The Case for Software Patentability, An Interview with David Kappos


David Kappos

It is difficult to comprehend how rational people could at all question the patent eligibility of software in 2015. Innovative software is all around us, from our automobiles, to the Internet, to advanced robotics, and even green technologies like windmills. Yet, we find ourselves caught in a bizarro Twilight Zone when it comes to software patentability.

Recently the Partnership for American Innovation released a new white paper title Why Software Implemented Inventions Are — And Must Remain — Patent Eligible. Among other things, the paper highlights the valuable economic and technological benefits of software-implemented innovations, and the critical role patents play in protecting them. In fact, according to a 2012 study by the U.S. Government, IP-intensive industries, all of which depend heavily on software-implemented innovation, support at least 55 million jobs, contribute $5.8 trillion to our economy, and account for 38 percent of America’s gross domestic product. Despite this enormous contribution to GDP, many software innovations cannot be patented, and many of those software innovations that have been patented are having their patents stripped as the direct result of the Supreme Court’s ruling in Alice v. CLS Bank.

The aforementioned PAI paper was co-authored by former USPTO director David Kappos, and former chief IP counsel to the Senate Judiciary Committee, Aaron Cooper. The paper argues that patents on software-implemented inventions are accomplishing the constitutional imperative to promote progress. They have led to tremendous growth in the software industry, foster open innovation and interoperability, and are incentivizing solutions to unique problems.

When I learned of the PAI paper I reached out to David Kappos to see if he had time for a conversation about software on the record. He agreed. What follows is part 1 of our 2 part interview, which took place via on September 14, 2015.

Without further ado, here is my interview with David Kappos.


QUINN: Thanks, Dave, for taking the time to chat with me today. I know that you have recently put out a white paper along with the Partnership for American Innovation regarding software and software protection, so I wanted to touch base with you about that. Can you tell me a little bit about that study and the paper?

KAPPOS: I’d be delighted to, Gene. And thanks for getting on the phone for this conversation. The concern that lots of us have, including the members of the Partnership for American Innovation, is that there is in some ways a lack of understanding about the innovation that goes into so much software these days – historically, too, but especially these days — and a lack of understanding about the ubiquitous role that software now plays in our lives. And that it would be good to improve understanding levels about those aspects of software, the importance of software, and the importance of having a strong patent system that is available to protect innovation that happens to be deployed or implemented in software. So basic business concerns and business interests are what motivated putting this paper together.

QUINN: I think it should be just self-evident how important software is but people just don’t seem to appreciate it. I heard a commercial not long ago and I just wrote an article about it, with respect to GE windmills and wind turbines they said that they were improving windmill farms. And one of the things the innovation does is it allows the turbine to follow the wind. And I thought to myself well the only way you’re going to get a turbine to move into a position that maximizes wind capacity, that’s all software. It takes software to figure it out and to accomplish it. I suppose you could tilt and move every windmill every hour or so depending on the direction of the wind, but that’s just unrealistic. Every innovation is smart this, smart that, and none of them can exist without software.

KAPPOS: I agree that’s exactly right. The GE windmills are a great example. And whether you’re talking about windmill turbines or whatever, it is the tools that are used to design those things — to design the blades on the windmills, to design the structures that stand them up — that are all tremendously dependent on software. So it really is ubiquitous well beyond casual observation or understanding.

QUINN: You don’t even need to go into that level of abstraction. If you took the software out of your car your car wouldn’t work anymore. All you hear talk about today from automobile companies is autonomous driving. What do people really think is doing the driving? It is all software. We have a software innovation economy and I really worry that the direction we’re going in is going to just shut things down and cause this innovation to stop because we’re not talking about the software that can be created over a weekend. We’re talking about software that takes large teams of engineers years to complete.

KAPPOS: Right, right. That’s one of the points that the Partnership for American Innovation wanted to make. You know, companies like Microsoft and Apple and GE — all of whom are members along with IBM, Ford, DuPont and Pfizer as well as smaller companies like Many Worlds and Second Sight — all of them are engaged in the hard work of making major, I’ll call it bone-grinding innovations. Second Sight is literally coming up with electro mechanical and implantable human interfacing medical technology that enables blind people to see. And like you said, Gene, serious software development involving lots of super smart people and putting in tremendous amount of time with a lot of specialized expertise, devising solutions to very important problems. You know, enabling blind people to see — it’s hard to imagine a more tangible, practical and important problem than that. Or in the case of Ford helping cars sense and respond to road conditions and traffic conditions to keep people safe and avoid accidents, not to mention park themselves, all of which can be done now. Very serious heavyweight innovation and it’s all not just dependent on software but largely based on software.

