U.S. District Judge James F. Holderman has been assigned 101 patent cases since 2000. For judges wary of today’s latest innovations, Holderman’s centennial must feel cringe-worthy. To him, it may as well have been a celebration.
“I love this stuff,” Holderman said.
Many law professors and would-be patent law reformers point to experience among trial court judges — like that of Holderman — as the best shot to create a system that protects and incentivizes the so-called knowledge economy of the 21st century.
The former chief judge of the Northern District of Illinois, Holderman has traveled to China to teach and to discuss patents with that country’s State Intellectual Property Office. He is an IP law lecturer at the University of Illinois College of Law. He was president of the Chicago chapter of the patent law-focused bar group, The Richard Linn American Inn of Court. The Intellectual Property Law Association of Chicago has honored him for his work on patent and IP cases.
The complexity of today’s technology and the increasing pace of innovation has stressed the U.S. patent system, raising questions over whether the general patent law framework — including its use of generalist trial court judges and a specialized appeals court — is well-suited to handle the leaps and bounds made by inventors today and in the future.
“Is technology getting more complex? There is no question about that,” Holderman said. “It’s not only getting more complex, it’s getting more complex at a faster pace.”
That’s why in 2011, Holderman opted in to the Patent Pilot Program, which allows judges in 14 district courts to choose whether they will take patent cases. The idea is that by becoming more familiar with patent law, trial court judges will make better rulings about what new inventions are patentable, how much those patents are worth or how to guide the parties to settlement more quickly. It is a nod to the idea that patent law is somehow more difficult to interpret and apply than other areas.
Since the inception of the pilot in 2011, the median time to end a patent case in the Northern District of Illinois decreased by 61 days, according to data from patent analytics firm Lex Machina.
Despite the quicker cases, there remains a healthy debate among judges, professors and practitioners about how effective a patent system can be when it asks generalist judges to decide hyper-technical questions. In the absence of wide-ranging patent reform, some professors believe the prescription for a healthy patent system — and, in turn, a strong economy — is to get more judges to approach such cases with an attitude like that of Holderman.
In their book, “The Patent Crisis and How the Courts Can Solve It,” Dan Burk and Mark Lemley, professors at the University of California at Irvine School of Law and Stanford Law School, respectively, argue that trial court judges are best suited to apply patent laws to evolving technology. The question is whether they are bold enough to wield powers in the statute that they say are going unused.
“You have to have a statute that is flexible enough to adapt to all kinds of technologies, (and) we think the core of the patent statute does that and is very sound,” Burk said.
“But to really do this right, judges have to look at each case, assess the technology that comes its way and say, ‘How do I make the patent statute work for that?’”
Tension, technology and the law
The friction created by rapidly evolving technology and the technology-agnostic U.S. patent system has not gone unnoticed by Congress or the courts.
On Oct. 1, 1982, the U.S. Court of Appeals for the Federal Circuit was created to hear all patent appeals. This was an effort to create legal consistency across the country, but it also bred patent specialization in the justices. Like the Federal Circuit, the Patent Pilot Program stems from the idea that familiarity with patent law will breed better or faster results.
After largely ignoring patent law for the better part of two decades, the U.S. Supreme Court has been busy rejecting what has grown out of the Federal Circuit. A study by patent lawyer Roy Hofer says 92 percent of patent-related Federal Circuit rulings were overturned as of 2008 — and it has fared no better since.
In the term that ended in July last year, the Supreme Court decided six patent cases — a record for one term, said Emory University School of Law’s Timothy R. Holbrook.
Not once did the Supreme Court side with the Federal Circuit in the cases Nautilus v. Biosig Instruments, Octane Fitness v. ICON Health & Fitness, Highmark v. Allcare Health Management System, Alice Corp. v. CLS International Bank, Limelight v. Akamai and Teva v. Sandoz. With the exception of the Teva case’s 7-2 decision, all the rulings were unanimous.
As technology builds upon itself, some say one particular burden on trial court judges is the confusing (at least to the layman) requirement that an invention be “non-obvious” to a “person having ordinary skill” in a particular field in order to receive a patent. It is difficult enough to know what you don’t know. The so-called PHOSITA standard asks judges, usually of a liberal arts background, to determine what a Ph.D. in chemistry or a software engineer doesn’t know.
One federal judge likened the PHOSITA concept to “an inside joke that I’m not in on,” according to Dave Schwartz, a patent law professor at IIT Chicago-Kent College of Law, who recalled the judge making that quip to a crowd at an IP roundtable discussion.
Such confusion is a problem because in the knowledge economy, a properly functioning patent system is arguably more important than ever. Make patents too strong, and the competition that drives innovation may be stifled. Too weak, and the inventor’s incentive to innovate will wane.
The more common critique of the current balance is that patent rights are too strong. That may be at least in part due to a generalist judge’s default position to assume a new or complex technology is patentable. A new bit of nanotechnology may be no more obvious to a generalist judge than what happens when he or she pushes “send” on a text message.
