“Peer to Patent”: Collective Intelligence and Intellectual Property Reform
Beth Simone Noveck[1]
Part I: Introduction
The patent system has not responded to the changing business, technology, and legal realities over the last two decades and as such does not now provide the support needed to foster innovation and fairness in intellectual asset management that the system was built to insure. The good news is that the United States Patent Office is actively seeking ways to remedy it and ensure that only worthwhile inventions receive the patent monopoly. Currently, underpaid[2] and overwhelmed,[3] patent examiners struggle under the burden of 350,000 patent applications per year[4] and a backlog of 600,000.[5] Though supposedly expert,[6] patent examiners are not versed in all the scientific disciplines.[7] Until relatively recently, even though software applications were pouring in, the Patent Office did not recognize training in computer science as a legitimate qualification.[8] The same is true now for nanotech and other state of the art sciences. Multiple patents have been given for the same invention or patents awarded for inventions discovered previously.[9] The patent examination process is one part of the challenge. The judicial review process that is intended to check regulatory dysfunction is not helping. The Federal Circuit, the specialty patent appeals court, rules in favor of patent holders[10] more often than not and awards large financial judgments for patent enforcement, spawning a new industry of predatory “patent trolls,” patenting for litigation not innovation.[11]
There are numerous reform initiatives but, for the most part, they attempt to fix the patent system ex post without addressing the ills that plague the examination process. The Patent Reform Act of 2005[12] would, inter alia, limit damages for infringement and institute a post-grant opposition procedure. Other “open source patent” projects focus on challenging undeserving patents after the fact,[13] promoting socially responsible licensing of patented technologies[14] or encouraging inventors to donate or license patented technology, especially AIDS and other life-saving drugs, to the public domain prior to patent expiration.[15] Yet another proposal suggests that we adopt different patent term lengths for different industries.[16]
But what if we could reform patenting ex ante and ensure better applications before the cost of litigation? What if we could make it easier to ensure that only the most worthwhile inventions got twenty years of monopoly rights? What if we could offer a way to protect the inventor’s investment while still safeguarding the marketplace of ideas from bad patents? What if we could give the scientific community a voice in determining whether an invention was truly novel or obvious? What if we could make informed decisions about the scientifically complex issues posed by patent law before the fact?
This article puts forward a proposal to reform the patent system by designing a new legal and software framework for patent examination. Such an institution builds upon what we have learned from the explosion of online and offline systems for generation collaboration, harnessing social reputation, collaborative filtering and other existing technology to create a web-based peer review system of scientific experts ruling on innovation. The goal of such a system would be to help the Patent Examiner find the right references and have access to those who can advise on how to combine them akin to having expert witnesses before the fact. This will help to winnow bad patents, narrow the claims of good patents and improve the quality of those inventions awarded a patent.
The proposal stems from a central theoretical insight: the key problem with the current patent process (and our regulatory structures, generally) is that its notion of expertise is mired in outdated technologies. Our administrative structures have been constructed around certain beliefs, namely that centralized administrators have the best access to information, that expert bureaucrats are the only way to produce dispassionate decisions and that making decisions in the public interest requires keeping the public at bay. As even Robert Reich has pointed out, agency decision making, by separating legal decision making from scientific expertise, “confounds the ideal of scientific policy-making on which the legitimacy of regulatory agencies is based.”[17] Given that public consultation has been difficult and time consuming, we have continued to build legal institutions around insular bureaucratic expertise that is premised upon a bygone material reality. Current practices sideline scientific expertise in part because of the difficulty in communicating with the public. We continue to trust in bureaucratic experience that does not work, instead of in the collective intelligence that the Net now makes possible.[18]
Technology presents us with the opportunity to move away from a conception of isolated administrative expertise. The Internet has already enabled large-scale collaborative processes like open source programming and the creation of the Wikipedia encyclopedia. This article argues for a vision of collaborative expertise whereby we connect the know-how of a large, trained and dedicated governmental staff with legal expertise to the wisdom of experts with deep scientific, subject-matter expertise. Using communication technology, we can create a new mechanism for distributed decision making on a large scale[19] that separates legal from scientific decisions. With procedures in place to distribute but interconnect these two forms of expertise, we can create new legal mechanisms to serve the public interest and reform, not only the patent process, but administrative decision making more broadly. The idea of scientific citizen juries, blue ribbon panels or advisory committees[20] is not new. But the suggestion to use newly available social reputation software – think Slashdot karma or eBay reputation points -- to make such panels big enough, diverse enough and democratic enough to assist the patent examiner is.
