Why we need software patents, part 3 of 3

Why we need software patents, part 3 of 3

Summary: 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.

TOPICS: Patents

In last week’s ZDNet Great Debate, I argued in defense of software patents. This is not a popular position among heavy internet users, even though it is the current state of the law and industry practice, and in the popular vote I lost 89% to 11%. (I’m relieved to say that the moderator declared me the winner.) These past few days I've been doing three posts about the basic premises of the argument for software patents. The talkback comments on the first two installments have echoed the 9:1 ratio pretty closely.

In the first installment of this series, I argued that the distinction between software and hardware patents is nonsensical. In the second installment, I argued that the intellectual property laws are necessary to encourage innovation. The last part of this argument is simply that the patent system, as presently constituted, is a good starting point. It would be destructive and disruptive to scrap it. 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'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.

These types of changes could improve the software patents system without causing the kind of massive economic disruption that abolition would (quite aside from the fact that abolition is an intellectually unsound position).

If time and space permit, I will try to circle back for part 4 of 3, to address some of the questions and fallacies that have come up in the talkback comments.

See also:

Topic: Patents

Steven Shaw

About Steven Shaw

Steven Shaw used to be a litigation attorney at Cravath, Swaine &gMoore, a New York law firm, and is now the online community managergfor eGullet.org and the Director of New Media Studies at thegInternational Culinary Center.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Trouble Is, Your Arguments Are Wrong

    Your first point was just special pleading: you were happy to use the argument that "software is just like hardware" to extend patent restrictions on hardware to software, but you went mysteriously silent when it was pointed out that the same argument should also work the other way, so if hardware cannot be copyrighted, why should software be?

    And your argument that "intellectual property" is necessary to encourage innovation is simply not borne out by the facts that have been steadily accumulated by researchers who have actually been looking into how economics works in the real world. Anticompetitive protectionist measures do not encourage a competitive free market.

    And the existing US patent system is in a totally dysfunctional state, where rejected patent applications can simply be resubmitted until they are accepted, or simply hoarded as a weapon to run a protection shakedown racket.

    And to top it all off, the US is trying to export its dysfunctional systems to the rest of the world, by pressuring them into accepting "free-trade agreements" that favour its "intellectual property" industries. Thankfully, you see the growing public discontent over ACTA, the suspicions over TPP, the political rise of "Pirate" parties in Europe and elsewhere--hopefully that will help put an end to this insanity.
    • and...

      I'd really like to hear the argument why "abolition is an intellectually unsound position" when most discoveries and inventions in the past weren't subservient to such a dysfunctional system... I guess in your world JP Morgan is a visionary and an innovator! In mine that would be Nicholas Tesla.
    • Some balls this guy has

      89% disagreement with his argument in the polls and he's still going to tell us why we're wrong and he's right. Just accept it!
      • Balls? Maybe salary?

      • That aspect doesn't bother me

        Majority opinion (even large majority opinion) doesn't determine truth, so if he really thinks we're all seriously misguided on a matter of major importance, he should try to change our minds. I was disappointed, however, to see that his arguments were so poor. I think any of half a dozen sympathetic Talkbackers could have done better.
        John L. Ries
    • Re; "intellectual property"

      This term is in itself a pack of lies.
      To insist of calling privileges property is insane.
      They are privileges granted by governments.

      These privileges expire.
      Property does NOT expire.
  • Patents Should Be Difficult To Get

    The goal of the Patent Office should be to find a way to [b]Deny[/b] patent applications. Extraordinary effort should be put into finding prior art. Obviousness should be given great consideration. And patent applications with only incremental changes should be tossed like the trash that they are.

    Making a patent almost impossible to obtain would increase the value of a patent. We don't need to abolish patents. We need to abolish silly, obvious, incremental patents.
    • Non-Obvious

      The non-obivous, which is supposed to be in the patent system already, would need to actually be implemented.

      Extracting a telephone number from a block of text and doing something with it is OBVIOUS! We've been doing extracting data from text for decades, it is pretty much a staple of programming. And once you've extracted it, doing something with that extracted data is obviously what comes next...

      Any half way decent programmer could do that in a few minutes and using existing libraries and interfaces into other applications would allow him to pass the number on with ease to a contact database or the phone app.

      It definitely isn't rocket science, it is an afternoon programming in an introductory class...

      And that is my main argument against software patents, 99% of those I see coming up in court are patents for the bleeding obvious or are frivolous or so wide ranging that they should have kicked out during the inital vetting process...
    • A talkbacker on his part two of the series posted

      a link to a website and study that I think anyone interested in this subject should read.


