X
More Topics

Software patents: Broken system or needed for innovation?

They're bought and sold like commodities. Can the patent system be fixed - or should it be scrapped?
Ed Burnette

Ed Burnette

Broken

or

Needed

Steven Shaw

Steven Shaw

Best Argument: Needed

The moderator has delivered a final verdict.

Opening Statements

Patents are a minefield

Ed Burnette: The goals of the patent system are commendable: to encourage and reward innovation, and to make information about new inventions available to other practitioners. This may have worked for physical objects, but when applied to software, the system is actually having the opposite effect.

Patents do not reward the innovators; instead, the only ones rewarded are the lawyers and those who game the system. Developers are routinely accused of "stealing" and hauled into court because somebody else came up with the same (often obvious) solution. Patents are bought and sold like commodities, with none of the proceeds benefiting the original author.

Because willful knowledge could trigger extra damages in a lawsuit, developers are seldom allowed to read patents. Even if we could, the legalease they contain is so convoluted and obscure that we can't hope to learn anything useful from them.

Patents are a minefield: broken, unfair, and ultimately self-defeating.

Software needs protection

Steven Shaw:  The Founders considered protection of intellectual property so fundamental to the new nation of the United States that they wrote it into the Constitution (Article I, Section 8, Clause 8). Not in an amendment to the Constitution, like free speech. It's in the actual document.

Without protection of intellectual property, the Founders knew, there would be no incentive for creators of inventions to "promote the Progress of Science." Their plan, hatched in the 18th Century, was a foundation of the explosion of American creativity and inventiveness of the next two centuries.

Today, there are those who believe software should not be afforded the protection of the patent laws. They are wrong.  While some people would continue to write software without the option of patent protection, there would be a radical contraction in software development without the patent laws to guarantee profit to creators of new software.

The Rebuttal

  • Great Debate Moderator

    Here's a patent portfolio

    If you were a CEO how would you use a patent portfolio?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    As little as possible

    At this point, patent portfolios are a necessary evil. So until patents are outlawed, you have to be an outlaw and keep your own stockpile to use in case of an attack. This tactic doesn't always work, especially against non-practicing-enties who have nothing to lose if you countersue them. But they do mitigate the risk of suits from real companies like Microsoft, Oracle, and Apple. The approach taken by Twitter is best. In April they annonced the "Innovator's Patent Agreement" (IPA) http://blog.twitter.com/2012/04/introducing-innovators-patent-agreement.html . Under this agreement, the inventors themselves get to decide how their inventions are used. Furthermore, control flows with the patents so if the patents are sold, the inventor retains control. "With the IPA, employees can be assured that their patents will be used only as a shield rather than as a weapon". Twitter's vision is "a world that fosters innovation, rather than dampening it". That's one I hope we all share.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Depends

    It depends on what I'm the CEO of. In some companies, it's important for the patent portfolio to cover what the company does. So if I ran Google, for example, I'd buy and buy and build my portfolio to include all patents covering anything I do and anything I'm thinking about doing.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Open source and patents

    Has open source helped inflate patent valuations or kept them in check?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    In check

    Free and Open source is one of the few things keeping patents in check at the moment. While I don't agree with many of the philosophies of the Free Software Foundation, they have certainly been a leader in the anti-patent effort.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    No data

    I'm not aware of evidence that open source has affected patent valuations one way or the other. There may be data out there. I haven't seen it.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Do the patent gatekeepers know what they're doing?

    Do the patent authorities have the know-how and expertise to realistically approve software patent applications?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    They have a resource problem

    It's not so much a matter of expertise as it is of resources. According to a study by Bessen and Maskin in 2004 [ http://www.researchoninnovation.org/iippap2.pdf ], the US Patent office had granted nearly 200,000 software patents, with the number rising over 20,000 per year (over 15% of all patents granted). The patent office just doesn't have the manpower to deal with this, so the result is that many thousands of patents have been approved with insufficient review as to their obviousness and prior art. The office has essentially punted review of the patents down the road to the courts when patent owners decide to sue so called infringers. Defending against a patent suit is an insanely expensive proposition, so many defendants find it cheaper to settle. Just look at how many Android licensees Microsoft lined up before Barnes and Noble drew the line and decided to fight back. Rather than let that one go to court, Microsoft ended up paying $300 million on a joint partnership to make the case go away.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Mostly, yes

    If you apply for an electrical engineering patent, your application will be assigned to a patent examiner with an electrical engineering background. It's the same with software patents: the US Patent and Trademark Office (USPTO) hires a ton of people with computer science backgrounds and education to act as patent examiners for computer software and hardware patent applications. As with any large, complex operation, you'll get mismatches and mistakes, but fundamentally the USPTO does not hire stupid people.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Litigation as a defense

    Do you view patent litigation as a defensive strategy with merit?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    No, but what other choice is there?

