ABA Section of Business Law
Business Law Today
January/February 2001 (Volume 10, Number 3)
An idea whose time has come:
Patenting software and business methods
Culmination of a 25-year debate
By PETER TOREN
T rue or false: Patenting software or business methods will put a chokehold on Internet innovation.
Some recent critics have focused on the issuance of patents that they claim do not meet the statutory tests of novelty and nonobviousness. Others, such as the Stanford University law professor, Lawrence Lessig, and the founder of Amazon.com, Jeffrey Bezos whose company owns the patent for its "one-click" technology, worry that the granting of such patents may weaken innovation in cyberspace and, accordingly, question the wisdom of granting such patents.
In response, these critics have proposed such radical solutions as having Congress declare a moratorium on the offensive use of software and business-methods patents or shortening the lifespans of software and business methods patents from the existing 20 years to three or five.
Are these high-profile critics on the right track? Their position ignores the fact that the patentability of software and business methods has not occurred overnight. It is not the product of a single court decision as many critics claim, but the culmination of more than 25 years of evolution. Further, such critics fail to take into account that the real world has largely rendered the academic debate over the patentability of software and business methods moot: More than 40,000 software or software-related patents have been issued in the United States and thousands more are issued every year.
More important, besides being rendered moot by the sheer number of patents being issued in these areas, the criticism has gone on far too long and obscures the more important and fundamental issue: What steps should be undertaken to improve the quality of software and business methods patents?
Critics of software and business methods patents generally cite the 1998 decision by the Court of Appeals for the Federal Circuit in State Street Bank & Trust v. Signature Financial Services (149 F.3d 1368 (Fed Cir.), cert. den. 119 S. Ct. 851 (1999)), as the root of the problem. Contrary to this understanding, however, is that there are more than 100 years of history for business methods patents and more than 30 years of history for software patents in the United States. The State Street decision does not represent a radical departure from existing law, but is the result of a decades-long debate that involved multiple parties including the U.S. Supreme Court and the Patent and Trademark Office (PTO).
That debate regarding the patentability of software actually began in the mid-1960s with a Presidential Commission on the Patent System, recommending that the U.S. patent laws be amended to provide that computer programs not be patentable. Fortunately, this recommendation was never enacted by Congress, but the PTO quickly espoused this position and rejected most applications for software patents. However, over the next 30 years, a number of decisions by the Supreme Court and the U.S. Court for Customs and Patent Appeals, the federal court responsible for hearing appeals from decisions by the PTO, effectively lowered the threshold for obtaining patent protection for software.
In the wake of these decisions and, in particular, a 1981 Supreme Court decision, the number of U.S. software patents began to increase. From 1982 through 1995, the PTO granted 381 patents containing the word "software" in their titles. The number of such patents was limited during that time, however, because the PTO examiners still would begin analyzing a software patent application to determine if the claim recited a "mathematical algorithm." If it did, the application would generally be rejected without further inquiry as to whether the statutory requirements such as novelty, utility and nonobviousness had been satisfied.
The pace greatly accelerated in 1996 after the PTO published its Examination Guidelines for Computer-Related Inventions . These guidelines specifically state that PTO personnel "will no longer begin examination by determining if a claim recites a mathematical algorithm." From Jan. 1, 1996 through August 1999, the PTO has issued 570 patents with the word "software" in the title of the patent, almost 50 percent more than it had in the 20 years preceding the guidelines. Thus, the approach of treating software as unpatentable subject matter had been overcome more than two years prior to the 1998 State Street decision.
The debate regarding the patentability of business methods lasted even longer than the debate over software. As noted by the State Street court, the "business method" exception "merely represented the application of some general, but no longer applicable legal principle." ( State Street, 149 F.3d at 1375)
With the adoption of the 1952 Patent Act, the court argued, business methods patents should have been subject to the same legal requirements for patentability as applied to any other process or method.
The court then embarked on a painstaking analysis of the prior case law to show that the patent courts had never invoked the so-called business methods exception to deem an invention unpatentable. Thus, according to the State Street court, the patentability of software and business methods should be evaluated under the same standards used for all patentable subject matter, namely, that the subject matter must be useful, novel and nonobvious and shall be based on a written description that enables one skilled in the art to make and use the invention and sets forth the best mode contemplated by the inventor for carrying out the invention.
The history of patents and their impact on innovation also suggest that the concern that software and business methods patents will detrimentally affect the development of e-commerce and the Internet is equally unfounded. In a recent book, Rembrandts in the Attic , Kevin G. Rivette and David Kline analyze the issuance of U.S. patents from 1850 through 1998. During this 150-year span, they point to four distinct periods of increased patent activity, each corresponding to the introduction and growth of particular technologies and related industries.
The first began in the late 1860s and lasted through the 1880s rapid advances in the emerging steam, railroad and electric power industries. The second occurred from about 1902 through 1916 the birth of the auto and aircraft industries. The third occurred from the mid-1960s through the late 1970s the "revolution" in plastics and other synthetic materials, and the growth in the aerospace and computer industries.
The fourth began in the mid-1980s and is continuing today the emergence of microprocessor-based information sciences and recombinant DNA-based life sciences. As much as patents played a role in the three prior periods of rapid technological evolution, Rivette and Kline argue, so too should they play a role in the current knowledge-based evolution of the Internet.
Critics have also suggested that the issuance of undeniably weak software and business methods patents those that even a proponent of strong patent protection would have good reason to be skeptical about is a basis to justify their claim that the scope of patent protection for software and business methods should be denied or severely limited.