      I was always of the opinion that the IP protection system was seriously broken, in need of major reforms, and was too far reaching. Absolute abolition of the system, however seemed to be far too radical an idea as I also believe that people's right to be compensated fairly for their works should be protected. The authors of this website and study make a very good case for absolute abolition of copyight and patent laws and do what the other side fails to do: They support their position with credible historical and modern case studies that show that the IP laws are not only unnecessary they are harmful to the greater goals of progress, creativity and innovation.
      • As it stands

        the inventors are generally are not fairly compensated for their work. It tends to be executives at large companies and their hoardes of lawyers who benefit from the system.
    • Difficult? Sure. Near-impossible? No.

      One problem with what you are suggesting. This type of system will gear itself so only those with the resources to argue their case will have a chance. This will actually stifle the little guy who wants to patent a great new innovation.
      • How does this differ from the current system?

  • Why stop at software? How about patenting legal arguments?

    To protect and encourage innovation in the legal profession, of course.
    • Patenting legal arguments is actually inherent to the legal system

      I may be wrong with the US but In my country there are things called precedents which are essentially the same as legal system patents. The main difference is that a lawyer can't sue another lawyer for using a similar argument that has become a precedent. In fact most people will probably realise that a Lawyer will not endorse suing another lawyer but will encourage any behaviour for others to sue others, I tend to think that is why Mr Shaw likes patents, it makes lawyers money even if he is an ex-lawyer.
      • Precedents are not patents

        Once a legal precedent is set, anyone can cite it.
        John L. Ries
    • Yes!

      Let's patent affirmative defense! If that's too broad - insanity then!
    • LOL!

      Now thats a a IP system I would love to see enacted while they tear down all others!
  • 9:1 ratio is another problem

    There is the other problem. If approximately 90% of software engineers are innovating without taking patents then that is another reason patents should be not be allowed for software. It would mean the patents do not accurately describe the prior art and leads to the approximately 10% of software engineers patenting existing technology and current standard of art practices.

    Take for an example Apple patent # 5,946,647 which detects structures and assigns actions to them. It was filed in 1996 and awarded in 1998. The HTML specification defines structures (HTML Tags) and one of them is the anchor or link structure which is defined to have the action of navigating to that link. This was described by 1991. The browsers prior to 1996 not only had this single action implemented, but also had multiple actions in the context menu on it and on the image structures. Word processors since 1989 had implemented spell checkers which detected structures (misspelled words) and presented multiple actions to the user such as replacing it with various different correct spellings, ignoring it, and adding it to the dictionary. We could continue listing potential prior art such as email clients and games like "Legend of Zelda" (1985). There's papers from 1989 describing using regexes to detect phone numbers. Now, it appears Apple is trying to hold the world hostage on such a basic principle of software engineering.

    I think that software patents are so broken, that we may need to toss the bath water to see if there's a baby there to save.
  • You agree that the threshold should be higher

    yet you do not come up with ways to achieve that goal. Right now the only way to prove that a threshold has not been met is by allowing the patent to be approved and let litigation prove the matter wrong. The fact of the matter is that only large corporations can afford to litigate patent matters, and the burden of proof lays on the innocent.

    To fix that, the patent office is going to have to do a more diligent job. But how can they accomplish that? Any solution you can come up with will cost a lot of money. And how will they raise that money? Well one way is to raise federal taxes in general, and we know how well the American population will take to that idea. The other is to raise the application fee, which again would raise the bar so high that only large corporations can afford to play the game.

    There is no method of properly maintaining a patent system that is fair to all. It is too complicated to manage, a fact that has been proven time and time again in the real world. The best way to protect the works of a programmer is through copyrights and trademarks. Copyrights and trademarks should be allowed to include provisions that protect the both the source code and the overall design and look of the software. These are things that any jury with common sense should be able to decide a violation has or has not been made in a court of law at a minimal legal cost. This also gives you your reasonable minimum threshold at a minimal cost, if you can't prove to a normal jury in a normal trial what your protected work is exactly and how someone violated it then it does not deserve to be protected.
    Michael Kelly
    • The Fix Is In Making Patents Applications Difficult To Get Approved

      By making it almost impossible to get a patent application turned into a patent the system will self-regulate. If the chance that your patent application will be approved is less than one in a million then the number of patent applications will plummet. The only patent applications submitted will be those that have merit. This will allow the patent office to do their due diligence (something that is NOT done now).