These are interesting times for patent lawyers.
The U.S. Patent and Trademark Office has proposed new transparency rules and will soon run a pilot program to introduce glossaries to patent specifications. Plus, the Supreme Court has heard six patent cases this term. These include blockbusters like Alice v. CLS Bank, with the court considering the patentability of computer-implemented inventions, and Nautilus v. Biosig Instruments, to determine when a patent is invalid for being overly vague.
To appreciate how unusual this is, the Supreme Court has more patent cases this term than during the entire 1980s. Congress is also active. In December, the House of Representatives passed the Innovation Act by an overwhelming 325-91 margin. The Senate is currently hashing out its version of patent reform and President Barack Obama has said he is eager to sign a bill.
How did we get here? Advocates for patent reform like the Electronic Frontier Foundation would like to believe we sparked this movement. But the reality is patent trolls themselves deserve most of the credit. By filing so many suits and threatening so many end-users of technology including burger joints, cafés, funeral homes, podcasters and supermarkets, patent trolls created a crisis that demanded a response.
Opponents of reform insist that the rise in patent trolling is a ‘myth.’ They are wrong. Multiple studies, such as those from professors James Bessen and Michael Meurer, Brian Love, Colleen Chien, and Robin Feldman, confirm that the last decade has seen a massive and sustained rise in trolling. Unified Patents recently released a snapshot of February’s patent litigation data that confirmed trolls are bringing more than 50 percent of patent cases, with 82 percent of the suits against high-tech defendants. The patent troll explosion is real, and it won’t go away on its own.
So how should we respond? Recent legislative proposals have focused on procedural tweaks to the litigation system. It is very cheap to file a bare-bones patent complaint. Under current Federal Circuit law, plaintiffs are not even required to specify which claims they are asserting or which products allegedly infringe. Yet a defendant can easily burn through a million dollars defending the suit. So, even if the underlying case is weak, trolls can use the cost of defense to pressure defendants to settle.
The Innovation Act passed by the House attempts to fix this imbalance through heightened pleading requirements and fee shifting. The bill would also reduce discovery costs by limiting requests to certain core documents. And it provides some relief to end-users by allowing them to stay their cases if the manufacturer is willing to step in. We hope that any Senate bill will include similar provisions. These kinds of procedural reforms should make patent trolling a less attractive business model. They would be especially effective against the bottom-feeder patent trolls bringing weak cases to extort settlements well below the cost of defense.
The main drawback with current legislative proposals is that they fail to address the root cause of problems in the patent system: the flood of low quality software patents. Patent trolling is fueled by poor quality patents with vague claims.
This explosion of software patents is also behind a wasteful and counterproductive arms race between operating companies. In 2011, both Apple and Google spent more on patent litigation and acquisitions than on research and development. Patents are supposed to promote R&D, not supplant it.
There are many reasons why software patents tend to be of low quality. First, the PTO does not have the time and resources to adequately determine validity. With only a few hours to spend on each application, it is impossible to expect examiners to find the most relevant prior art. Massive repositories of software—like GitHub—are mostly ignored.
The patent office also does not have the resources to confirm basic facts like whether the applicant really did the work he or she claims. It recently awarded a patent for a method of preparing stem cells even though the underlying work was widely known to be fraudulent; the purported inventor had even spent time in jail. As James Grimmelmann, a law professor at the University of Maryland, observed: “The USPTO is an armory handing out legal howitzers on the honor system. What could possibly go wrong?”
The second major problem with software patents is vague claims. Software lacks the universal lexicon of other fields. A chemical molecule can be described in a way that will be instantly understood by all experts in the field. But a computer program can be described at many different levels of abstraction.
Even worse, software patents often claim functions rather than specific solutions (e.g. “a computer processor configured to do X”). We would never tolerate pharmaceutical patents claimed this way (e.g. “a molecule configured to treat disease X”). The lax standards applied to software has led to a huge number of patents that are both overbroad and vague.
To fix patent quality, we need reform in the courts, legislature and at the USPTO. First, to deal with bad patents that have already issued, Congress should expand opportunities to challenge software patents more cheaply at the patent office; the original draft of the Innovation Act included such a provision. And Congress should empower the PTO to reject low quality applications.
While more resources for examination would help, it will not be enough to simply throw money at the problem. We also need to remove the incentives—like the fact that applicants get unlimited do-overs in the form of continuation applications—that make it hard for the USPTO to police bad patents.
The judiciary can do its part simply by diligently enforcing the Patent Act’s existing provisions. When applying section 112(f)’s limits on functional claiming, courts should look at the reality of how software is claimed, rather than checking to see if the claims include a few easily avoided magic words. And the Supreme Court has an opportunity right now in Nautilus v. Biosig Instruments to demand far more clarity in patent claims. In the other major case of the term, Alice v. CLS Bank, the Supreme Court should hold, at the very least, that adding ‘on a computer’ or ‘on the Internet’ to otherwise abstract ideas is not enough to confer patent eligibility.
All three branches of government have a role to play in fixing the patent system. Only a combination of reforms—ranging from improving the PTO’s prior art searches, to legislation clamping down on litigation abuse, to judicial decisions promoting claim clarity—will bring meaningful change.
Daniel Nazer is a staff attorney and Mark Cuban Chair to Eliminate Stupid Patents at the Electronic Frontier Foundation.