A Statistical Analysis of the Patent Bar: Where Are the Software-savvy Patent Attorneys?

• Jurisdiction: United States,Federal
• Citation: Vol. 11 No. 2009
• Publication year: 2009

Ralph D. Clifford, Thomas G. Field, Jr., & Jon R. Cavicchi ⁰

Among the many factors that impact the declining quality of U.S. patents is the increasing disconnect between the technological education patent bar members have and the fields in which patents are being written. Based on an empirical study, the authors show that too few patent attorneys and agents have relevant experience in the most often patented areas today, such as computer science. An examination of the qualification practices of the U.S. Patent and Trademark Office ("PTO") suggests that an institutional bias exists within the PTO that prevents software-savvy individuals from registering with the Office. The paper concludes with suggestions of how the identified problems can be corrected.

I. Introduction

A decline in the quality of U.S. patents¹ has been widely perceived. Often, patents that claim inventions of dubious novelty are issued while those describing important advancements are held hostage to the inability of the U.S. Patent and Trademark Office to competently process more recently developed technologies.² As a consequence, marketplace competition is directly and adversely affected.³

Commentators from academia, government, media and industry have identified several significant contributing factors to this problem.⁴ One causative factor is the increasing difficulty in locating prior art in the face of the changing technologies that are submitted to the Office.⁵ A second causative factor is the competence (or lack thereof) of the PTO to examine applications seeking protection for newer technologies with which the Office has little experience.⁶ The third causative factor, and the primary topic of this paper, is the increasingly inappropriate technical credentials held by the parties responsible for drafting the vast majority of patents: the patent attorney or agent (collectively referred to as the "patent bar").⁷

The technical credentials of the patent bar are examined using a four-step analysis. First, based on an extensive empirical study, we determine the current composition of the patent bar. Second, we evaluate the current rate of demand for different types of patents in the so-called information age. As expected, software patents are being sought at a blistering, ever increasing pace. Third, we determine, in comparative terms, whether the growth of the software-savvy fraction of the patent bar is matching the demand; we conclude that it is not. Fourth, we discuss the causes of this imbalanced patent bar composition. From our analysis, the PTO's unique credentialing process and the requirement that applicants pass a unique agency-administered bar examination (hereinafter "patent bar exam") excludes many of the brightest and best-trained computer science graduates from patent bar membership.⁸

Building on the analysis and our conclusions from it, we consider how the PTO has rejected past suggestions of potential problems. Accordingly, we propose ways that the PTO might be induced to solve the problem by recalibrating its criteria for admission into the patent bar.

II. The Non-Legal Education of Current Patent Bar Members

To determine the educational background of those admitted to practice before the PTO in patent cases, a Freedom of Information Act⁹ request was submitted to the Office for selected information from each application form filed by all current members of the patent bar.¹⁰ The information sought included all educational institutions listed by the applicant and the degrees earned with the date it was awarded as well as the date upon which the applicant was admitted to the patent bar.¹¹ in response, over the next year and a half, the PTO provided 54,897 pages of scanned images of the application forms filed by the 26,735 then-current members (as of May 5, 2006) of the patent bar.¹² Converting the PDF files into a computer-readable form, proofreading and normalizing the data,¹³ building a database, and doing the statistical analysis took considerable time.¹⁴

The top-fifty reported nonlegal fields of study of the people admitted to the patent bar are reflected in Table A. The most common degrees are in the hard sciences and engineering.¹⁵ When practitioners are grouped into five broad categories as reflected in Table B, an interesting, but not unexpected, pattern emerges.¹⁶ Approximately one quarter of the membership was primarily trained in chemical, mechanical, and biological studies; one sixth falls into the electrical sciences; and only one twentieth was educated in computer technology.¹⁷ In other words, approximately ninety percent of the patent bar is trained in chemical, mechanical, biological or electrical fields while less than five percent are trained directly in computer-related fields.¹⁸

That such a breakdown is unsurprising, however, does not mean that it meets current societal needs.