“An enormous amount is at stake,” said Jonathan Masur, a University of Chicago Law School professor. “And getting it wrong in either direction is a real problem.”
‘The law lags behind’
Before even addressing whether the patent law system is well-suited to today’s innovations, some question the idea that today’s technology is actually more difficult for the system to digest. It is not as if patent law was designed without considering advances in technology — that’s the very thing it attempts to govern.
Lawyers and academics frequently cite past success as indicative of similar results today. To the layman in the 1870s, Alexander Graham Bell’s telephone may have been just as mind-bending as the Internet was to the common man of the 1990s.
“The background knowledge grows more and more, and science advances,” said IIT Chicago-Kent’s Schwartz. “But when we look at any invention under the microscope, I think there is always something you can wrap your hands around and understand.”
Arthur Gollwitzer, a patent litigator at Michael Best & Friedrich, said courts can sometimes struggle with “bursts” of new technology, such as the dawn of the Internet age. But he disputes the idea that this creates a problem so fundamental that it requires a vast shift in policy, or that previous “bursts” haven’t caused similar problems.
“Can the patent system keep up with evolving technology? Absolutely,” he said. “Are there growing pains sometimes? Probably.”
University of Chicago’s Masur believes the court is mired in an extended period of indigestion. But he disagrees with Gollwitzer, believing reform is necessary.
“I think the courts first started encountering real problems coping with new technologies 20 or 30 years ago, and they have not caught up,” Masur said. “Technology has continued to get more complex, but the court hasn’t acquired any new tools.”
Prior to a Supreme Court ruling in 2013, patents were granted for 30 years to companies who isolated strands of DNA. To argue that gene sequencing was patentable, lawyers often relied on a ruling by New York federal district judge Learned Hand from 1911 — 40 years before DNA’s double-helix structure was deciphered.
That precedent fell out of favor in 2012 when Jon Harkness, a patent lawyer, dug into the century-old court documents to determine the lawyers appearing before Hand never argued over the ability to patent biological substances. Had the law reacted quicker, eliminating monoplies in this area, tests to determine genetic cancers or other diseases would have been drastically less expensive.
A wider variety of tests may even exist. Companies such as Myriad Genetics — whose patent was wiped out in the 2013 case — may have sought to develop more tests for more diseases in order to attain the same level of profits that their patent-granted monopoly allowed. A patent can be an incentive; the monopoly it grants an excuse for sloth.
Software is also a good example. It has been widespread since the 1980s, but questions over its patentability are just now reaching the Supreme Court. In Alice Corp. v. CLS Bank International, the court ruled last year that abstract ideas implemented using a computer aren’t patent-worthy. A brief, 17-page ruling that not once uses the word “software,” Alice has nonetheless cast a pall that is obscuring the value of patents in the area. It also led to a marked reduction in patent litigation in California last year. An earlier ruling could have avoided countless fees spent on litigation.
“The law lags behind,” said Brent Hawkins, a McDermott Will & Emery partner. “That’s the bigger issue, and that’s what handcuffs the court sometimes. It hadn’t fully contemplated certain types of technology being the subject of a patent.”
And while the Supreme Court seems to think the appellate court is confused on a number of patent law issues, the appellate court has a similar opinion of trial courts in a substantial number of cases.
A study by IIT Chicago-Kent’s Schwartz found that 32 percent of all claim-construction terms — the actual interpretation of what a patent applies to — were wrongly construed by trial court judges, according to Federal Circuit rulings.
Those results have not been tested against prior years, so it is impossible to say today’s technology is the reason so many cases are reversed. A rise in litigation over new types of technology, though, has occurred at the same time.
A study by Schwartz, Lemley and John Allison, expected to be published later this year, shows that nearly 50 percent of patent lawsuits today involve software and electronics. Chemical and biotech patents account for another 20 percent of the patents being litigated.
Experience at the trial level
A number of tweaks have been proposed as means toward achieving better results.
Some academics advocate the creation of a specialized patent trial court packed with tech-minded jurists. The professors argue they could draw better lines in the sands of invention — what deserves patent protection and what doesn’t.
While that idea has been adopted in Europe, it is far off here, in no small part due to some high practical hurdles — it’s unlikely that a budget or enough science-minded judges exist to populate such a court.
Attempts have been made to create different patent rules for different technologies, but those have mostly only proven that a patent statute must include flexibility to adapt to new inventions. For instance, Congress passed the Semiconductor Chip Protection Act in 1984 after that industry lobbied for stronger patent protection. One study says that rapid innovation in the field means the act provided “virtually no real chip protection.”
Meanwhile, others are hopeful about the Patent Pilot Program and the experience it fosters among trial court judges.
The Northern District of Illinois is one of 14 U.S. District Courts participating in the program. The program allows trial judges to opt in to a docket weighted more heavily with patent cases. Since the inception of the program, the median time it takes to end a patent case in the Northern District of Illinois decreased by roughly two months, according to data from Lex Machina, the patent analytics firm.