QUINN: Yeah, you know, I don’t know whether you’ve gone back and thought about it in this way but I’m going to try and throw something at you and I’d love to get your thoughts on it – where did we go wrong? And by that question I want to take you back – do you remember the first time you saw Pong, the video game Pong?

KAPPOS: Oh, sure, sure, yeah.

QUINN: It was revolutionary. I remember the first time I saw that and it was like, oh, my god, this is crazy. How can you possibly do that?

KAPPOS: Yeah, it was so cool. Remember we’d spend hours playing the game.

QUINN: You would spend hours and hours and that was just the first generation. You look at things like that today it’s like people don’t have that same appreciation we had when we first saw Pong. Today developers are doing things that are infinitely more difficult to do than Pong. But back 35, 40 years ago when that came out it really captured the imagination. Somehow society has moved to the point where it just seems that they think innovation happens and they don’t understand what goes into making it happen. Take for example the Apple pinch and swipe and zoom. I mean that took years and years of development. And then people trivialize it saying “oh, that’s cool and interesting but that has to be easy to do.” And it’s just not easy to do. Where is the disconnect? How did the narrative get so far off the rails? Have you any thoughts on that?

KAPPOS: Yes, without commenting on your specific example, I have given a little thought to why people think software innovation is “easy,” why people are so hostile to patent protection for inventions that are implemented in software. And it’s interesting that you bring it up, Gene, because I gave a presentation recently, and the topic of it was the future of innovation. So I decided to say innovation or intellectual property in an innovation-driven world, what could possibly go wrong? With a picture of Alfred from Mad Magazine. And the point there was to say you would think this should be pretty straightforward. You’ve got a society that’s driven by innovation in this case, lots of it being software-based innovation, and you know as we do that the only system for protecting innovation, for capturing value from innovation, is the intellectual property system. In any sane rational world you’d look at that and say, “Well, this is straightforward.” Of course you’ve got to have a really strong balanced IP system particularly to protect the kinds of innovations that are the most cherished, which would include software as one category not to mention biotech as another category. You’d say of course we’re going to champion these and protect them. What could possibly go wrong? Well, something did go wrong.

I think there is a fair criticism in these trends. And one of the fair points of criticism is that in the early days of patent protection for software, and I’m going back now to when I was first starting in the field in the 1980s, it was hard to search for prior art and there was a kind of a land grab that went on that culminated in probably too many patents issuing at the time that were not equipped to meet what we now think of as the standards for patent hood. In fairness to the USPTO, you couldn’t expect them to examine using the KSR guidelines before KSR was decided or any of the other more recent opinions, Nautilus and its ilk that have come about recently that have helped to draw tighter boundaries and enabled the office to ensure that only really innovative inventions are getting patented. But before those days you know we wound up with some patents out there in the software area that in retrospect I think we’d all say we wish hadn’t have been granted. And you had some lawsuits that drew a lot of negative attention and that then starts the ball rolling.

Had we not had those events then I think probably again in fairness the software industry and the software programmers would not have become so very anti-patent and concerned about parts of their profession becoming off limits because of trivial patents. Had there not been trivial patents or at least not as many of them we would have not seen anywhere near as much concern and the field might have grown up to be more like others. That’s one point. And the other point, Gene, that I’d mention again in fairness to everyone is that as you and I know software is algorithmic in nature and it’s process-based in nature and it’s abstract in nature. And software has gone through one level of abstraction after another with what started out as simple code becoming processes and sub routines and then objects and on and on and on. With each new generation of programming we introduce a new level of abstraction. And the combination of the shifting nomenclature, the levels of abstraction, the process nature of software has all really and I’ll say conspired together to make it really hard for either the patent draftsmen or the patent examiner or the public later on to really get a handle on describing software innovation in sufficiently clear patent claims. So that’s my view. In part we got off to a bad start and in part we’re dealing with a subject matter that’s inherently challenging that has gotten us into the fix we’re in.

QUINN: There’s one thread that ran through what you were just talking about that always irritates me to no end when I hear it from the critics of software. They talk about nomenclature for example. They use the terms out of context. They use them in every day ways. But that’s not what the terms mean in a legal discussion. And then they look at the way that these things are described and they laugh and say, oh, well, that’s easy. And it’s just not easy. Somewhere along the way it seems that we’ve gotten to the point where a patent in order to be valid and enforceable has to be an instrument that Joe Average could actually understand. But we live in a world where I don’t know if that’s possible with the level of sophisticated innovation we’re talking about. And then the other thing that just runs in my head is that in my experience a lot of the programmers who complain about software patent protection the loudest will stop reading once they look at the title of the invention.


