Dealing with mad patent disease

Upside-down world

Indeed, mad patent disease has flipped the patent system on its head. It is no longer a method for protecting and publishing innovations so others can find, license and use them. Now it has become a numbers game in which the biggest holder wins and few have an incentive to read--let alone use--the innovations captured in a patent.

Intel Corp., the world's biggest chip maker, estimates that as many as 600,000 existing U.S. patents relate in some way to its business. That number is growing by as much 80,000 a year, the company believes.

"If we have to start reading through nearly 100,000 patents a year, I won't even be able to hire enough patent attorneys to do it," said David Simon, chief patent counsel for Intel.

In this climate, many fear being charged with willfully infringing patents or omitting prior art in patent applications, a charge known as inequitable conduct. So Intel and other companies have put strict procedures in place to control which patents its engineers can read.

Steve Perlman, a serial entrepreneur in San Francisco, tells a story about how inane the corporate IP lockdown can be. Gemstar, which owns a set of patents around electronic program guides, threatened an infringement suit against Perlman's mid-1990s startup WebTV Networks. So Perlman spent a weekend reading through Gemstar's patents. He found only one that related to the company's technology, and he was able to figure out a work-around so WebTV did not infringe the patent.

When Microsoft Corp. later acquired his company, Perlman was told Microsoft had licensed the Gemstar patents so it would have freedom to operate in that space. Microsoft executives insisted that the WebTV box use the license to ensure it would face no threat of a patent suit--even though Perlman had already done the engineering work to ensure it would not.

Mark Lemley, a professor at Stanford Law School, characterizes the problem crisply. "Both researchers and companies in component industries simply ignore patents," Lemley stated in an article published last year in the Michigan State Law Review.

In its current state, the Patent Reform Act of 2009 (S.515) won't cure the disease. The proposed bill tries to limit willful infringement, but it's unlikely those limits would be enough to inspire companies to lift their restrictions on which patents their engineers can read. The bill also tries to limit use of inequitable conduct, a provision that lets third parties essentially kill a patent if they can show the applicant knew about but did not reveal some prior art.

Other provisions seek to limit damage awards by restricting so-called rocket dockets (such as the Eastern District of Texas), which tend to favor plaintiffs in patent infringement suits. One much-debated provision asks judges to provide specific guidance to juries about awards based on past case law.

Meanwhile, some observers believe two provisions in the legislation could exacerbate mad patent disease.

For one, the bill would shift the system from granting patents to the first to invent a technology to acknowledging the first to file a patent on it. The provision is geared to bringing the U.S. system more in line with those in Europe and Asia. But some say the net effect of such a shift would be a rush to file applications that ultimately would have little or no commercial significance.

The legislation also would redefine the process for challenging patents after they have been granted. Some say the new provisions could overwhelm the U.S. Patent and Trademark Office with added casework and create more uncertainty about whether a patent will remain valid.

The bill does not directly address the fact the patent office is already overwhelmed, bogged down in nearly half a million new applications a year and a backlog of about 750,000 existing applications.

So what's an engineer to do? As is the case with any political process, the first steps are to get informed, get engaged and let decision makers know your views.

Various sides are already jostling to amend the draft bill that proponents hope to bring up for a vote to the full Senate soon. If it goes through, the House of Representatives will need to pass a companion bill.

Speaking with one voice

Letting representatives know how you feel as an individual is one small step. Organizing into a more formal coalition is another, bigger one.

Engineers lack an organized voice on patent reform, according to entrepreneur Perlman. Technology business lobbying organization TechNet takes a corporate view, the IEEE is divided on the topic, and other coalitions are split along industry lines.

Engineers also need to ensure they are heard within their companies. In the rush for numbers, patent quality has gone out the window. Many blame the overwhelmed patent office, which has hired 1,200 patent examiners a year for the past three years and has created a new training academy to try to catch up.

Companies themselves could be better arbiters of quality, and they could start by giving engineers better compensation for their work on patents. Rather than pay inventors an honorarium (the typical case today), companies could tie patent compensation more closely to the real licensing revenues those patents generate. The latter approach is already in practice in Germany and Japan.

Engineers should also be encouraged to be involved in the patent application process, keeping an eye on the technical accuracy of the work defining claims and finding prior art.

In addition, Kevin Gillette, the outgoing chairman of the patent office's Patent Public Advisory Committee, has invited engineers to get involved in that group's work. The PPAC has published 17 recommendations for improving the patent process, including mandating an interview with the applicant so the examiner can better understand the technology for which the patent is sought.

Such actions are not a cure in themselves, but they are steps in the right direction. To get to a cure, engineers might have to accept some hard concessions.

In his article, Lemley of Stanford sketched out several big leaps he believes it would take to get back to a system where engineers proactively read, license and use patents.

First, the patent office would need to process and publish applications much more quickly so they are commercially relevant. To do that, applicants would probably have to accept strict limits on the number of claims and continuations they can file.

A recent court decision has cleared the way for the patent office to do just that, but the move has been a controver– sial proposal.

In addition, Lemley said, applicants might have to agree to a simplified challenge process that establishes the scope of their patent rights early on. And they should be required to publish terms of licensing for their technology, he added.

Finally, the Stanford professor suggests the law grant a kind of immunity against infringement to companies already using patented technology if they developed it independently. Such companies might still owe a reasonable royalty, he added.

But such steps are "radical, and it seems unlikely that they will happen," Lemley concluded in his article.

If engineers raise a strong voice on such measures, it could make a difference. There's a broad consensus emerging that patent madness has gotten out of hand, and a growing group of people wants a system in which individual patents have real value again for owners and users alike.