|
The patentability of software remains an issue debated only by a few players in the technology industry. Academics, the courts, policy makers, and general industry take as given that patents can and should be granted for software inventions. The question now is how to optimize patent policy to facilitate technological innovation.
In a recent paper, Mark Lemley and Dan Burk state: “we accept as given (that)… patent law has a positive role to play in fostering software innovation, but argue that this will not occur as the state of software patenting currently stands.” Designing Optimal Software Patents. IPRs In Frontier Industries: Software And Biotechnology, Hahn, ed., 2005. How well policy makers and the courts address software will determine if software patents “stimulate innovation investment or disastrously stif(le) innovative activity.” 89.
One of the patent doctrines Professors Lemley and Burk touch on is that of the obviousness standard, which Professor Duffy and others consider perhaps the most critical aspect of patent law. To Lemley and Burk, for software innovation the doctrine is especially important, as suggested previously by Robert Merges: …the non-obviousness doctrine should be adjusted commensurate with the costs of innovation in a given industry...
The higher the cost to bring an invention to market, the bigger the potential reward should be, in order to attract the necessary investment. Bigger potential rewards can be provided by increasing the likelihood of obtaining a patent, and the likelihood of obtaining a patent can be increased by lowering the obviousness barrier...
Conversely, the obviousness barrier should be heightened where the cost of innovation is lower and the need for a patent incentive is less. Because innovation in software is relatively less uncertain than in (other) industries …Merges’s economic framework suggests that the non-obviousness bar should be rather high. 92. Upon economic analysis of how innovation proceeds in the software industry, Lemley and Burk decide that the standard of obviousness should be lowered from that currently applied by the CAFC, based primarily on the prominence of cumulative innovation in the industry. The cumulative innovation model predicts that innovation will rely on many, small incremental improvements on existing inventions. Cumulative innovation theory suggests that patent protection for incremental software inventions should be relatively easy to acquire in order to reward incremental improvements, implying a somewhat lower obviousness threshold. 93. The recommendation from Lemley and Burk to lower the standard of obviousness for software seems non-intuitive and contrary to practices in the industry. Often, small incremental improvements occur much more rapidly than typical processing time at the USPTO, when a patent finally issues, several new generations of improvements could have passed (and already become irrelevant). Further, not every incremental improvement should be considered “innovative” to deserve patent protection. Too often, insignificant changes to code are deemed innovations, even when there is no market or economic impact. Granting such patents may result in the over-issuance of patents that make no technological contribution to innovation. Finally, that a single product may be covered by many patents is already a complication for some software developers. Issuing more patents by lowering the obviousness standard will only increase transaction and licensing costs in the industry.
posted by Noel Le @ 1:19 PM | Patents
Link to this Entry |
Printer-Friendly |
Email a Comment | Post a Comment(0)
|