QUINN: I don’t know how we can have an honest discussion if all you are going to do is read the title of a document that will be many dozens of pages long, if not over 100 pages long. I’m not saying that anything that you said was incorrect. I think there are patents out there that are unfortunately granted and that didn’t help at all. And the EFF points to those all the time. And the ones that they point to usually make you question how that particular patent got issued, you know? But we seem to be grading the patent system by the obvious failures, which in the software area strikes me as being few and far between. I mean it’s been 10, 15 years since these things were freely given out by the Patent Office. But that’s what you still hear. You still hear members of Congress saying that as well. Do you have any thoughts on that?

KAPPOS: Indeed, the issue you’re pointing to now I’ve also considered. And again, perhaps taking a little bit of a philosophical approach to all of this. I think you’re right, first of all. People frequently don’t read beyond the title of a patent and are frequently ill-equipped to understand even well written, very valid and defensible patents. I’ve seen some get demonized and trivialized and I found myself shaking my head also saying, well, if you would have read and understood the claims people would be saying very different things about this patent. And to me that whole line of discussion fits into a category of recognizing that at some level the patent system has always struggled with being operating at the intersection between the law, technology, and the future, which is a difficult intersection to operate at. And therefore it always has had a certain level of controversy associated with it as people have struggled to understand inventions expressed in patents and express their frustration with their inability to understand.

So there’s an element in all of this in my view that goes to the inherent difficulty with the patent system. It’s been around since the 1600s when the British first started issuing patents. It was around when Thomas Jefferson and James Madison started examining patents. They were struggling with some of the same stuff. And, oh, by the way as you and I know there were some patents that were being issued even back then that we would now call software patents. We all are really struggling to this day and I’m sure it’ll continue indefinitely in the future. It is that complicated. If you’re not really a deeply experienced scientist or someone who is used to reading patents, unless you’re able to and equipped to invest a lot in the domain, the domain being the relevant technology as well as understanding how patents read, you’re going to be perplexed. And people are going to exhibit that as frustration, which we will be dealing with. I just don’t really see a way to change that.

As you know artifacts have come about and Congress has passed laws from time to time, to require a title, for example. Require an abstract. Include a background. I mean all of these things in various forms are actually, at least somewhat, written in plain English. At least they’re not as hard to understand as patent claims. Yet nothing seems to do any good and for me at the end of the day where I find myself coming to is that is just the way it is. It’s an issue that’s dogged the patent system for hundreds of years and it probably will continue to dog the patent system.

CLICK HERE to CONTINUE READING… In part 2 of our conversation we pick up discussing why the label “software patent” is really a misnomer, the unfortunate reality that the patent system is not in a good place presently, the reality that software is just a medium of expression and how ridiculous it is that an algorithm can be patented when it is implemented in hardware but not when it is implemented in software.


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

10 comments so far.

  • [Avatar for Edward Heller]
    Edward Heller
    October 1, 2015 11:48 am

    Night, I am definitely anti-business method patents, not doubt about that.

    But, that is not what Kappos and Gene are talking about here. Every example given is of the kind of software patent I totally endorse — where software lies at the heart of the improved machine, manufacturing process, etc.

  • [Avatar for step back]
    step back
    September 30, 2015 04:49 pm

    Nelson Avek and Night Writer,

    Great discussion by both of you at comments #1 and #2.

    People have no respect for one another and no comprehension what it takes to make things work in different fields of expertise.

    I have worked in many fields including:
    1) Hardware (from design of microcode functionality to control of large electric motors in noisy environments);
    2) Software (from design of microcode functionality to design of end user application software); and
    3) Mind manipulation by use of words (a.k.a. lawyering).

    It has been my experience that folk in the hardware end of the business have no comprehension for what it takes to make things work in the software end and thus have no respect for it.

    Similarly, folk in the software end of the business have no comprehension for what it takes to make things work in the hardware end and thus have no respect for it.

    Ultimately, for people in the lawyering business, words are just words to be tossed about for whatever persuasive end purpose deemed necessary and they couldn’t care less as to what the lowly peons in either the hardware or software realms do.

