An Examination of Software Patents

EDITORIAL NOTE: What follows are the prepared remarks of David Kappos, Under Secretary of of Commerce for Intellectual Property and Director of the United States Patent and Trademark Office, delivered at the Center for American Progress in Washington, DC, on November 20, 2012.  This is reprinted here with permission.  

Thank you, Winnie, for that kind introduction. Good morning, everyone. It’s great to be here at the Center for American Progress. I’m pleased to be able to talk about intellectual property and the role that intellectual property rights play in enabling innovative goods and services to come to market. And specifically, I’m going to focus my remarks on software patents and the so-called smartphone “patent wars,” which have become front page news in the last year or so.

It is increasingly clear that intellectual property, or IP, is a key driver of economic growth, exports, and job creation. IP rights are the global currency for creating value for products and services, for all innovators, in all markets. And the protection provided by patents is critical to the innovation ecosystem. In fact, last spring, the U.S. Commerce Department released a report that found IP-intensive industries support at least 40 million jobs and contributes more than $5 trillion to our economy, accounting for 35 percent of America’s gross domestic product. So it is in this context that we are seeing multi-billion dollar acquisitions of patent portfolios and a number of high profile patent lawsuits, involving some of the most innovative companies on the planet, who are producing some of the most popular technologies ever created.

The competition for market share in high-value sectors like smartphones and tablets is nothing new. These devices can have many patented components and features, containing a blend of software, hardware, and design patents. Yet software patents have tended to be the focus of controversy and some critics go so far as to argue that software shouldn’t be patentable at all.

Now, when we talk about software patents, what exactly do we mean? Let me be clear: patents aren’t issued merely for lines of code. Patents are issued for process and apparatus, which are determined to be novel and non-obvious. Patents are not granted for abstract ideas. But they are available to protect innovations, such as those enabling—automated language translation, voice recognition, and video compression, all involving major technological advances, all of which can be implemented in software. Because many breathtaking software-implemented innovations power our modern world, at levels of efficiency and performance unthinkable even just a few years ago, patent protection is every bit as well-deserved for software-implemented innovation as for the innovations that enabled man to fly, and before that for the innovations that enabled man to light the dark with electricity, and before that for the innovations that enabled the industrial revolution. Every bit as well-deserved. But it is equally important that patent protection be properly tailored in scope, so that programmers can write code and engineers can design devices without fear of unfounded accusations of infringement. And we know that inconsistency in software patent issuance causes uncertainty in the marketplace and can cause threats of litigation that in turn can stifle innovation and deter new market entrants.

At the USPTO, we recognize this as a serious concern, but we also observe that the various dire reports and commentary have omitted a critical component—the facts. So we decided to get the facts, undertaking our own study to look at the U.S. patents involved in some of the highest profile litigation among major firms in the smartphone industry. We found that in the vast majority of these cases, over 80 percent, the courts have construed the software patents at issue as valid. And an ongoing look at statistics in USPTO bear out a similar conclusion—rejections in software patent applications taken to our appeals board are upheld at a slightly higher rate than for the office as a whole, and those few decisions appealed to the Federal Circuit are affirmed 95 percent of the time. So to those commenting on the smart-phone patent wars with categorical statements that blame the “broken” system on bad software patents, I say—get the facts—they don’t support your position.

Now in fairness, we’ve struggled over the years at the USPTO with patentability determinations for software-implemented patent applications. And while the courts have consistently upheld the patentability of software as a general proposition, it is fair to say they have struggled as well, at all levels, and for many years, with various tests for patent eligibility and functional claiming and inventiveness standards. So with all of these challenges, should we treat software differently than hardware when it comes to patentability? No. Absolutely not and for several reasons.

Look at just a few of the incredible innovations this nation’s inventors have produced in recent years in which patented software played a role, equally as important as the hardware with which it interacts. We all take GPS for granted now; in fact, most of us rely on our smartphones for that service. Like the smartphone in which the technology is now housed, GPS technology involved many innovations integrated in hardware and software. Or take the da Vinci Surgical System: it’s a robotic arm that mimics a surgeon’s hand movements to the finest detail, allowing tiny incisions that ensure reduced tissue damage and quicker patient recovery. Thousands of these devices are in use today, and more than 200,000 surgeries have been performed with them.

These breakthrough products rely heavily on innovations in both hardware and software. Why would we tell the team working on a clamp that holds a scalpel their innovation is worthy of protection, but tell the programmers whose algorithm guides that clamp with unerring precision their innovation is not? And what amount of venture capital would that team of innovators raise if half of their innovation was free to be lifted by their competitors the moment they put it in the marketplace? Discrimination against a form of innovation that is increasingly critical to technological advancement, indeed that in many areas dominates technological advancement, makes no sense. So to those reporting and commenting on the smart-phone patent wars as if to suggest that the system is broken: let’s move beyond flippant rhetoric and instead engage in thoughtful discussion.