We have arrived at a unique moment in history when five factors converge to make this kind of community patent reform (and collaborative expertise) proposal possible: first, the state of patenting has become so problematic as to meet with almost universal opprobrium; second patent applications are published after eighteen months independent of grant, making it possible to consider open peer review;[21] third, peer review is widely practiced in the public sector (e.g. EPA, NIH, NSF); fourth, we finally have the social reputation and social networking technology to make peer review on this scale possible; and, fifth, we have the expertise with such endeavors as Wikipedia, Slashdot, Yahoo Answers, Linux, Apache and many more such collaborative decision making systems, both online and off, to be able to design and construct a new legal institution that brings the wisdom of experts to bear.
This proposal argues that this new type of digital institution that combines scientific and legal decision making using online tools will improve the quality of patent examination. Instead of one examiner, an application can have 1000! If we do so, the new generation of social software that enables commons-based production might not simply make it easier to get a date, it could bring about a better, fairer form of governmental decision making for patents and improve upon the constitutional promise to promote the progress of science and the useful arts in our democracy by ensuring that only worthy ideas receive the odious monopoly of which the first patent examiner, Thomas Jefferson, complained.[22]
How will it work?
The proposal is for a process that preserves the constitutional and statutory standards of patent law while remaking the legal practices that enable them through software. The so-called “peer to patent” Community Patent Review system would have two parts: it would make it possible for the community of practice in a given area of art to suggest prior art to the patent examiner; and, in the second part, it would convene panels of experts having skill in the art to assist the examiner. There is no doubt that, unlike other volunteer, peer review projects in academia or on-line projects such as Slashdot, a peer review system for patents implicates large fortunes and vicious competition. Such an institution must be designed to harness enlightened self-interest to drive participation.
This Article lays out a detailed draft blueprint for the Community Patent system. The next Part discusses the current state of patent law in the United States and the problems to which the current examination procedure gives rise. In short, review by a lone examiner is crippling our ability to assess innovation. Part Three provides some background on the existing procedure for patent examination while Part Four then explains the community patent process (or what I have nicknamed “peer to patent” peer review). It provides a draft specification of its features so that technologists can begin translating the design of the legal process into functioning software. Part Five responds to critiques and questions ranging from why will people participate to how will courts review the results. The Article concludes with a final section entitled “From Proposal to Prototype to Pilot” laying out the details of a pilot implementation to be adopted by the United States Patent Office. The aim is to create and deploy the peer to patent system, first in a pilot project to test the system on software and biotech patents and then, after refinement, full implementation by the United States Patent Office, phasing in deployment in those areas where the Patent Office is most overworked and understaffed.
There is some irony, of course, to creating open, peer review processes to augment closed, proprietary patent grants. We are applying techniques and lessons of open source programming to an arena that is the anathema to most open source adherents. Arguably, we are engaging the community in closing off innovation from the community. This makes sense, first, because undoing patents altogether, including software patents, is not politically in the offing. Therefore we must start by reforming the processes and making existing patents better quality. Second, this process invariably creates more information and enhances sharing of information about invention and discovery within the creative community. Third, opening up patents to community review will have unforeseen consequences on patent law and the nature of patents. We can hope that these openness reforms will set in motion a process of reform which will, not only improve patent examination, but by engaging the community of practice more actively in the fate of innovation, will change the way we perceive patents in our economy.
This also has far reaching implications and applications beyond the patent process. It implies a fundamental rethinking of our assumptions about bureaucracy and expertise. While Congress passes four hundred items of general legislation per year, federal agencies (of which the patent office is one) enact 4000-8000 regulations that translate those laws into the specific rules for everyday living: requiring seatbelts for cars, limiting the parts per billion of chemicals in the air, dictating the width of doorways in new home construction, restricting the use of dirty words on television. And federal agencies are only just the tip of the iceberg of work done by state and local agencies. While we have democratically-elected legislatures, we rely upon these non-elected “expert” bureaucrats because they are not swayed by political considerations. They have access to better, more scientific, more objective information. Or do they?
Done right, peer review for patents may point the way toward a new model for the administrative state. It is not the New Deal notion of the bureaucratic expert administrator nor the interest group representation of mid-century nor the analytic management model of recent years.[23] Each of these models suffers from the democratic deficit created by reliance on centralized information and expertise. Peer to Patent argues for a new paradigm of collective governance by the innovation community.
