A few weeks ago we interviewed Julie Samuels, the Mark Cuban Chair to Eliminate Stupid Patents at the Electronic Frontier Foundation (EFF), about software patents and patent trolls. Our conversation focused both on why we have a software problem and how the patent trolls are a part of the A generalized term referring to patents whose subject matter extends to computer-implemented code, which have been the subject of great controversy, including but not limited to how they interact with open source software. Although software patents are often denigrated, there is no accepted definition. However, there are a variety of methods for identifying software patents for empirical analysis. See Bessen, A issue. We also discussed potential solution to the problem including the SHIELD Act, Mark Lemley’s functional claiming proposal, shorter terms and some other ideas that have been proposed.
EFF recently launched Defend Innovation, a grassroots project focussed on fixing the patent system. It has seven calls for action:
- A patent covering software should be shorter: no more than five years from the application date.
- If the patent is invalid or there’s no infringement, the trolls should have to pay the legal fees.
- Patent applicants should be required to provide an example of running software code for each claim in the patent.
- Infringers should avoid liability if they indepently arrive at the patented invention.
- Patents and licenses should be public right away. Patent owners should be required to keep their public records up-to-date.
- The law should limit damages so that a patent owner can’t collect millions if the patent reflected only a tiny fraction of the defendant’s product.
- Congress should commission a study and hold hearings to examine whether software patents actually benefit our economy at all.
The transcript of the interview is below the fold.
Josh: EFF has been involved in patent reform efforts for a long time. However, over the last year, there seems to be more of a discussion on software patents specifically, and problems in the system. Why do you think the focus on software patents is starting to get so much attention?
Julie: I think a few things have happened that really brought the issue front and center. First, the trolls started targeting the little guys; all of a sudden it wasn’t just big companies suing each other in court. We saw trolls go after individual app developers, cafes, nursing homes, and growing start-ups. Among other things, this made the issue “good tv”.
Second, last year’s successful defeat of SOPA and PIPA really moved tech policy concerns up on many people’s radars–particularly among policymakers and others on Capitol Hill. All of a sudden, people who hadn’t given much thought to what doesn’t work in the tech community started paying attention, and in that community there seems to be some consensus that software patents are the next battle.
Josh: You say it makes for “good tv” – do you have a specific example of that?
Julie: Sure. You have to look no further than This American Life (When Patents Attack!). That hour of radio did more for our work than almost anything I can remember.
Josh: That certainly was a great piece of journalism. You also mentioned in your answer that there is consensus this is the next battle – who are the types of groups and people that make up that consensus?
Julie: Those people are the ones who are actually inventing and creating. Every software engineer I talk to abhors software patents and wishes they would get out of the way of innovation. And it’s not just the engineers and the tinkerers, it’s the small business owners–the people best positioned to grow our economy and create jobs. Sure, there are lots of policymakers, professors, and lawyers (yours truly included) calling for change, but we are just bearing the message from those who really feel the negative effects of a broken patent system and are stuck paying a misguided tax on innovation.
Josh: It’s interesting so many software year engineers are so passionate about this issue but yet work for companies that file the patents and perpetuate the problem. What do you think is the cause of the disconnect?
Julie: Not enough engineers in the legal departments! Seriously, what’s happening is legal departments feel pressure to “patent up” to protect the company from litigation. We’re actually seeing a second wave of this kind of behavior, and it’s really dangerous. It creates incentives to get as many patents as possible, often times regardless of their quality. And when those companies hit hard financial times, those patents end up on the open market, where they are often picked up by trolls.
Josh: Patent quality — that is a good segue to the bigger issue — what do you see as the current problems with software patents?
Julie: First, the patent system shouldn’t be one-size-fits-all. That just doesn’t make sense. Different kinds of technology should be treated differently. For example, look at pharmaceuticals: it often can cost up to $1 billion to develop a drug, get it through the FDA, and bring it to market. So you can understand why, as a society, we’d want to incentivize that kind of research and development.
But then think about software. It’s fundamentally different. It doesn’t take the same kind of financial outlay or investment of time. It’s not to say that coding is easy, but we’ve seen no proof that a 20-year government-granted monopoly is necessary to incentivize development. Many of the largest software companies never relied on software patents to grow their businesses–look at Google, Facebook, and Twitter, for instance.
Which brings me to a larger point that I really wish we talked about more. The patent system is in our Constitution. It’s clear what the Founders intended: a system that would incentivize innovation for society. I think that we need to take a step back and consider our current laws and policy to determine if that’s how the system currently works. I would argue it does not.
Josh: There is a lot in that answer. You mention the difference between software/Internet and pharmaceutical. We have estimated the amount of patents in a smartphone to be approximately 250,000. The life sciences are very different. Can you discuss this more.
Julie: That’s an important point. While approximately 250,000 patents may cover a smartphone, only 20-30 patents usually cover a pharmaceutical invention. Think about that difference: 20 or 30 patents is still a lot to navigate, but it’s doable. How could a company (or an individual!) ever know what is in 250,000 patents? How could they make risk assessments about growing their businesses and releasing new products? It’s completely unfeasible. Tim Lee and Christina Mulligan wrote a great article about this problem.
Josh: Why is that such an important difference — what are the risks created by that uncertainty? A lot of people don’t necessarily understand that.
Julie: You have to think for a second about one of the most important elements of the patent system: it’s called the notice function. The basic idea is that in return for a 20-year monopoly, an inventor must dedicate his invention to society and explain the best way to practice that invention. By clearly explaining it, the inventor also puts the rest of the world on notice as to what people can do without infringing anyone’s patent rights. So if there are, let’s say, 20-30 patents covering a product, an inventor (or her lawyer) can read those patents, hopefully understand them, and make informed decisions about what she can make, do, or sell without risking infringement. But that becomes impossible when you start talking about hundreds of thousands of patents, particularly when those patents are software patents, which are notoriously broad and vague and therefore hard to understand.
