The problem with PUR lies in a perilous deviation from the basic exchange that takes place under the patent bargain -- in exchange for making a new, novel invention public, the government grants the inventor a limited exclusivity over the invention for a fixed period of time. Upon invention, the inventor has three basic options to consider: 1) participate in the patent system’s quid pro quo, teaching and sharing the invention with the public, 2) forgo patent rights, but simply publish the invention (called "defensive publishing") so that others will learn about the technology, thereby creating prior art so that others cannot patent the idea, or 3) keep the invention secret, but risk someone else inventing a similar invention and obtaining patent protection and the exclusive rights that come with it, creating a disincentive for keeping inventions secret.
PUR as contemplated by the FTF proponents would fundamentally alter this dynamic as it would allow an accused infringer to demonstrate their prior use of the patented technology to avoid infringement of a subsequently patented invention, even if the accused had not disclosed the invention to the public. It would transfer enormous risks to those who participate in the patent system from those who do not (for there would be no way of knowing who was hoarding what secrets and where). It would fundamentally shift the "reward" from those who make their inventions public to those who keep their ideas secret. The patent bargain would be broken, as no exclusivity would be assured in exchange for disclosure by the patentee.
Arguments of "fairness" to those who make substantial investments in secret are simply a non sequitur. Means for achieving fairness have already been established by society’s acceptance of the patent bargain. One cannot have it both ways.
In essence, PUR would make prior use akin to prior art -- equating, wholly illogically, a secret with a public good. And worse, proponents of PUR actually want (and have tried pushing for) PUR to go further -- considering it as prior art in validity analysis of the patent. Doing so would allow the secret to be used not only to avoid infringement, but to summarily invalidate an otherwise properly granted patent! This would allow private and secret prior use to destroy a public patent right -- and eliminate the patent holder’s rights to that patent not only with respect to the prior user, but with respect to the rest of the world.
Read the whole article. Good points. Kudos also to Greg Aharonian who shared the essay in his outstanding PatNews newsletter today.