Thus, instead of viewing the issuance of weak patents as the result of system that needs fixing, critics have concluded that the problem lies with the nature of the patentable subject matter. In other words, according to these critics, since there are software and business methods patents that should not have been issued, the entire class of such patents is suspect.
The issuance of such weak patents, however, is not traceable to the nonstatutory nature of the subject matter, but to the inability of the system to properly evaluate the merits of the patent application. More specifically, the issuance of such patents has resulted primarily from the fact that the PTO was not equipped to handle the flood of software and business methods patents and also because of flaws in the patent-application system.
To begin with, the PTO was extremely slow in hiring patent examiners qualified in the software and business methods areas. That meant that examiners who did not have extensive experience and knowledge about these subject matters were determining the novelty and nonobviousness of such applications.
Second, the PTOs classification system has not historically been equipped to handle software patents. Instead of classifying patent applications as software, such applications were classified based on how the software was to be used. This made it very difficult for the examiners to find prior art.
Third, since software applications are relatively new, the PTO has not compiled a sufficiently large prior-art library. This problem is compounded by the fact that many software inventions may not be described in industry publications. In fact, software inventions, unlike those in more established scientific areas such as biotechnology and electrical engineering, are generally not described in professional journals that are subject to peer review. This makes it especially difficult for any one person to find prior art.
In addition, the patent application system was not designed to overcome these PTO shortcomings. For example, under present law, the presumption of the validity of a patent can only be overcome with "clear and convincing" evidence that either some previous invention or publication contained every element of the patent claim (novelty) or that the elements can be found by combining two or more prior-art references (nonobviousness).
This, combined with the high cost of patent litigation, makes it likely that companies that have been accused of patent infringement may decide it is more economical to settle a claim than it is to fight, even if the claim is based on a weak software or business methods patent.
The solution to the issuance of weak patents in the software and business methods area is not to bar the granting of such patents or to reduce their terms, but to institute changes in the patent process that would lead to improvement in the quality of business methods and software patents. In fact, the PTO already has begun to institute a number of changes that are designed to lead to an improvement in the quality of such patents.
It has issued an action plan that, among other things, calls for enhanced technical training for examiners, revised examination guidelines for computer-related inventions and expanded search activities for all patent applications in the business methods class.
The PTO also has announced plans to revamp the procedures by which a party may ask it to reexamine a patent on the basis of undisclosed prior art. Specifically, the plan provides that all reexamination proceedings will be handled by an examiner other than the one who handled the initial application and will also test the use of an expert panel to decide whether to reissue patents that have been examined.
A third policy change will require that prior art submitted by nonapplicants be reviewed at a higher level as a possible basis for director-ordered reexaminations. Currently, prior art submitted by the public is put in the file and not reviewed. Director-ordered reexaminations occur when patents have a sufficient level of notoriety and raise a concern across an entire industry.
While the above changes are certainly steps in the right direction, a number of other reforms could be undertaken to ensure that the patent system will continue to foster and encourage innovation.
Congress should ensure that the PTO has the resources to be able to hire and train a sufficient number of patent examiners in the software and business methods area and to create the necessary prior-art database to be able to properly evaluate such applications. This could be as simple as allowing the PTO to keep the income from filing fees since the federal government currently diverts millions of dollars every year to other programs.
The PTO should broaden the requirements to qualify as a patent examiner to persons who have a background in computers or business but may lack the level of scientific and technical expertise that is currently required.
The scope of the doctrine of equivalents in software-infringement cases should be narrowed to avoid preventing legitimate efforts to design around existing software patents. This doctrine allows a finding of infringement even when the accused product does not literally satisfy each of the patents elements, but does accomplish substantially the same result in substantially the same way. For a variety of reasons, it has been suggested that the doctrines impact in this area may be greater than in other areas and may result in limiting software innovation.
The defendants burden of proof to rebut the presumption of patent validity should be lowered from "clear and convincing" to a "preponderance of the evidence." This should encourage accused infringers to contest the validity of a patent rather than determining that it is simply cheaper and less risky to obtain a license from the patent owner.
All software and business methods patent applications should be made public 18 months after filing and any party should be permitted to respond to publication by submitting prior art or by filing a protest. Unlike the present law that permits the applicant to opt out by claiming that he or she does not intend to file the application abroad, this proposal would allow interested parties to bring to the PTOs attention prior art that may bear on the patentability of the claimed invention. The filing of such prior art or opposition also should not prejudice a partys right to later pursue civil remedies.
Congress should commission a study of the economics of Internet patents to determine if software and business methods patents are retarding the growth of the Internet. If that is the case, Congress could then consider more drastic steps to curtail the issuance of such patents.
The rigorous patent-application process has previously worked in other areas and should be given the opportunity to work in the Internet arena. While the patent system may need some fine tuning, the long history of patent law in the United States suggests that there is no sound reason not to protect software and business methods patents. There is no legal or other basis on which to distinguish these types of patents from all other subject matters that are eligible for patent protection.
History shows us that the granting of patents will likely lead to greater innovation. While the system is not perfect, we should have faith in it.
Toren is a partner with Brown & Wood LLP in New York City and teaches a course in cyberlaw at Hofstra University. He formerly was a trial lawyer with the Computer Crime and Intellectual Property Section of the Criminal Division of the U.S. Department of Justice. His e-mail is: firstname.lastname@example.org