Original: http://www.heckel.org/Heckel/ACM%20Paper/acaddeb.htm

IPC LogoIntellectual Property Creators         

The Academic Debate: Considered Opinion and Advocacy

The Mainstream View

Donald Chisum has written the standard reference on patent law[9] and is frequently quoted in judicial decisions. His expertise is in integrating patent decisions into a coherent view of the patent law across mechanical, electrical, chemical and other technologies as they are handed down. His reputation rests on the soundness of his analysis in predicting how courts will rule.

Chisum views software as one of many technologies and says software is almost as patentable as anything else and to the degree that it is not, it should be[8]. He has two concerns. First, the current decisions uphold most software as patentable in a way that forces patent lawyers and the technical community to focus on legal technicalities rather than the technical ones. Second, he questions whether lack of patentability will create underinvestment in software innovation as compared to other technologies.

His approach is similar to Judge Schwarzer's in determining how to calibrate the credibility of possible "junk science" in a courtroom. The question is not "Does this make sense in isolation?" but "How does it fit into an organized body of knowledge?"[44]

The Contrarian's View

The League provides no citations for its position in Against Software Patents. But a similar article by the same authors[15] has four citations: Mitchell Kapor's congressional testimony and articles by lawyers Pamela Samuelson[39, 41] and Brian Kahin[26] let us see how the anti-software patent position fits in with "an organized body of knowledge."

While patent lawyers conversant with software are in virtual unanimity that software should be patentable, neither of these lawyers is a patent lawyer.

As background we should consider the legal education most lawyers get. First, very few study patent law and are not exposed to the fundamental concept that patent rights are property rights. Second, most law students do take antitrust law where they learn that monopolies are illegal. Later, when they discover patents they often mistakenly see monopolies. Third, law students are taught to be advocates. Their job is not to make a considered opinion, but to present their case as best they can; someone else-the judge, Congress, the public- makes decisions.

This author's opinion is that those who took an anti-patent point of view did it as a knee jerk intuitive reaction. Away from the mainline of software business and patent law, distrustful of monopolies and inexperienced with intellectual property, these lawyers and software developers found support in one another. This view gained respectability as a contrarian's view and is always welcome in legal journals and at conferences.

Samuelson argues that the basis of software patentability is weak and says the primary issue is a policy one[39]. She presents two arguments against software patentability and Kahin a third:

A. The industry has flourished and innovated without patents.

A more accurate statement would be: "The industry has flourished and innovated with little realization in the software community that patent protection was available."

The growth of the software industry was not due to differences between software and other technologies, but to its synergetic relationship with the computer industry and the new business opportunities created by the computer's rapidly decreasing costs. Until recently software has been seen by computer companies as a loss leader. Software created the demand for computers which was, and still is, the dominant industry: but software developers were to be kept fat, dumb and happy: salaries were high, patents were not mentioned, and there was lots of technology to play with.

Accidental Empires shows that the personal computer software successes were achieved by amateurs who were lucky enough to be in the right place at the right time[12]. The successful early PC software companies (a) marketed innovations pioneered by others and (b) aggressively pursued their own intellectual property rights: Microsoft's MS/DOS is a derivative of CP/M; and Lotus' 1-2-3 of VisiCalc. Had Digital Research, VisiCorp and Software Arts asserted intellectual property rights as aggressively as Microsoft and Lotus, they might not have been eclipsed by them.

In the early stage of the industrial life cycle, the first person in his garage who acts on the opportunity starts with the biggest and most established company in the business. When there is no established competition, new companies can compete without patents. As the industry matures it becomes difficult for new companies to enter the market without a sustainable advantage such as patents.

The free enterprise system rewarded entrepreneurs whose personal computer software companies were successful, as it should, but that success should not blind us to its nature-bringing the innovations of others to market. We expect that if patents had been more widely used by the software industry, the true innovators would have received a fairer share of the rewards, thus rewarding innovation as well as business savvy. Had patents been more widespread, the software industry would have been more profitable, it would have grown with less of a boom-and-bust cycle albeit less rapidly and there would have been a greater diversity of software product categories and features.

To base a software intellectual property system just on the experience of the last ten years would be like raising teenagers in the expectation that their childhood will be repeated?

B. Many in the software industry say software patents will discourage innovation.

Samuelson says,

"... it is primarily from the widespread concerns about the effects of patents from within the industry and the technical community that she has pursued this study questioning the patent protection for computer program related inventions."[39]

Samuelson addresses this perception by conducting a survey to see how widespread the perception is, at least in the related area of user interface copyrights[42], rather than attempting to determine if the perception reflects reality. In reporting the perception she gives it more credence, reinforcing any error, and creating more concern. It would seem to be more constructive to research the effect of patents on innovation and business formation in other industries to see how it might apply to software, and analyze software patents that exist for their effect on innovation and new business formation as we do.

