This week, the Supreme Court handed down a decision that offers a glimmer of hope to Vonage, the Internet telephony pioneer that is fighting for its life after losing a patent lawsuit against Verizon. Vonage pioneered the Internet telephony market and has enticed more than two million customers away from Verizon and other telephone industry incumbents. But, although Verizon hasn’t been able to stop Vonage’s momentum in the marketplace, they’ve found more success in the courtroom. In March, a Virginia jury ordered Vonage to pay Verizon $58 million for infringing three Verizon patents.
The case raises some troubling questions about America’s patent system, which seems to allow a deep-pocketed incumbent to drive an innovative competitor out of business. Vonage could go under if the verdict is not overturned, because it would most likely be impossible to continue offering its core products without Verizon’s permission. But on Monday, in the case of KSR v. Teleflex, the Supreme Court made it easier for defendants to challenge obvious patents. The law does not allow patents on obvious inventions, but in recent years the courts have not adequately defended this limitation on patentability. Given the extraordinary breadth of Verizon’s patents, and the fact that very similar technology had been developed by others prior to Verizon’s patent application, the Supreme Court’s decision may give Vonage the legal ammunition to get Verizon’s patents invalidated.
One of Verizon’s patents covers devices that allow users to connect to Internet telephony services wirelessly. The other two cover servers that translate between Internet addresses and phone numbers. The latter patents present a particularly difficult challenge for Vonage because they strike at the core of Vonage’s products. It’s impossible to offer a full-featured Internet telephony product without translating between Internet addresses and phone numbers, so if the courts uphold the Virginia judge’s broad interpretation of those patents, it could give Verizon the legal ammunition to shut down not just Vonage but most of its other Internet telephony competitors as well.
Theoretically, the patent office is only supposed to award patents for “non-obvious” inventions, and the concept of converting between an IP address and a phone number certainly seems obvious. In an ideal world, the patents that were issued would be narrow enough that companies could “invent around” others’ patents if they were unable to come to acceptable licensing terms.
Unfortunately, our patent system has long since departed from this ideal. In recent decades, the courts have dramatically lowered the bar for obviousness. As a result, some of the patents being granted are so broad that inventing around them is practically impossible. The patents that allowed NTP to win a $612 million settlement from BlackBerry maker Research in Motion, for example, essentially covered the concept of transmitting new email notifications wirelessly. There’s no way RIM could have “invented around” that.
Technology companies have responded to this proliferation of bad patents by engaging in the patent equivalent of nuclear stockpiling. By obtaining dozens, hundreds, or even thousands of patents, a company can develop a credible deterrent against patent lawsuits: if someone sues it for patent infringement, it will be able to find a patent the other company has infringed and counter sue. Vonage’s fundamental mistake was that it chose not to join this arms race. As a result, when Verizon sued, Vonage was completely defenseless.
Software patents are particularly ripe for abuse because software is assembled from modular components. If the patent system allows those components to be patented, it becomes almost impossible to develop a software product without infringing numerous patents. Moreover, because of the complexity of software, it is often prohibitively expensive to even find all the patents a given software product might be infringing. So even a software firm that wanted to find and license all of the patents relevant to its products would likely be unable to do so.
Because of the high cost and uncertainty of the patent system, most software companies don’t even try to find patents they might be infringing. Instead, they sign cross-licensing deals with as many companies as possible, and they pray that the remaining companies won’t sue them before they’ve had time to develop a patent war-chest of their own.
This is great for patent lawyers, but it’s not clear how it promotes “the progress of science and useful arts” as called for by the Constitution. The genius of the free-market system is that companies compete for customers based on the quality of their products, and not legal advantages. Companies are free to introduce new products without having to ask anyone’s permission. Patents on obvious ideas undermine this competition. If the courts uphold Verizon’s broad interpretation of its patents, no one will be permitted to participate in the Internet telephony market without Verizon’s permission. Not only will consumers be harmed through higher prices, but the pace of progress may also be slowed, as potential innovators are locked out entirely.
In Monday’s decision, the Supreme Court reinvigorated the obviousness requirement, holding that “granting patent protection to advances that would occur in the ordinary course without real innovation retards progress.” Verizon’s patent certainly seems to fit that pattern. The Internet’s function is to transmit data, and to a computer, sound is just another type of data. So it was only a matter of time before the Internet became fast enough to support widespread use of Internet telephony. And the specific process described in Verizon’s patents-translating between a telephone number and an Internet address-is an obvious step in the process of exchanging data between the old-fashioned telephone network and the Internet.
A more fundamental reform would be for Congress to amend patent law to clearly state that software is not eligible for patent protection. The de facto patentability of software is a recent judicial innovation, and there’s growing evidence that it’s been bad for the software industry. Congress should put an end to this ill-considered experiment by reinstating the rule that software is not eligible for patent protection. The software industry was extremely innovative in the decades before the legalization of software patents, and there’s every reason to think that eliminating software patents would have beneficial effects.
Patent stockpiling is a wasteful and counterproductive form of competition. If the courts or Congress don’t do more to rein them in, we’re likely to see a lot more cases of software companies being forced to spend their resources on patent lawyers instead of engineers. The Supreme Court’s Teleflex decision gives lower courts the opportunity to rein in such abuses of the patent system, but the lower courts must now seize the opportunity. And if they don’t, Congress should step in to address the problem.