Software patents, like all patents, are a form of innovation currency. They are also ecosystem enablers, and job creators. The innovation protected by software patents is highly integrated with hardware. All of it must remain eligible for protection. The current software patent “war” is hardly the first patent war—and unlikely to be the last in our nation’s patent history. Whenever breakthrough technologies come onto the scene, market players find themselves joined in the marketplace by new entrants. The first instinct of the breakthrough innovators is to bring patents into play. This is not only understandable, it is appropriate. Those who invest in breakthrough innovation have a right to expect others to respect their resultant IP. However, in the end, as history has shown time and time again, the players ultimately end up agreeing to pro-consumer solutions via licenses, cross-licenses or joint development agreements allowing core technologies to be shared.

We saw this in the middle part of the 19th century with sewing machines. We saw it later in the 19th century with the telegraph and electricity. We saw it in the first part of the 20th century with airplanes. And we are seeing it again now with consumer electronics devices including not just smartphones, but tablets and game consoles. So I’ve explained why it’s important to have IP protection to encourage innovation in the highly integrated information technology area where many innovations include algorithms implemented in software. And I’ve provided a bit of history on how past patent conflicts have sorted themselves out.

So does that mean we’re done? End of speech? Should we just accept the problems, given the importance of the innovation and the illogic of discriminating against great technology that happens to be implemented in software? Of course not. The right point of inquiry is quality. By getting that right, we grant patents only for great algorithmic ideas worthy of protection, and not for everything else. This administration and its innovation agency understand that low-quality patents do no good for anyone. Low quality patents lead to disputes, uncertainty, and lost opportunity. Quality is central to our mission. All of this especially for software.

So we’ve been working on the underlying drivers for software patent quality from the beginning. Knowing that you get what you measure, starting in the summer of 2009 we assembled a taskforce to devise a comprehensive new set of quality metrics. That work culminated in 2010, when we rolled out the most complete, broad, objective patent quality measurement system on the planet – seven metrics, which we report eagerly to the public. The Lexus of quality metrics. And what do those metrics show? Patent quality isn’t broken at all. In fact, our decisions on both allowances and rejections correctly comply with all laws and regulations over 96 percent of the time.

Also in 2010, we strengthened our guidelines used by patent examiners to determine which inventions are eligible for patent protection. These are our so-called 101 Guidelines, which apply especially heavily in the software area. In 2011, we published our first ever comprehensive guidelines for review of claim clarity—an issue that comes up regularly with software patents. These so-called 112 Guidelines have measurably improved the clarity of patents in the software area. In fact, our nine-month review of these guidelines showed a 20+ percent increase in Section 112 rejections, even while an increased number of those issues have been resolved informally through a record 170,000 hours of interviews conducted with patent applicants in the last financial year. We’ve also given all of our examiners more time to review every patent application. This improves quality across the board, including of course for software patents. We have also reached out to experts in the software industry to provide technical training to our patent examiners, so they are up to date on the latest developments. More than 17,000 hours of technical training was provided to examiners in FY2012 alone, including 32 software-related courses presented by academic and industry experts.

Outreach will be even easier now with our Silicon Valley satellite office. We’ve recently hired a leader for that office, a renowned intellectual property attorney who is well-known and widely respected in Silicon Valley in the software field and beyond, Michelle Lee. So that’s what we at the USPTO have been doing to improve software patent quality.

And the courts are doing their part to improve quality as well. In particular, we have benefited greatly from the seminal 2007 Supreme Court case, KSR v. Teleflex, as well as subsequent decisions by the U.S. Court of Appeals for the Federal Circuit. KSR and later Federal Circuit cases have strengthened the legal doctrine of obviousness. In other words, they narrowed what is considered patentable. Software experts have long observed that programming is incremental in nature, with modest improvements not worthy of patent protection. KSR gave us the ability to recognize this valid observation and incorporate it in our examination process.

While we’ve made progress through court decisions, USPTO’s reexamination process is another method for challenging software patents. The reexamination process has proved to be a particularly effective tool in weeding out invalid patents with much less expense and disruption than litigation. It allows market participants to determine which patents merit a challenge, and, for these patents, the process provides a cost-effective resolution with the benefit of USPTO’s technical expertise. On two occasions during the last 3½ years we have taken a rather extraordinary step—conducting Director-ordered reexamination of issued patents. Both were software patents, and both were rejected upon re-examination. So these are steps we’ve taken over the past three-plus years. But now that the America Invents Act, or AIA, has gone into effect, we are empowered to do a lot more.