It makes as much sense to devise theories of innovation by polling the software community as it does to devise laws of physics by polling people who walk. The intuitive answer is not necessarily the correct one. Ask people, "Assume you are walking at a steady pace holding a ball, and you drop the ball. Will the ball hit the ground in front of you, in back of you, or next to you?" most people will say "in back of me."[32] A physicist will tell you that Newton's laws predict, "next to me," and can perform an experiment to prove it.

In contrast to Chisum and the mainstream of patent law where software is viewed from a broad perspective of many technologies, Samuelson views the law from inside the software community looking out. She has always advocated narrow protection- arguing as late as 1984 that CONTU's recommendation that software in machine readable form be copyrightable was ill considered [40 ]

Samuelson acknowledges that the conditions that promoted software innovation until now may be different from those that will promote it in the future.

Samuelson proposes to design a special (sui generis) form of protection for software, saying, "It is possible to design a law that is appropriate to the kind of subject matter that software is." How can one be certain of one's ability to build a skyscraper, if one is uncertain if one is building on bedrock or marshland?

Writing laws is like designing systems. They will have desired and undesired, intended and unintended consequences. The current law, especially Benson forces patent lawyers into a Catch 22 dilemma: Write a straightforward patent, and get it rejected as a mathematical algorithm; write one that is patentable subject matter and it will not be straightforward. Attempting to make only software unpatentable will no more prevent practical software patents from issuing and being enforced than prohibition will eliminate alcoholism. Samuelson laments that "patent lawyers [do not] claim software-related inventions in a straightforward manner[39] ; but this is like blaming a program's users or the market for not behaving as expected-a sure sign of an inexperienced designer. Law, software, and marketing strategies must be designed the same way the Constitution was: based not on how we would like people to behave, but on how self interested people will actually behave.

Academics such as Samuelson who pontificate on software patents and want to create sui generis system of protection seem ready to reinvent the wheel before they understand how a wagon works or the infrastructure of the highway system. Most show no knowledge or understanding of the processes of innovation over hundreds of years in a variety of technologies. Most have little, if any, experience prosecuting (filing), analyzing, or litigating patents.They argue that software is different from other technologies, not from the perspective of the history of innovation and patents over a range of technologies, but from the myopic view of the development of a single technology, largely in a single software marketplace in an atypical decade-the personal computer market in the 1980s.

Ben Franklin described a professor who was so learned he knew the word for horse in five languages: equus in Latin, caballo in Spanish, cheval in French, cavallo in Italian, and Pferd, in German. He then went out to purchase one, and returned with a cow. Given your experience observing complex software systems develop and how the hype manifests itself in reality, when you hear sui generis protection systems being proposed as transport into the twenty-first century, ask yourself: "Will I ride horse or a cow?"

Taking a contrary position and fighting for it, as Samuelson does, is in the highest tradition of the law. In part it is because truth emerges from the debate and if there is no one to debate with, a poor sort of truth will emerge. Yesterday's contrarian view may emerge as the mainstream view. It is like having advocates for a programming language like FORTH. It forces identification and discussion of issues and influences the industry-PostScript is Forth-like. But software developers seriously consider programming in FORTH only in unusual circumstances.

C: Software helps disseminate information

Brian Kahin, a research fellow in the Science, Technology and Public Policy Program at Harvard University's Kennedy School of Government offers a different reason software should not be patentable [25]:

A deeper, more disturbing problem in patenting programs was barely evident before computers became ubiquitous personal tools ... the computer has developed into a medium for human expression and a mediator of human experience. Thus, what is increasingly at stake in software patents is the generation and flow of information.

This "barely evident" "problem" was addressed by Lincoln who, in his Lecture on Discoveries [23], said:

certain inventions and discoveries occurred, of particular value on account of their efficiency in facilitating all other inventions and discoveries. Of these were the arts of writing and of printing-the discovery of America, and the introduction of Patent-laws."

Lincoln not only believed that the patent laws encouraged innovation, but he anticipated Mr. Kahin in realizing that inventions which promote dissemination of information are particularly important. Since Lincoln's speech, patents seem to have encouraged many inventions which engender the "generation and flow of information": the telephone (Bell), phonograph (Edison), movie camera (Edison), Xerography (Carlson), radio (Armstrong and Marconi), Phototypesetting (Scheffer), and TV (Philo T. Farnsworth). Even the birth of patenting, Aldus Manutius, the famous Venetian scholar printer for whom the desktop publishing company is named, received two patents-one on a form of Greek type-and a ten year monopoly to use the italic font which he invented .

Having discovered that he is treading in the footsteps of Lincoln, albeit in the opposite direction, Kahin might, in planning his future travel, consult the work of his colleagues at the Kennedy School of Government which has guided our analysis[33].

 

(Last updated on 10/06/98)

IPClogoIntellectual Property Creators                       www.ipcreators.org
101 First St., Suite 425
Los Altos, CA 94022
Voice: 650/948-8350
© Paul Heckel 1995-7