    The trick is to persuade the arrogant and gullible Supremes:
    Oh high and mighty know it alls. Though are truly the wisest in all matters. Holy and infallible be thy words. Behold that so many have said software is abstract. Therefore truly it must be so. Enchant for us your holy words declaring all software to be abstract and not eligible. Listen not to the crafty and tricky draftsmen on the other side (lest yea be fooled by the truth). Thank you very much.

  • [Avatar for Night Writer]
    Night Writer
    September 30, 2015 01:05 pm

    Edward, (1) You think Benson is great, and (2) there is no definition of a business method. Please. Your posts are consistently anti-software patent.

  • [Avatar for Edward Heller]
    Edward Heller
    September 30, 2015 12:30 pm

    Curious, but Kappos and Gene are not talking about business methods. Every example they give, every one, is about improvements to machines, and improved methods of manufacturing.

  • [Avatar for Eric Berend]
    Eric Berend
    September 30, 2015 09:55 am

    What’s worse about this whole U.S. patent laws fiasco?

    Is it that the whole concept of software and hardware as a particular invention is poorly captured in terms of conventional appliance? Note: please see my comment, #7 in, that analyzes that actual operation of software-as-invention; something that should have been established before making ANY attempt to “reform” the U.S. patent laws.

    Or, is that other disciplines of invention, those based upon electrical circuits, mechanical action and chemical activity are carelessly or deliberately derogated with contempt for their historical prowess?

    Or: is it simply that the meddling of powers-that-be in U.S. legislature, at the behest of particular industrial interests, is deliberately intended to empower patent infringers and thereby destroy opportunity in general for individual inventors?

    Whatever one’s opinion about the first two questions may be, regardless of the absurd calumnies and obfuscations, there can be little doubt that the real effect is the success of the third.

    Which means that there can be little doubt, that this is and has been a main and true goal of this destructive movement from the beginning, whatever may also be promulgated regarding software patents and so-called “patent trolls”.

    In the end, the true ‘patent trolls’ ARE the infringing interests, here. All of the nonsense about “patent quality”, “patent trolls”, “defending innovators”, protecting small business”, etc. is directed towards ONE GOAL:


  • [Avatar for Night Writer]
    Night Writer
    September 30, 2015 08:57 am

    >>new machines, methods of manufacture, and the like, where the heart of the invention is improved software

    Ned, software is a component and should be separately patentable without “being at the heart of the invention.” That construct that you push over and over again is —of course–in polar opposition to this article.

  • [Avatar for Curious]
    September 30, 2015 08:16 am

    Software is an improvement to a machine — clearly statutory subject matter under 35 USC 101 (“Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof“) Whether the software performs automated trading on an exchange, improves the security of message exchange, adjusts the pitch of turbine blades, or makes funny cat videos even more funny shouldn’t matter. This just goes to utility, which is a very low bar to hurdle.

  • [Avatar for Edward Heller]
    Edward Heller
    September 30, 2015 07:35 am

    Gene, Seriously, I know of no one, let alone the Supreme Court, who questions the patentability of new machines, methods of manufacture, and the like, where the heart of the invention is improved software.

  • [Avatar for Nelson Avek]
    Nelson Avek
    September 30, 2015 01:49 am

    I spent more than 2 years in learning development and another 5 years in becoming enough good to produce something patentable. People easily spit out that software innovation is easy as pie, but the truth is, it takes away the sleep of a developer to build something for the living. Their hard work is worth protecting. However, there are some shitty people out there abusing the whole system with crap patents as well.
    It was an informative post, thanks to you Gene.

  • [Avatar for Night Writer]
    Night Writer
    September 29, 2015 10:21 pm

    Great article. I wonder too why people think that software is easy. I remember taking a class on expert systems in the mid 1980’s with a professor who had just received his Ph.D. from Stanford in Computer Science for an expert system design. It was tough work to get expert systems working.

    Now we have Fed. Cir. judges dismissing expert systems as if they are nothing more than two sticks being put together.

    The other thing that I remember is teaching computer science back in 1980’s. I remember how the older generation didn’t understand computers at all and something as simple as a menu would drive them into hysterics.

    But, now, somehow or another, there is this mental model of computer that can just do anything you tell them to. Stevens, of course, infamously wrote that you just write down what you want the computer to do and give it to a clerk to program.

    But, in reality, the cutting edge of software is as tough as it ever was. One thing I have noticed at the Fed. Cir. is you seldom hear from the non-science judges the word hindsight.

    Just musings….