On September 16th of this year, we implemented most of the final rules for the AIA. Signed into law by President Obama a year earlier, the AIA is the most significant reform to the U.S. patent system since 1836. Many of its provisions are extremely well suited to software patents. For instance, third party submissions of prior art. Given the enormous amount of prior art information now available and its innumerable sources, it can be challenging for a patent examiner to find all the relevant prior art for every patent application. Worse still for software, where much prior art is in the form of previously written software, which is difficult to find and more difficult to understand unless you wrote it. And worse yet for software, where shifting terminology results in near-endless synonyms that frustrate even the most diligent searcher.

But now, under the AIA—for the first time in the history of our patent system—third parties can submit prior art to our examiners, referencing a specific application and including an explanation of how the prior art relates to the application. So now anyone can participate in the patent system and contribute to higher patent quality. We’re already seeing those third-party submissions coming in just two months after implementing this provision of the AIA. As of last Tuesday, we had received nearly 150 third-party submissions covering all types of technologies, including software.

And we are pleased to see innovative private sector efforts emerge to leverage this new AIA provision to further improve patent quality. One such initiative has already begun crowdsourcing searches for software prior art. It’s called Ask Patents and is an online network hosted by Stack Exchange, where software experts engage in robust discussions of possible prior art for given applications, then submit the best prior art along with helpful commentary. And finally, for all cases in which we receive third party submissions, we’re giving our examiners extra time in the examination process to carefully consider the submitted prior art and commentary. So the third-party submission provision of AIA has every prospect to improve the software patent landscape going forward. But what about patents already issued over the past 20-odd years?

The AIA gives us three new procedures to help manufacturers, service providers, and patentees confirm the validity of issued patents while weeding out overly broad ones. All three are handled by our new Patent Trial and Appeal Board, made up of administrative patent judges with phenomenal legal and scientific pedigrees. The new proceedings are known as post-grant opposition, inter partes review, and covered business method patents review. To start, each proceeding is statutorily mandated to complete in one year, which will save many millions of dollars in litigation costs and ensure resolution far faster than was possible previously either in the USPTO or district courts. One of those proceedings—post-grant opposition—allows third parties to challenge an issued patent on any ground, including basic eligibility and clarity, two areas of particular concern with software patents.

Another of the new proceedings focuses on business method patents. Since many of the more problematic software patents issued during the last two decades were drafted broadly to cover not just specific algorithms but the business problems they solve, we expect the business method review procedure to be very useful in addressing overbroad software patents issued in the past. In our rulemaking process, we implemented this provision to ensure its availability for just these kinds of patented software-implemented business processes. Moreover, since the business method review procedure sweeps in all grounds of patentability including statutory subject matter and clarity—areas of the law that have been tightened by the courts in recent years—it gives us the opportunity to address patents that would not have been granted under the current law.

So I think it’s fair to say the USPTO has done, and is doing a lot under the AIA and our own management processes, to address the challenges posed by problematic software patents. And bear in mind that the key provisions of the AIA have just gone into effect two months ago, and our initiatives within USPTO are just newly effective as well. So to the commentators declaring the system is “broken” I say: give it a rest already, and give the AIA a chance to work. Give it a chance to even get started. But we’re not done. Not nearly.

For starters, there are fresh legislative suggestions circulating. One suggestion, made by those who believe the USPTO should have more tools to review existing software patents, is to expand covered business method patent review process to include all software patents. Another legislative proposal aims at curbing speculative patent suits that leverage litigation cost to extract settlements based on borderline frivolous infringement claims. The SHIELD Act, legislation introduced in Congress earlier this year, would seek to deter such law suits by requiring the loser of an infringement case to pay court costs and attorney fees for the winner.

Beyond legislation, there remain gaps within the regulatory system that still need to be addressed. A root cause of problems with our current environment for software patents—and indeed all patents—is simply deciphering ownership. At the heart of a well-functioning innovation environment is accurate information about who owns what assets, so that license rights can be confirmed or sought, and unproductive effort simply avoided. By analogy, if you want to build a house on a piece of land, you check property records to determine who owns the land, so you know who to contact about purchasing or leasing the land. Of course, the same goes, in theory, for patent properties. But in our current environment, it can be a real challenge for those who find themselves facing a possible infringement suit to determine the owner of the patent they may be infringing.

This is especially true in the software area, where innovation is extremely diverse and broad-based among large and small entrants everywhere, and where patent applications and patents are frequently bought and sold. USPTO is working to solve this problem as well. Last year we issued a call for comments on the possibility of the agency requiring disclosure of the real party in interest for published applications and issued patents. And taking those comments into account, we’re moving forward. On January 11th, the USPTO will be hosting a roundtable event at our headquarters in Alexandria, Virginia, on proposed requirements along these lines. We hope to see many of you at this roundtable, and we welcome your feedback and suggestions on this important initiative.