    Ideally we wouldn't have to defend against mostly frivolous suits. Failing that, the best way would be to try to get each patent invalidated through prior art and obviousness. But that takes time. A defensive portfolio is considered by many to be necessary for countersuits.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Yes, with modifications

    Yes, though there should be some changes made. There are two main problems with patent litigation: First, it is expensive. Second, the issues are often too complex for most juries or even judges to grasp. The solution is to have the issues litigated in expert courts, and to develop expedited procedures for patent litigation and push alternate dispute resolution mechanisms (such as arbitration) designed to bring the price into a more realistic realm.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Intellectual Ventures: Brilliant model?

    Where do funds like Intellectual Ventures fit into the patent landscape?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    Trolls

    These companies are just adding fuel to the fire, blowing up the bubble more and more. In one sense, they are not doing anything wrong. They are operating within the constraints of the law, and playing by the rules. It's a business model. But those rules are wrong and need to be changed. Whenever Nathan Myhrvold of Intellectual Ventures makes a public appearance, he is heckled and booed by irate onlookers. But I think that we should save our ire for the system that makes his business possible.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    It's not a fund

    Just as background: Intellectual Ventures is the company founded by Dr. Nathan Myhrvold, former Chief Technology Officer of Microsoft. I wouldn't call Intellectual Ventures simply a fund. It's a company with a lot of facets. I've crossed paths with Myhrvold on several occasions, and he's surely the smartest person I've ever met (which doesn't mean much, but Bill Gates has said the same thing about Myhrvold -- and that means something). So it's not surprising to me that his company is too complicated for the average newspaper reporter to understand or write about coherently. Myhrvold and I have mostly spoken about food and wine, because he is also a thought leader in the culinary world (the guy has a lot of areas of interest and expertise; paleontology, too), so I have no special insight into the secret world of Intellectual Ventures. But if you want to get an idea of his vision for Intellectual Ventures, and all the parts of Intellectual Ventures -- including an amazing prototype lab -- go read the company's "about" page http://www.intellectualventures.com/aboutus.aspx . In the popular press, Intellectual Ventures and Myhrvold have been covered mostly for their efforts to protect their purchased patents, aka "patent trolling," which as I mentioned before is a nonsensical designation, but if you avoid that oversimplification you can see that it's a remarkable, visionary company.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Bubble trouble

    Are patent valuations currently a bubble?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    Definitely true

    An economic bubble is defined as "trade in high volumes at prices that are considerably at variance with intrinsic values". Google paid over $12 billion for Motorola, largely for their patent portfolio. Microsoft, Apple, and RIM paid $4.5 billion for 6,000 patents from Nortel, and Microsoft paid $1 billion for 800 AOL patents. Google also bought over 2,000 patents from IBM in 2011 for an undisclosed amount. These prices far outweigh any intrinsic value the ideas might have, even if they were all valid.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    The opposite

    I think a lot of patents are currently undervalued because enforcement is such a pain and major expense. Most every patent valuation currently includes a discount for difficulty and expense of enforcement. But whether patent values will someday drop precipitously depends on events: If the anti-software-patent people have their way, software patents will become worthless and there will be a self-fulfilling bursting of a bubble.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Trading patents

    Do you agree that intellectual property and patents are something that can be traded like any commodity?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    Can? Yes. Should? No.

    They are certainly being traded that way now. I don't agree with the practice though, because it takes control out of the hands of the original creator. Ideas and thoughts are not property! But an inventor should have a say as to how their invention is used. If they were encouraged to write a patent for purely defensive purposes, but then the patent is sold to another company which is sold to another company, that promise is no longer honored. That's not right.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    I don't accept the terminology but agree with the idea

    The term commodity implies that something is generic, like oil or frozen concentrated orange juice. Intellectual property is very diverse, so I wouldn't call it a commodity. I suppose some clever person at Goldman Sachs could aggregate a lot of intellectual property and create a commodity-like financial instrument (for all I know it has already been done), but the underlying intellectual property is decidedly not a commodity. Still, yes, absolutely, I agree that intellectual property is and should be something that can be bought and sold. If you can't buy, sell, and do what you want with your property then you don't really own it, and without the ability to own property the whole system collapses.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Aren't patents a commodity today?

    Seems like a) everyone has one; b) they are traded via auctions and c) used against rivals like an oil embargo or something similar.

