Writers, Programmers, and Patents

by on March 26, 2007 · 10 comments

Don Marti has an excellent analogy to help illustrate what’s wrong with software patents:

If Victor invents something, and I describe it in prose, I’m not infringing. If he invents something and I build it as hardware, I am. But if I do something in between between hardware and prose—”software”— where do you draw the line of where he can sue me? If Dr. David S. Touretzky doesn’t know where you draw the line between “speech” and “device” how should the courts know?

All of the arguments for software patents work just as well for prose patents. Just as a software patent covers the algorithm, not the code, a prose patent could cover the literary device, sequence of topics, or ideas used to produce some effect on the reader…

The debate over software patents isn’t just an attempt to set one arbitrary line between the patentable and the unpatentable. It’s about resisting the slide toward higher and higher transaction costs that happens when patents creep into places where they don’t make sense. We have algorithm patents but not prose patents because lawyers and judges use analogies and other prose inventions more than they use algorithms.

Quite so. I think the reason you see such violent and near-unanimous dislike for software patents among computer programmers is that it’s not an abstraction for them. For most people, software is just a magical icon that sits on their desktop and does stuff when they double click on it. The question of whether software should be covered by patents is akin to debates over who owns the moon: intellectually interesting, but not really relevant to their day-to-day lives. But what computer programmers see is that widespread enforcement of software patents would mean that a significant portion of their professional lives would suddenly require regular consultation with lawyers. This pisses them off in precisely the same way—and for precisely the same reasons—that patents on plot devices, analogies, literary styles, and other prose concepts would piss off writers.

  • dimitris

    I think it’s fine to treat prose as special. In fact, prose may hold a key to highlighting the big software/business process patent bug, obviousness, to laypersons.

    Non-programmers don’t understand the problem in its original domain, so do what mathematicians do: Transform it to a different domain:

    “Inventions” which can be communicated by prose that cannot reasonably claim copyright protection are inherently obvious:

    Hey, how about storing a shopper’s payment and shipping preferences as a cookie?

    Even if I claim copyright on the above prose, you can get around my one-click shopping “invention” by phrasing it as:

    At Tim’s E-store, we let you store a standing “Our Reference:” in your electronic purchase orders on our system, using your browser’s cookie system

    There are virtually infinite different, non-copyright-infringing texts that convey the one-click idea with enough information for a practitioner in the art to duplicate. So it’s obvious, as is, for example:

    We’ll ship a DVD to the customer, and the envelope will be reusable for the return trip. We’ll charge a flat fee per month for unlimited DVD rentals.

    (Ironically, it seems the latter “invention” precluded the “inventor” of the former from offering a competing service in the US. They did, in the UK, since the DVD business model patent was probably invalid. As was their one-click patent. Sweetness.)

    On the other hand, I can’t see how any prose that describes the IDEA encryption algorithm can be rephrased without producing a version that, for copyright purposes, is 100% derived. Neither I nor the lawyer are practitioners in the mathematical art, but we don’t need to be to understand that the mathematics in the description is what makes the invention work

    So, again, it’s not algorithms versus prose. It can be attacked, however, as the copyright relationship between two pieces of prose.

    Making the lawyers understand that is left as an exercise to the reader :-)

  • http://weblog.ipcentral.info/ Noel Le

    Hmmm. You make some good points here Tim, but one or two analogies make the argument less serious.

    ‘better than that pizzaright analogy you used one time though. Oh man…

  • dimitris

    I think it’s fine to treat prose as special. In fact, prose may hold a key to highlighting the big software/business process patent bug, obviousness, to laypersons.

    Non-programmers don’t understand the problem in its original domain, so do what mathematicians do: Transform it to a different domain:

    “Inventions” which can be communicated by prose that cannot reasonably claim copyright protection are inherently obvious:

    Hey, how about storing a shopper’s payment and shipping preferences as a cookie?

    Even if I claim copyright on the above prose, you can get around my one-click shopping “invention” by phrasing it as:

    At Tim’s E-store, we let you store a standing “Our Reference:” in your electronic purchase orders on our system, using your browser’s cookie system

    There are virtually infinite different, non-copyright-infringing texts that convey the one-click idea with enough information for a practitioner in the art to duplicate. So it’s obvious, as is, for example:

    We’ll ship a DVD to the customer, and the envelope will be reusable for the return trip. We’ll charge a flat fee per month for unlimited DVD rentals.

    (Ironically, it seems the latter “invention” precluded the “inventor” of the former from offering a competing service in the US. They did, in the UK, since the DVD business model patent was probably invalid. As was their one-click patent. Sweetness.)