The completeness of the patent record, including ownership, is essential to a functioning innovation market. Transparent and accurate markets attract investment, and repel gaming. It’s time for the patent system to man up.  And the benefits for the software area will be substantial. There are other areas where we’re working on projects that will help further improve our handling of software patents. The cooperative patent classification system will enable us to find software prior art from the U.S. and other countries much more effectively. It launches in just over a month, January 2013. A new information technology system featuring state-of-the-art search capabilities for our examiners is under development, and it will also assist greatly in finding software art. And there’s more, but we’ve already covered a lot of ground today, and I want to leave time for discussion. I hope I’ve been able to shed a little light on a subject that has mostly generated heat over the past year.

You know, the history of software patents is not a perfect one, although things are improving. Some of the most troublesome patents have expired; others can be challenged with new post-grant proceedings; and newer patents are quantifiably clearer, and aligned with current legal standards. But it’s important to note that, during the so-called smartphone patent wars, innovation continues at breakneck pace. A system like ours, in which innovation is happening faster than consumers can keep up, cannot fairly be characterized as “broken”. Nor can it be said that the U.S. is just a receiver of all this innovation. Most of the innovation is taking place right here. Broken? What?

The fact is, the explosion of innovation—and follow-on litigation—that we see across consumer electronics hardware and software is a direct reflection of how our patent system wires us for innovation. It’s both natural and reasonable that in a fast-growing, competitive market, innovators would seek to protect their breakthroughs using our patent system. While our IP system is not perfect, it is the envy of the world. It’s the strongest in the world, by far. That strength encourages investment and provides assurances to entrepreneurs as they enter the U.S. and global markets. At the same time, our focus on quality in the patent examination process overall ensures that patents are granted for true innovation, and not otherwise.

But for the evolving American patent system to realize its potential, the public needs to be an engaged stakeholder as well. I’ve outlined this morning some of the ways third parties can directly participate in the patent process. It doesn’t cost a dime to submit prior art; just thoughtful consideration of an application of interest, and use of our online tool. And the cost of pursuing a post-grant proceeding with our Patent Board is far cheaper than challenging a patent in court.

For those who feel more needs to be done, we encourage you to keep reaching out to us at the USPTO, as well as to other actors who also have an important role to play. The USPTO administers the laws, while Congress and the courts write the laws and interpret them, respectively. Working together, we can find the right balance for software patents. We can find a balance that ensures market certainty, encourages investment and research and product development, and guarantees that patents issued going forward are appropriately tailored.

Looking at the bigger picture, our nation has had a 230-plus year love affair with innovation. Our founding fathers enshrined patent rights in our Constitution, an affirmative right here, that in other countries is only issued grudgingly. It’s one of the few, if not only, clauses in the Constitution that gives Congress the right to create personal property. A right of the people, not an exception taken at the discretion of the government. Pretty central to our democracy. Open-ended. Maximally inclusive. This love affair with innovation has produced the strongest innovation environment the world has ever seen.

The questions raised over the last year aren’t about software patents. They’re about the basic premise of any patent system. Our patent system is our country’s investment plan—a giant 401k through which we pay a little extra now for more great innovations in the future. So the question for our innovation society is simple: do we demand today’s innovation on the cheap—via a weaker patent system that excludes subject matter and risks under-incenting innovation—or do we moderate today’s consumption with investment in a strong patent system so our children will enjoy even greater innovations?

While that question may be simple, the solutions to such complex challenges rarely are. Software patentability is a case in point. But the U.S. patent system is great because it struggles with complex challenges. For more than 200 years, when faced with challenges to the core of our patent system, we have not chosen the easy way out. We have chosen to struggle with complex challenges. Our courts have been asked—begged—repeatedly for generations to give up on software patents, and declare some simple rule. Congress the same. Both have chosen the wiser path, to continue struggling instead. And we are a better country—with a better patent system—as a result.

Thank you for your time today. I am happy to take questions.

This was originally published on the USPTO website at:


Warning & Disclaimer: The pages, articles and comments on do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of

Join the Discussion

3 comments so far.

  • [Avatar for Stan E. Delo]
    Stan E. Delo
    December 1, 2012 11:56 pm

    Remarkable presentation, which David is so very good at. He made the software patent issues much clearer for me in several senses, looking at them from a very broad perspective that perhaps only he could have. I wish he could stay a bit longer, but then again Time is just God’s way of keeping everything from happening all at once I suppose.

  • [Avatar for Roland]
    December 1, 2012 05:32 pm

    Many thanks Gene for bringing David Kappos’s unabridged remarks to our attention.

  • [Avatar for Anon]
    December 1, 2012 09:46 am

    Nice coverage – thanks Gene.

    Impressed with the “reprinted with permission” byline.