From Jan. 1, 2000, through Sept. 18, 2011, the median time to termination in a patent case was 236 days. That fell to 175 days from Sept. 19, 2011, through March 25 this year.
When presented with those numbers, Holderman shied away from attributing them to the pilot project, saying it was too early to tell the effect it had on the court. Still, he said he continues to support the idea that experience is highly valuable in patent law.
“I do think the developing of expertise among judges in the patent area is something that will ultimately be very beneficial to our country,” Holderman said. “Because it is a highly complex, unique area of the law. Antitrust or securities litigation are complicated, no question about it. But they don’t spring from a monopoly granted by the government to encourage innovation.”
The academics, however, are somewhat split on whether experience helps trial court judges reach better results.
A study by Schwartz says that judges do not get any better at claim construction — measured by reversal rates at the Federal Circuit — with more expertise. It is so idiosyncratic that a steep learning curve exists for each patent.
But a 2013 note in the Harvard Journal of Law & Technology disputes that idea, pointing to evidence that shows judges with more experience fare better on appeal. It says a district court judge that has had 126 patent-infringement suits (the 75th percentile of experience) has a 16 percent lower chance of being overruled than a judge with only 31 patent cases (the 25th percentile).
A case in Holderman’s courtroom is a good example of the tactics a skilled patent jurist can use to drive the parties to settlement.
An owner of Wi-Fi patents had sued hundreds of small business owners. The cases were consolidated in front of Holderman, and the makers of wireless routers, including Cisco Systems Inc., stepped in to defend and file a counterclaim on behalf of the coffee shops that were initially sued for infringing Wi-Fi-related patents by offering the service to customers.
The case featured what some consider the most distressing aspects of the so-called patent troll problem — a standard technology patent was used as an attempt to elicit small payments from thousands of business owners.
Holderman focused first on calculating damages. His analysis found the patents were worth 9.56 cents per Wi-Fi chip. That quickly led to a confidential settlement that ended the litigation.
Neither side sought appellate review, and the Federal Circuit separately pointed to the opinion as an example for calculating damages in standard technology cases. The case was In re Innovatio Ventures IP LLC, No. 11 C 9308
Holderman said he once thought specialized patent trial courts were a good idea. In a 2002 edition of The Journal of Law, Technology and Policy from the University of Illinois College of Law, Holderman wrote: “We need a specialized trial court unencumbered by the burdens and distractions faced by generalist judges.”
But he has since changed his mind. He thinks experience in patent trials will provide better results than judges who would exclusively hear patent cases.
“It is difficult for generalist judges, but if they have enough time and enough motivation, like anyone, generalist judges can learn to address highly technical material in a fair and just way,” Holderman said in an April interview.
In defense of the PHOSITA
But successful trial court rulings still largely depend on a very difficult task: Judges must determine whether slight innovations in complex technologies are unique enough to deserve a patent.
To do this, they compare the invention to what a hypothetical peer of the inventor would know at the time.
This is known as “the person having ordinary skill in the art.”
U. of C.’s Masur calls that task “almost impossible.”
“Imagine if you were sitting in a room listening to two experts in theoretical physics debate some point about theoretical physics,” Masur said. “You’d have no idea who was right. It’s almost impossible. And that’s not a valid basis for making a decision, and I think judges run into that problem all the time.”
In a separate interview, Burk countered: “That’s why they get the black robes, right?”
Burk, like others, compares PHOSITA rulings in patent cases to highly technical decisions judges must make in medical-malpractice cases. Additionally, he said it is difficult to envision something that replaces the PHOSITA standard. The flexibility it is designed to provide for new technologies is the crux of a good system, he said.
In Burk’s reading of the patent statute, the PHOSITA standard holds the key to allowing patent law to apply varying levels of protection to different types of technology and new innovations. It tells district court judges to treat different types of technology as their own separate field.
“The Federal Circuit has told us that the person of ordinary skill in biotechnology has a Ph.D.,” Burk said. “They haven’t said anything about software, but you can get some pretty good software out of two teenagers in a garage.”
Burk said the main impediment is judges who are reluctant to aggressively use the statute.
For instance, it allows judges to appoint their own expert — paid for by the parties — if they are having difficulty choosing between two persuasive ideas. But judges rarely do it, citing an adversarial judicial tradition in the U.S. that defers to information presented by the parties.
And judges could also make their own rulings about what the PHOSITA knew based on how much patent protection they believe an industry requires. If software companies plan to sell a product for six months before replacing it, it probably follows that the ideas behind the new technology are obvious to the average Stanford student and not deserving of a patent.
Burk’s advice to judges is simple: Get aggressive.
“It’s not popular to really get your hands dirty and use the statute to reach some outcomes, but we all know that that’s what judges do,” Burk said.
“The reality is, Congress has given the courts the tools to make this work. It requires some effort, but particularly if judges are willing to use tools they otherwise don’t use, like independent experts, the statute can work a lot better.”
It will only be clear if “better” is good enough the next time a revolutionary technology comes down the line, which is perhaps the one certainty in the patent reform debate.