Many companies choose to just ignore the patents. It’s certainly easier and cheaper than the alternative.
Josh: Which leads to the inevitable An entity in the business of being infringed — by analogy to the mythological troll that exacted payments from the unwary. Cf. NPE, PAE, PME. See Reitzig and Henkel, Patent Trolls, the Sustainability of ‘Locking-in-to-Extort’ Strategies, and Implications for Innovating Firms. suits and shakedowns.
Julie: Exactly. But to be clear, even if a company had the resources to review all of the potential patents that cover its products, it likely would still face the trolls suits (or threats of suits). The patents that many of the trolls assert are so broad and vague, or don’t even come close to covering the products at issue. But it’s still cheaper to settle than go to court.
Josh: Switching gears a little. Looking at the broader A generalized term referring to patents whose subject matter extends to computer-implemented code, which have been the subject of great controversy, including but not limited to how they interact with open source software. Although software patents are often denigrated, there is no accepted definition. However, there are a variety of methods for identifying software patents for empirical analysis. See Bessen, A problem — what do you see as some ways of solving it?
Julie: There are lots of potential fixes out there that would solve the problem. One answer that’s already floating around Congress is a fee-shifting bill called the SHIELD Act. Right now, it’s really easy for the trolls to bring lawsuits and really hard and expensive for potential defendants to fight back. The SHIELD Act would allow courts to make the loser pay for the costs of the suit, which should deter some of the worst lawsuits. We’re in favor of anything that makes the An entity in the business of being infringed — by analogy to the mythological troll that exacted payments from the unwary. Cf. NPE, PAE, PME. See Reitzig and Henkel, Patent Trolls, the Sustainability of ‘Locking-in-to-Extort’ Strategies, and Implications for Innovating Firms. business model less attractive, and the SHIELD Act would go a long way toward that.
But changes won’t just come from Congress. Courts and the Patent Office have an impact, too. There’s a positive movement to fix what’s called “functional claiming“, which would essentially require a patentee to claim its specific invention and only that, limiting the negative effects that patent can have down the road. If the patents were less broad, we’d all be better off.
There are also some promising self-help remedies out there. One example is the Defensive Patent License (DPL), a scheme that would allow parties to come together and dedicate their patents to defensive uses only.Twitter also has proposed a promising fix, its Innovators Patent Agreement (IPA), a pact between Twitter and its engineers that the patents Twitter gets from its employees will never be used offensively, no matter whose hands they end up in down the road.
Josh: You mentioned functional claiming and over-broadness. Over-broadness seems to be a running theme throughout this debate. Ultimately, if patents keep on getting issued on things that were not intended to be patented, we will be stuck with a broken system. Beyond functional claiming, are there other ways to get at over-broadness in your mind?
Julie: Functional claiming is an elegant fix because it’s already in the law, which is to say that the law already requires that the patents be less broad, but for some reason, the Patent Office and courts have let that requirement slide some, at least with regard to software and Internet patents. Of course, Congress could fix this by introducing legislation reinforcing that the patents should be less broad, and cover only what the invention covers. But there are other solutions, too, that don’t just have to do with the broad nature of the patents. For instance, their lifespan could be shorter. Or innocent infringers–those who came to the invention independently–could have a defense against claims of infringement.
Josh: What are your favorite examples of overbroad patents that have created problems in the wrong hands?
Julie: There’s one particularly egregious one that’s working its way through the court system right now in a case called Ultramercial v. Hulu. The patent there covers watching a video on the Internet before you can see copyrighted content. That’s it! To make matters worse, the See CAFC admitted that the idea was in theory too abstract to be patented, but nonetheless held the patent valid because, essentially, it would be really “hard” to make that work on the Internet. This even though the patentee never explained how it would actually practice the invention on the Internet!
Josh: As we look at the broader A generalized term referring to patents whose subject matter extends to computer-implemented code, which have been the subject of great controversy, including but not limited to how they interact with open source software. Although software patents are often denigrated, there is no accepted definition. However, there are a variety of methods for identifying software patents for empirical analysis. See Bessen, A problem, one of the issues critics of the reform movement raise is how do deal with patents on things where software is a component without invalidating the whole patent. How do you respond to that?
Julie: If the invention is limited to a specific software operation tied to a particular component, I’m less concerned about it. Because, of course, then its downstream impact would be less dangerous (and again, more limited).
Josh: Are there any current court cases (like CLS) where you see an opportunity for the Supreme Court or Fed Circuit to reign in software patents?
Julie: CLS Bank v. Alice is definitely the case we’re watching the most closely right now, particularly because the full See CAFC is hearing it (as opposed to the more typical three-judge panel). The case I discussed before, Ultramercial, is also really important, though it appears we won’t hear anything on that until CLS Bank is decided. (What’s interesting about Ultramercial is that the Supreme Court already asked the See CAFC to rehear it in light of the recent unanimous ruling in Myriad v. Prometheus).
Josh: We are running out of time. So, how would you like to end this?
Julie: First, thanks for this opportunity, it’s been a lot of fun. Second, and most important, is that we’re at a crucial moment in this fight. Policymakers and the courts are paying attention. There is an appetite to fix these problems. I’d encourage people to get involved, to follow the work we’re doing at Defend Innovation and the work you’re doing at Patent Progress. Also, I’d ask people to email their members of Congress and ask them to support the SHIELD Act, which they can do by clicking here.