[1] Please send comments to bnoveck@nyls.edu.
[2] A patent examiner gets paid less than half the salary of a first-year associate prosecuting patents for a large law firm. According to the United States Patent and Trademark Office, a patent examiner starts at levels between GS-5 to GS-9, where the salary is between $35,548 and $49,729. GS-13 earns $78,018. See http://www.usptocareers.gov/home.asp (follow “FAQ” for the list of grades; and follow “Job Search” for salaries at grade level). Entry level salaries for first year associates in large New York law firms are around $125,000. Entry-level Attorney Salaries Remain Stable, The Daily Record of Rochester, August 30, 2005; see also NALP: The Association for Legal Career Professionals, Salaries and Compensation: Private Sector, at http://www.nalp.org/content/index.php?pid=147.
[3] America has twice the number of applications with the same number of examiners as the European Patent Office. See Adam B. Jaffe & Josh Lerner, Innovation and Its Discontents: How our Broken Patent System is Enduring Innovation and Progress, and What to Do about It 11-18, 131 (2004).
[4] The total of 382,139 patent applications in 2004 is more than doubled from 176,264 in 1990. See http://www.uspto.gov/web/offices/ac/ido/oeip/taf/us_stat.pdf (last visited Sept. 30, 2005).
[5] USPTO Oversight Hearing, Before The Subcommittee on Courts, The Internet, and Intellectual Property Committee On The Judiciary United States House of Representatives (2005) (The Honorable Jon W. Dudas, Under Secretary Of Commerce for Intellectual Property and Director Of The United States Patent and Trademark Office) at http://judiciary.house.gov/media/pdfs/dudas090805.pdf (“without any change to the system, the backlog of applications awaiting a first review by an examiner is expected to grow from the current level of approximately 600,000 to over 1,000,000 by 2010.”) (last visited Sept. 30, 2005).
[6] Required to become a patent examiner is required “a degree from an accredited college or university in Electrical Engineering, Computer Engineering, Computer Science, Mechanical Engineering, Chemical Engineering, Material Science Engineering, Biology and Organic Chemistry.” No advanced degree is required. See http://www.usptocareers.gov/home.asp (follow “FAQ”) (last visited Sept. 30, 2005).
[7] While there is legal training, there is no continuing science education program for patent examiners. “According to agency officials, examiners automatically maintain currency with their technical fields by just doing their job of examining applications, which they believe contains the most cutting-edge information. However, patent examiners and supervisory patent examiners disagreed and said that the literature they review in applications is outdated, particularly in rapidly evolving technologies. USPTO offers some voluntary in-house training, but the agency could provide no data on the extent to which examiners have taken advantage of such training. Moreover, patent examiners told us that they are reluctant to attend such training, given the time demands involved. In contrast, USPTO’s policy requires examiners to attend extensive training provided by the agency on legal issues on which examiners are periodically tested.” United States Government Accountability Office, Report to Congressional Committees, USPTO Has Made Progress in Hiring Examiners, but Challenges to Retention Remain, at 6 (2005), at http://www.gao.gov/cgi-bin/getrpt?GAO-05-720.
[8] Computer science became a recognized qualification in 1994 even though the USPTO was granting software patents since the early 1980s. Diamond v. Diehr, 450 U.S. 175, 177 (1981) (affirming the Court of Custom and Patent Appeals decision granting a patent for a computerized “industrial process for the molding of rubber products.”). See In re Alappat, 33 F.3d 1526, 1545 (Fed. Cir. 1994) (“a computer operating pursuant to software may represent patentable subject matter, provided, of course, that the claimed subject matter meets all of the other requirements of Title 35. In any case, a computer, like a rasterizer, is apparatus not mathematics.”).
[9] See Adam B. Jaffe & Josh Lerner, Innovation and Its Discontents: How our Broken Patent System is Enduring Innovation and Progress, and What to Do about It 4 (2004).