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    Unfortunately yes

    Lawyers have turned patents into a game, a way to make a quick profit. Just like banks and brokers have turned the stock market into a casino. Patents are no longer about innovation, just like stocks are no longer about underlying fundamentals.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Not really

    a) there are a lot of patents but not billions; it's still unusual for someone to own one -- do you?; b) fine art is sold at auction but is not a commodity; c) that would only be true if one owner had most of the world's patents, which is not the case.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Second hand patents

    From a talkback: A patent that is purchased second hand should expire a bit sooner. The buyer is the present owner but not its creator. Agree or disagree? And why?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    That's not the answer

    The law is complicated enough without adding conditions to the expiration time. A better idea would be to make the author's rights non-transferable. That way he or she could control what happens to their creation no matter how many buyouts occur in the interim.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Disagree

    All that accomplishes is that it makes patents worth less money, because if you sell them the purchaser can't get their full value. I see no point.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Patent trolls: Fiction or fact

    Do patent trolls really exist?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    Absolutely

    Patents are like weapons of mass destruction. If two large nation-states have them, there is a measure of balance and safety because both sides lose in case of a conflict. But if a lone terrorist gets one, all bets are off. Similarly, when one company with a large portfolio sues another company with a large portfolio, the result more often than not is some kind of cross licensing agreement. A patent troll, more formally known as a "non-practicing entity" or NPE is like the terrorist. You can't threaten them with a countersuit that could cause them to stop producing their products because they don't have any. You either have to pay them their ransom, or try to get their patent invalidated. Meanwhile they can get an injunction to halt your product and revenue stream. Lately even large practicing enties are behaving like trolls because they so dominate the other party. In this case, the defendants cannot hope to mount successful defense or counter-offense before their money runs out or their market opportunity is lost. There's a great story from Gary Reback printed in Forbes in 2002 about how IBM shook down Sun Microsystems. I encourage you to check it out. Microsoft is behaving the same way (see the Microsoft v. Barnes&Noble accounts), but they won't even let you know which patents you are allegedly infringing.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    No

    That's like saying someone who buys and sells real estate as an investment, rather than living on it, is a troll when he sues trespassers. Private ownership of property is the bedrock of capitalism. The ability to own property, more than any other factor, has been responsible for economic growth and resulting improvements in quality of life in the past two centuries. Whether that property is real property like land, personal property like clothes, or intellectual property like patents, makes no difference. Without the ability to own any of those types of property privately, people will no longer have the profit motive to motivate them. It's certainly an option to trust innovation to the goodness of people's hearts or to open-source-type business models, but there is no example in history of that working on a society-wide, economy-wide basis. As much of a fan as I am of Linux, I don't for a second think Linux is a good model for governing the world. Once we decide that intellectual property is property and can be bought and sold, the whole negative connotation associated with being in the business of buying and selling intellectual property evaporates.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    How long would it take to implement your fix?

    Years, days, months?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    First step is quick, after that, could take a while.

    All it would take is one good Supreme Court decision, or a courageous act of Congress, for the situation to be fixed in the US. Given resistance to software patents in places outside the US like the EU, I think a change in the US policy and a removal of pro-patent US pressure would be felt around the world and an equillibrium of common sense would eventually over. It took about 20 years to get ourselves into this mess. Hopefully it would take a lot less time to get ourselves out. We came very close in the 2010 Bilski decision, but the justices didn't go the extra step to scrap software and method patents altogether. Patent writers just had to learn new tricks beyond the so called "machine-or-transformation test"; new ways to word things so what are essentially ideas could still be patented. Another good decision was KSR vs. Teleflex in 2007, which said that patents can't be granted if the invention would be obvious to a "person having ordinary skill in the art". In practice, though, according to patent expert Mark Nowotarski of Markets, Patents & Alliances LLC, the flexible approach of KSR cuts both ways. He writes: "KSR vs Teleflex is turning out to be a surprisingly powerful tool for helping patent practitioners persuasively argue that their clients??? inventions are not obvious."

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Minutes...years...

    Ten minutes to raise the bar on granting patents. Several years to change the court system.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Software vs tech patents

    Why are software patents different than other technology patents---materials, inventions etc?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    Software is an Expression of the mind

    Software, math, algorithms, and business methods are all expressions of the mind, and the mind should be free to explore new ideas and frontiers without being encumbered by the artificial constraints of patents. Physical objects start out as an idea, of course, but have to be distilled down to something in the real world that you can see and touch. That's the thing you can patent. We have copyrights for writing, musical performances, artwork, and other forms of expression that are more mental than physical. It makes no sense for software to be both copyrightable and patentable. That is too restrictive. Copyrights are enough.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    They aren't

    This is the biggest confusion people have about software patents. A lot of people think of software makers as "publishers," like book and music publishers. They are not. When you play recorded music, you hear music -- that's the end of it. Software is more like a machine: it does something, performs a function. Almost any software program could just as easily be built as hardware. So for example if you want to do voice or handwriting analysis, you have two choices: a hardware implementation or a software implementation. The underlying flow chart of what your invention does is going to be the same either way. The thought process, the collaboration, these things are all structurally the same whether you're building a hardware physical machine or a software virtual machine. Only the implementation differs.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Fixes

    What would you fix about the system as it stands today?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    No software patents, period.

