than a bad directive,” commented one.
But what has been left behind in Europe, say patent specialists, is a fragmented regime with national bodies and the European Patent Office applying a variety of different tests when it comes to the patenting of software-related inventions.
The EPO’s approach, says John Collins at Marks Clerk, a London law firm specialising in patents, is based around assessing whether a product is automatically excluded from patentability because it falls within certain categories. It then considers whether the product is new and inventive and, in the case of a computer-related invention, represents a “technical solution to a technical problem”.
In some member states, however, patent offices approach the same question slightly differently. For example, they may apply different sets of exclusions, or in the case of the UK, focus first on the claimed “invention” itself, consider the prior art, and so on.
Lawyers worry about the potential results. “The courts say that this leads to the same result. But my concern is that if you have two different ways of getting to the goal, you may not end up in the same spot,” comments Mr Collins.
That, in turn, raises the prospect of “forum shopping”, in terms of where companies seek to fight their patent battles, and the unattractive possibility that decisions over a specific invention may be inconsistent within the EU.
Some lawyers are already murmuring about the need to resurrect the directive. But, as yet, no one seems to have quite recovered from the previous struggle. “In my view, the appetite has somewhat diminished,” says Mr Collins.
All this has been watched with some bemusement from the US, where any computer-related invention is patentable, provided it is new, useful and produces a concrete, tangible result.
This position stems partly from a landmark case in the early-1980s, when the US Supreme Court construed the patent statute to permit the patenting of computer programs if they were part of a process.
But it also owes much to a legal fight waged in the late-1990s by State Street. The financial services group complained that a competitor had infringed its patented idea for applying hub-and-spoke technology to mutual funds.
An appeal court agreed, deciding that the rule book’s special exemption for business processes was “an unwarranted encumbrance that should be discarded as error-prone, redundant and obsolete”.
The court also held that mathematical algorithms were patentable if their application produced a “useful, concrete and tangible result”. Not surprisingly, the judgment prompted a flurry of new patents.