The deadline for filing amicus briefs in support of the Federal Circuit’s attempt to trim back business method patents in Bilski passed on October 2. Many briefs have been filed, and much fuss has been made in the tech community, for business method patents are linked to the problem of software patents. Many software patents, such as Amazon’s 1-click order patent, are for business methods.
If the courts ultimately trim back business method patents, will this take some of the pressure off both tech and the patent system? Not as much as many in the tech community or the patent community would hope, for reasons I examine below. Patent reform is now being driven by business constituencies, and these constituencies are not good at all at working on big picture institutional problems. There, in short, is a not-seeing-forest-for trees problem.
Business method patents as such involve processes related to market research, loans, billing, underwriting, and so on. Software patents are just that, patents on software–perhaps software built into hardware, perhaps stand-alone software intended to run on any computer. In a nutshell, problems have arisen with these patents. Such patents are hard to define. There are many of them, perhaps too many. Large producer firms become a target of patent owners, some of whom seem to buy up patents mainly to litigate them.
Some caveats: These problems are not unique to software or business methods; they also arise with biotech, nanotech, bioinformatics, and so on. The problem is one of degree, not of kind. And no one has shown conclusively that business method or software patents are harmful on balance; several studies of software patents suggest that the good and bad balances out, or that software patents are a weak positive. Nevertheless, there are problems, and solutions are being sought.
Many solutions look, as does Bilkski, to subject matter limits and to sectoral reforms–getting rid of software or business method patents, or tailoring the law just for them. Another type of solution is more general, affecting the patent system as a whole, like changes to the presumption of patent validity.
Sectoral and subject matter reforms can get traction inside the Beltway. This approach promises gain with no pain; if software and business method patents just go away, pharma and biotech are unscathed, and might not bother to oppose the changes.
Note, though, that getting rid of business method patents (the best line of argument is probably that these are ultimately, like a piece of music or a book, playing to a human and not to a machine audience) does not necessarily get rid of software patents. Software’s main purpose is to make a machine do something involving electricity and optics and such; whatever the literary qualities of source code it makes little sense to distinguish it from object code (nor would doing so get programmers off the hook if all they did was produce source code… they would simply face liablity for contributory patent infringement as opposed to liability for direct infringement, on the same principle that the maker of a drug that turns into a patented product in the patient’s stomach faces liability for contributory infringement).
Furthermore, any sectoral approaches leave biotech, nanotech, and technologies beyond out in the cold, which makes no sense from a policy standpoint. It also might tend to be obsolete rather quickly, like the chip mask laws. Sectoral differences can come and go. For the moment, pharma clearly needs patents because it has big up-front research costs and must recoup those. But this might change… if medicines were tailor-made for individual DNA, for example, expensive clinical trials would look quite different and might become nearly obsolete. Software could change, biotech could change, nanotech could change, finance could change.
More generally, finding sensible rules to limit patentable subject matter is likely to be an uphill battle. Most economic studies find that the patent system as a whole is a positive for both developed and, generally, for developing countries, though for some sectors (software) a more weak positive than others. So one doesn’t necessarily want to default it to “off” for new technology. A law intended to cover new technology by its nature cannot always describe exactly what that technology should or should not look like in advance. A sectoral approach is doomed to make rules by hindsight, a particularly unsatisfactory outlook for forward-looking innovation. The law needs to be general… and a general law is likely to catch a lot of borderline stuff.
The better approaches to reform therefore are big picture changes that look at how to improve the system as a whole. These include adding more courts to hear patent appeals (so the Federal Circuit has someone to talk to); increasing the use of fee shifting in the patent system (or the legal system more generally) so that defendants are more willing to litigate strong cases; and rolling back the presumption of patent validity. But companies for which the patent system is a positive now are opposed to any change.
The better, more general, approach is therefore doomed, for the present, to take a back seat to fussing over whether or not an idea is too abstract to be useful; exactly what a patentable “process” is; whether source code should be treated like object code; whether code running in a machine is different from code merely being transmitted, and other arcane inquries, the results of which are unlikely to prove durable.
Having no dog in this fight, I can only watch it all unfold, and reserve the right to change my mind about any of these after reading the next 100 studies.
Further Reading:
“Patents and Innovation: Friends or Foes,” by Francois Leveque and Yann Meniere.
“Defining Software Patents – A Research Field Guide,” by Anne Layne-Farrar.
Josh Lerner and Feng Zhu, “What is the Impact of Software Patent Shifts?: Evidence from Lotus v. Borland.”
Martin Campbell-Kelly, “Not All Bad: A Historical Perspective on Software Patents,” Michigan Telecommunications and Technology Law Review, Vol. 11.
Iain Cockburn, Megan MacGarvie, “Entry, Exit and Patenting in the Software Industry.”
Anne Layne-Farrar and others, “Do Companies that Patent More Today Have Higher Revenues Tomorrow?”
The Technology Liberation Front is the tech policy blog dedicated to keeping politicians' hands off the 'net and everything else related to technology.