    Patents should not apply to software and business methods at all. For many years, they didn't. Patents on software started to be allowed in the 80's and early 90's, and only then because legal innovations not technical ones. Practically every software patent begins with the words "A method and apparatus for...". Ever wonder why? The wording is there to get around the restriction that for 200 years, patents had to be on physical objects. A computer is a physical object, so patent writers had to relate the software back to an apparatus to make it qualify. Instead, software should be protected by copyrights. Copyright has worked well for hundreds of years for media such as books and music, and has been extended naturally into the digital age. Copyright violations are clear cut. It's easy to tell, for example, if somebody has copied your novel, your sheet music, or your source code files. Two people can't write the same novel by accident. Of course, there are those who will try to game the copyright system too, but at least they can't get inside your head and try to control your ideas and insights.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Patents should represent more than minor innovation

    I'd be the first to say that the threshold for obtaining patents -- in all areas -- is too low right now. Patents should represent more than minor innovation. I don't think they need to be limited to historical, watershed breakthroughs, but the threshold should be higher than it is now. I would also push for streamlining the patent litigation system in several ways, for example through directing patent lawsuits to a specialized court system (the same way disputes over wills in several US states get funneled into probate or surrogate's court, and patent appeals go to the Federal Circuit) after first making an aggressive push to have those disputes settled by arbitration, mediation or negotiation.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Technical difficulties

    Sorting out. Back in a minute.

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

  • Great Debate Moderator

    Let's get real: You need software patents

    It has been argued that software patents are necessary to provide motivation to innovate. Do you agree with that argument? Why or why not?

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    Motivation is found elsewhere

    No. Innovation is done by the developers, by the men and women in the trenches. To us, patents are more of a hindrance than a help. Creating useful and stable software is hard enough without worrying about other parties we have never heard of coming in and accusing us of stealing. The real value of software should be in the code, not in pieces of paper that lawyers fight over. There's an art and a science to programming. The best motiviation is to see your creation make a difference in the world - to solve some difficult problem, to make somebody's life easier, or to enable new paradigms that weren't possible before. Patents add friction to that. They raise barriers, add uncertainty, and reduce communication and trust between others in your field. Before patents, ideas were freely shared in a vibrant community of conferences, journals, online forums, and standards. Now, patents are like a dark cloud hanging over everything.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Needed

    duplicate

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

  • Great Debate Moderator

    Mike check, test 1, 2, 3

    Ping back to make sure.

    lawrence-dignan.jpg

    Posted by Lawrence Dignan

    Ping

    All systems are go.

    ed-burnette-60x45.jpg

    Ed Burnette

    I am for Broken

    Mic check

    Present.

    steven-shaw-60x45.jpg

    Steven Shaw

    I am for Needed

Closing Statements

Just say no to software patents

ed-burnette.jpg

Ed Burnette

Suppose, as Steven argues, that intellectual property can be defined and described, and needs to be protected. Can it be Patented? I would argue that a Patent is the wrong way to protect such a thing, if it even exists. Copyrights are used for other products of the mind such as music, eBooks, and newspaper articles. To me, it makes the most sense to apply copyrights to software as well. Copyrights offer plenty of protection by themselves.

Can software patents be fixed? Over the years, we've heard many ideas such as shortening patent expirations, crowdsourcing prior art, making lawsuits less expensive, and more. My answer to all this is: Why bother? We can have innovation without patents. Some of our greatest innovations such as the World Wide web were not patented (thank goodness!). Patents are a drag on innovation, because they incentivize non-productive behavior. We'd be better off without them.

Software is more like a machine

steven-shaw.jpg

Steven Shaw

We've covered a lot of ground, but I want to circle back to what I think is the fundamental misunderstanding of software patents.

Software is not the same as mathematical formulas or literary text. Software is much more like a machine. A typewriter is no more of an invention than word processing software -- I'd actually argue that word processing software requires considerably more inventiveness. If a person comes up with an innovative new idea, the idea should be patentable regardless of whether the implementation is via a traditional mechanical machine or a virtual software machine. Once you understand the nature of software in this way, software patents make perfect sense.

There are improvements that can be made to the patent system, but they are the same types of incremental improvements every large system needs. Scrapping the system is just a bad idea.


Not the people's choice

lawrence-dignan-150x105.jpg

Lawrence Dignan

Steven Shaw is not a popular guy. Why? He's writing that software patents are necessary. Oops. Ed Burnette made a compelling case that copyright law -- rather than patents -- should hold sway over software. However, Shaw's argument held together well. His bottom line: "There are improvements that can be made to the patent system, but they are the same types of incremental improvements every large system needs. Scrapping the system is just a bad idea." I'll give Shaw a narrow victory. The crowd apparently disagrees.

 

Editorial standards