    On the other hand, I can’t see how any prose that describes the IDEA encryption algorithm can be rephrased without producing a version that, for copyright purposes, is 100% derived. Neither I nor the lawyer are practitioners in the mathematical art, but we don’t need to be to understand that the mathematics in the description is what makes the invention work

    So, again, it’s not algorithms versus prose. It can be attacked, however, as the copyright relationship between two pieces of prose.

    Making the lawyers understand that is left as an exercise to the reader :-)

  • http://weblog.ipcentral.info/ Noel Le

    Hmmm. You make some good points here Tim, but one or two analogies make the argument less serious.

    ‘better than that pizzaright analogy you used one time though. Oh man…

  • http://www.codemonkeyramblings.com MikeT

    I still think the ridiculously low cost of entry into software development, combined with the potential profit to be had, negate the need for patents. It’d be like arguing for 17+ year monopolies on drugs in a world where a life-saving drug doesn’t cost $1B to make, but $10M-$20M, and can be put on the market within no more than three years of getting started.

  • http://dsgazette.blogspot.com False Data

    It seems to me that the really difficult issues with line-drawing are likely to arise in areas such as embedded software. Should an FM radio be patentable? What about a software radio that has exactly the same functionality with fewer parts? Should a defibrillator be patentable? What about one that shapes the voltage based on heart activity? Would it make a difference if the voltage-shaping feature were implemented in hardware or in software? Should the voltage-shaping feature be independently patentable? Even if implemented entirely in software? The lines get fuzzy because the distinction between hardware and software is fuzzy–there is an area of overlap in which you can achieve the same goal using either. You could choose a bright line rule saying no feature is patentable if implemented entirely in software, even if it accomplishes the same goal as one implemented in hardware, but you’d need policy reasons for making such a distinction.

    The second point you raise about having to interact on a daily basis with lawyers is valid but not unique to software. We might all benefit from empirical data on whether patents are a net gain or net loss to innovation, coupled with research to identify the significant traits of the industry that determine what effect patents are likely to have. For example, patents might slow development in rapidly moving areas such as software in which the industry norm is to protect trade secrets, or it may be that the effort we put into preventing reverse engineering and guarding our trade secrets is more expensive than having the lawyers working with the programmers. We might have a gut feeling on this question–I certainly do–but too much legislation in this world gets made on gut feelings that turn out to be wrong.

  • http://www.codemonkeyramblings.com MikeT

    I still think the ridiculously low cost of entry into software development, combined with the potential profit to be had, negate the need for patents. It’d be like arguing for 17+ year monopolies on drugs in a world where a life-saving drug doesn’t cost $1B to make, but $10M-$20M, and can be put on the market within no more than three years of getting started.

  • http://dsgazette.blogspot.com False Data

    It seems to me that the really difficult issues with line-drawing are likely to arise in areas such as embedded software. Should an FM radio be patentable? What about a software radio that has exactly the same functionality with fewer parts? Should a defibrillator be patentable? What about one that shapes the voltage based on heart activity? Would it make a difference if the voltage-shaping feature were implemented in hardware or in software? Should the voltage-shaping feature be independently patentable? Even if implemented entirely in software? The lines get fuzzy because the distinction between hardware and software is fuzzy–there is an area of overlap in which you can achieve the same goal using either. You could choose a bright line rule saying no feature is patentable if implemented entirely in software, even if it accomplishes the same goal as one implemented in hardware, but you’d need policy reasons for making such a distinction.

    The second point you raise about having to interact on a daily basis with lawyers is valid but not unique to software. We might all benefit from empirical data on whether patents are a net gain or net loss to innovation, coupled with research to identify the significant traits of the industry that determine what effect patents are likely to have. For example, patents might slow development in rapidly moving areas such as software in which the industry norm is to protect trade secrets, or it may be that the effort we put into preventing reverse engineering and guarding our trade secrets is more expensive than having the lawyers working with the programmers. We might have a gut feeling on this question–I certainly do–but too much legislation in this world gets made on gut feelings that turn out to be wrong.

  • http://my.opera.com/Vorlath/blog/ Vorlath

    Software should have the same laws as music. Can you patent music if you put it on a CD meant to be played on a CD player? I hope not. Software is the same. Instructions are nothing more than notes that activates a specific response in the machine whenever the ‘reader head’ passes over it. End of story.

  • http://my.opera.com/Vorlath/blog/ Vorlath

    Software should have the same laws as music. Can you patent music if you put it on a CD meant to be played on a CD player? I hope not. Software is the same. Instructions are nothing more than notes that activates a specific response in the machine whenever the ‘reader head’ passes over it. End of story.

Previous post:

Next post: