Peer-to-Patent Pilot Launches

Red Hat is a proud sponsor of and contributor to the pilot program of the Peer-to-Patent project, which launched June 15, 2007. This project, developed by New York Law School’s Institute for Information Law and Policy with the cooperation of the United States Patent and Trademark Office (PTO), will test an approach for enabling the public to participate in the patent-examination process. While this pilot will have limited scope - one year, 250 software patents - we hope that the data we gather will demonstrate the value of public inclusion in the patent-examination process, enabling the PTO to someday extend it to all filed patent applications.

Understanding the significance of this pilot requires an understanding of the current patent-examination process. To get a patent on an invention, the inventor first files a patent application that both (1) describes the invention in sufficient detail that one “skilled in the art” can make and use the invention, and (2) includes one or more short, numbered paragraphs known as “claims,” each of which uses legalese to put the public on notice of exactly what the inventor considers to constitute his invention. Performing everything specified in any single claim of the patent results in infringement of the patent.

The law specifies that the PTO should award patents only for technologies that are new and nonobvious contributions over the “prior art”–that is, what people had done previously, as demonstrated by prior patents, documents or devices. Accordingly, the PTO examines patent applications to be sure that they meet the requirements of being new and nonobvious before granting patents on them. The PTO, however, has a large backlog of applications that require review, so a few years typically elapse before an Examiner at the PTO even picks up a filed patent application.

At that point, the Examiner searches for prior art demonstrating that the claims of the patent application do not define new and nonobvious technology. If he finds such prior art, he rejects one or more of the claims in a written document referred to as an “Office Action.” At that point, in a written reply, the applicant may either (1) amend the claims to more clearly define his invention over the prior art, or (2) explain to the Examiner why the rejected claims differ from what the cited prior art discloses. Several rounds of this back-and-forth process may occur. If the applicant ultimately persuades the Examiner that the claims contain a new and nonobvious contribution over the prior art, then the Examiner notifies the applicant of his right to a patent in a “Notice of Allowance.” If the applicant subsequently pays an issue fee, the applicant receives an actual patent a short time thereafter.

Those affected by patents have become increasingly aware of limitations on the PTO’s ability to do its examination job effectively. One problem is that the PTO is so overwhelmed with patent applications that Examiners do not have the time to adequately search, understand and apply the prior art to which they have reasonable access. In other cases, the Examiners do not even have reasonable access to certain kinds of prior art that would help them to reject patent claims that do not define new or nonobvious contributions over the prior art.

This is an especially tough problem with respect to software patents. Software was not generally accepted as patentable until the mid-1980′s, and it was not until the mid-1990′s that software patents were pursued by most companies in the software industry. Consequently, while Examiners can most easily find and apply prior art patents in the limited time they have to devote to examining any given patent application, much of the prior art for software technologies exists as actual programs or academic papers rather than patents.

The result? The PTO erroneously issues patents that should not have been granted because they do not meet the novelty and nonobviousness requirements of patentability. While procedures exist for challenging the validity of such patents, they are expensive. Such erroneously issued patents result in unnecessary and often exorbitant litigation, licensing and/or review costs.

The goal of the Peer-to-Patent project is to demonstrate a framework for helping the PTO do a better job of examining patent applications by including the public in the examination process. In the pilot, the public will have the ability to provide prior art for applications that are part of the pilot, as well as comment upon the applicability of identified prior art to the claims of the pending patent application. The project will then pass along the best of this information to the PTO to assist the Examiner in his patentability review.

Why is Red Hat so excited about this project? Red Hat has been a longtime critic of software patents. For reasons already noted, erroneously issued patents have been especially numerous in the software arts, leading to unnecessary costs in the software industry in analyzing, litigating and licensing patents that never should have issued in the first place. We hope that the Peer-to-Patent project demonstrates that the community review of patents is a viable way to improve patent quality and thereby deal with some of the problems that software patents pose.