III. The Changing Face of Technology

When the first patent act was adopted in 1790,¹⁹ inventions primarily covered the structure, manufacture, and use of mechanical devices.²⁰ The dominance of such physical technologies²¹ continued into the early twentieth century even as the use of electricity began to emerge.²² Since the end of the twentieth century, however, the increasing power and speed of computer technology has facilitated expanding uses of software²³ to accomplish tasks that would be otherwise unachievable.

Although nonmechanical technologies have been the subject of many patents over the last few decades,²⁴ the patentability of an invention that only processes information has not always been recognized²⁵ and today remains controversial.²⁶ Beginning at least as early as the Johnson administration,²⁷ patents for such technology have been repeatedly opposed by the PTO.²⁸ As this paper is written, the scope of patents that are appropriate for software- and business-related inventions is again before the Supreme Court in the Bilski²⁹ case.

In the Federal Circuit's en banc Bilski opinion, the court adopts a "machine-or-transformation" test supposedly derived from Supreme Court precedent but says:

future developments in technology and the sciences may present difficult challenges to the machine-or-transformation test, just as the widespread use of computers and the advent of the Internet has begun to challenge it in the past decade. Thus, we recognize that the Supreme Court may ultimately decide to alter or perhaps even set aside this test to accommodate emerging technologies. And we certainly do not rule out the possibility that this court may in the future refine or augment the test or how it is applied. At present, however, and certainly for the present case, we see no need for such a departure . . . .³⁰

Judge Mayer dissented, saying:

The en banc order in this case asked: "Whether it is appropriate to reconsider State Street Bank & Trust Co. v. Signature Financial Group, Inc., 149 F.3d 1368 (Fed. Cir. 1998), and AT & T Corp. v. Excel Communications, Inc., 172 F.3d 1352 (Fed. Cir. 1999), in this case and, if so, whether those cases should be overruled in any respect?" I would answer that question with an emphatic "yes." . . . Affording patent protection to business methods lacks constitutional and statutory support, serves to hinder rather than promote innovation and usurps that which rightfully belongs in the public domain.³¹

The final observation in that dissent is, however, difficult to reconcile with 35 U.S.C. § 273. It explicitly provides a defense for infringers of business method patents.³²

As found in the State Street Bank case referenced by Judge Mayer, most business-method patents are essentially software patents.³³ In stark contrast is the dissent of Judge Newman, who writes:

The now-discarded criterion of a "useful, concrete, and tangible result" has proved to be of ready and comprehensible applicability in a large variety of processes of the information and digital ages. The court in State Street Bank reinforced the thesis that there is no reason, in statute or policy, to exclude computer-implemented and information-based inventions from access to patentability. The holdings and reasoning of Alappat and State Street Bank guided the inventions of the electronic age into the patent system, while remaining faithful to the Diehr distinction between abstract ideas such as mathematical formulae and their application in a particular process for a specified purpose. And patentability has always required compliance with all of the requirements of the statute, including novelty, non-obviousness, utility, and the provisions of Section 112.³⁴

Regardless of the decision rendered by the Supreme Court in Bilski, it is clear that the patent bar will be called upon to continue to deal with software-based technologies for the foreseeable future. The relative magnitude of that task can be determined by examining the quantity of patents being issued in various areas of technology.³⁵ Using the data collected from the PTO, Table A12³⁶ lists the number of patents issued by year according to technological classes established by the Office.³⁷ A multi-way cross-tabulation³⁸ was constructed to connect each such class to the nonlegal background of individuals most apt to understand the underlying technology.³⁹ A summary appears in Table E.⁴⁰

The number of patents best associated with the top fifty undergraduate majors is presented for five time periods.⁴¹ Some technologies have greatly decreased in rank over time. The most significant changes were found in agriculture which dropped eight places in rank from twenty-fifth to thirty-third, ceramic engineering which fell nine places from twenty-third to thirty-second, food technology which decreased eleven places from twenty-fourth to thirty-fifth, metallurgical engineering which dropped ten places from sixteenth to twenty-sixth, and petroleum engineering which fell ten places from nineteenth to twenty-ninth.⁴² Other technologies have significantly increased during the time...