[10] “Judges appointed after 1982 voted to hold a patent valid 164 times out of 298, or 55.0% of the time. While judges appointed to the Federal Circuit, with its widely asserted pro-patent orientation, did vote slightly more often to uphold the validity of a patent than their predecessors, the numbers are quite similar. The Federal Circuit did hold patents valid much more often than its predecessor courts, but the difference cannot be attributed to judges appointed to that court at different times.” Mark Lemley & John R. Allison, How Federal Circuit Judges Vote in Patent Validity Cases, 27 Fla. St. U. L. Rev. 745 (2000). See also Rochelle Cooper Dreyfuss, The Federal Circuit: A Case Study in Specialized Courts, 64 N.Y.U. L. Rev. 1 (1989); Donald R. Dunner, The United States Court of Appeals for the Federal Circuit: Its First Three Years, 13 AIPLA Q.J. 185 (1985); As Professors Dan L. Burk and Mark Lemley remark, “the Federal Circuit has bent over backwards to find biotechnological inventions nonobvious, even if the prior art demonstrates a clear plan for producing the invention.” Dan L. Burk & Mark A. Lemley, Is Patent Law Technology-Specific?, 17 Berk. Tech. L.J. 1155, 1156 (2002).
[11] William M. Bulkeley, Aggressive Patent Litigants Pose Growing Threat to Big Business, Wall St. J., September 14, 2005, at A1 (discussion of impact of patent trolls on successful businesses). See, e.g., Martin LaMonica, Small Company Makes Big Claims On XML Patents, CNET (October 21, 2005) (small company seeking to extract licensing royalties from users of open standard XML), at http://news.com.com/Small+company+makes+big+claims+on+XML+patents/2100-1014_3-5905949.html?part=rss&tag=5905949&subj=news. This is one of myriad such cases of someone patenting an already known and fundamental technology and then seeking to extort royalties using a patent that should never have been granted.
[12] Patent Reform Act of 2005, H.R.2795, 109th Cong. (2005) (Introduced in House June 8, 2005).
[13] See the Adelphi Charter on Creativity, Innovation and Intellectual Property at http://www.adelphicharter.org/ (calling for new public interest standards for patenting). Free Ideas, The Economist (Oct. 13, 2005) (describing the Adelphi Charter). See also Electronic Frontier Foundation Patent Busting Project at http://www.eff.org/patent/ (last visited October 3, 2005) (“EFF is launching a Patent Busting Project to take on illegitimate patents that suppress non-commercial and small business innovation or limit free expression online. The Project has two components: A. Documenting the Damage…B. Challenging The Patents”); see also Sarah Boettinger and Dan L. Burk, Open Source Patenting, 1 J. of Intn’l Biotech L., 221-231, (2004).
[14] The goal of Science Commons is to “encourage stakeholders to create areas of
free access and inquiry using standardized licenses and other means: a 'Science
Commons' built out of voluntary private agreements.” at http://www.sciencecommons.org (last visited Sept. 30, 2005).
[15] Consumer Project on Technology, Patent Compulsory License Project, at http://www.cptech.org/ip/health/cl/
[16] See Mark A. Lemley, An Empirical Study of the Twenty-Year Patent Term, 22 AIPA Q.J. 369, 385 (1994) (assessing fixed and variable length patent terms).
[17] David Schoenbrod, Power without Responsibility 100 (Yale University Press 1993) (citing Robert B. Reich, Book Review of The Politics of Regulation, New Republic, June 14, 1980, at 36-37.).
[18] Yochai Benkler, Coase's Penguin, or, Linux and The Nature of the Firm, 112 Yale L. J. 369 (2002). See also The Digital Universe, available at http://www.digitaluniverse.net/ (offering collaborative web portals to foster expertise).
[19] Marko A. Rodriquez and Daniel J. Steinbock, The Anatomy of a Large Scale Collective Decisionmaking System, 2 (March 13, 2006) (directed, weighted semantic network that connects humans, their mental models, and their artifacts to enable swarms of particles to traverse the network and rank solutions to the problems facing the group), available at http://www.cse.ucsc.edu/~okram/research.html.
[20] Federal Advisory Committee Act (FACA), 5 U.S.C. App. § 2 (2005) (An act to authorize the establishment of a system governing the creation and operation of advisory committees in the executive branch of the Federal Government and for other purposes.)
[21] The Patent Reform Act of 2005, H.R.2795, 109th Cong. (2005) would make publication of patents universal. Under the current rule publication is the default but applicants may request an opt out by certifying that the invention disclosed in the application has not and will not be the subject of an application filed in another country, or under a multilateral international agreement, that requires publication of applications 18 months after filing.
[22] Along the same lines, Irving Wladawsky-Berger, IBM Vice President of Technology and Strategy said, “Through the U.S. Patent Office any idiot can get a patent for something that should never be granted a patent." InfoWorld March 7, 2005.
[23] Richard B. Stewart, Administrative Law in the Twenty-First Century, 78 N.Y.U.L. Rev. 437 (2003) (describing the major models of the administrative state in the twentieth century).