Interview: Public Patent Foundation's Dan Ravicher [LWN.net]

December 23, 2003

By Pamela Jones, Editor of Groklaw

While the SCO saga is absorbing our attention in the short term, many are concerned about software patents and they worry that the real test for GNU/Linux will be in the future, from patent lawsuits. There have been numerous patents granted that to programmers seem to have been wrongly issued. The Amazon One Click patent springs to mind. Now Microsoft has announced it will be charging for use of the FAT filesystem, and that too makes some worry.

The Public Patent Foundation has recently been established for the purpose, as its web site puts it, of protecting "civil liberties and free markets from wrongly issued patents and unsound patent policy by providing those persons and businesses otherwise economically, politically, and socially deprived of access to the system governing patents with representation, advocacy and education."

Dan Ravicher is the patent attorney -- and programmer, incidentally -- who started PubPat, and he is its Executive Director. He was kind enough to answer some questions about patents and the work his organization is doing to educate the public and counter patent abuses. He says he is looking into the Microsoft FAT patents situation and has about a hundred pieces of prior art which were not reviewed by the examiner which they are currently reviewing. Dan was kind enough to answer the following questions.

What made you decide to start your foundation and can you tell us what it does?

The patent system is being abused by private actors to the detriment of the mostly unaware public. Our health, our freedom, and our economic prosperity are all under assault from bogus rights meted out to the few with the power and expertise to game a system originally established hundreds of years ago to promote progress within society as a whole. The government, through primarily a captured patent office utterly failing to achieve its mission and skewed policies implement into patent law by Congress and the courts, is not just failing to defend the public interest from abuse of the patent system, but is complicit in and supportive of such efforts.

In information technology industries, abuse of the patent system means illicit restraint of civil liberties and unjustified disproportionate burdening of small businesses. In life science industries, abuse of the patent system has even more devastating results, including the exacerbation of pain and suffering by those who cannot afford medical technologies covered by undeserved patents. This situation is abhorrent and the Public Patent Foundation is beginning a campaign against such abuses.

PubPat's four core activities are (1) challenging patents that threaten the public's health, freedom, or other interests, (2) helping small businesses defend themselves from patents being asserted against them, (3) establishing patent commons within markets crippled by patent thickets, and (4) educating the public regarding these issues and advocating for reform of the patent system.

If you plan on contesting any patents, can you tell us what patents you have in mind currently? And what would the process involve, from your standpoint?

At the moment we have under consideration several patents, including Microsoft's FAT patents, the Optima patent on CD burning, and a patent on co-transformation and protein production. Upon completing our review, there are many ways to neutralize the harmful effects of a patent, including asking the Patent Office to revoke it and publicizing ways to avoid infringing it.

To expand on one of the examples above, the Microsoft FAT patents are part of Microsoft's first attempt at building a licensing line of business akin to the one rolled out by IBM several decades ago. This causes concern for us because Microsoft is an admitted monopolist with a proven track record of driving competition from various markets through any mechanism available to it. They may now be focussing on patents as yet another avenue to foreclose competition, including specifically that from free software.

Beyond these atmospheric concerns, our analysis of the FAT patents has produced a substantial amount of prior art that was not before the patent office when it issued those patents to Microsoft. For a company with a nefarious past to be seeking revenue for patents that very likely did not deserve to be issued, is a malign scenario indeed. PubPat intends to ensure that the public's interest in being protected from such bahavior is properly represented.

Should there be software patents at all?

Many feel passionately about this issue. As a empiricist, I infrequently speak in categorical broad-brush terms unless presented with sound data and analysis to support a particular conclusion. With respect to software patents, everyone can agree that none which fail to meet the requirements of novelty and unobviousness should be granted or maintained. Beyond that, I have grave concerns about the lengthy term of patents being applied to technologies with short life cycles, especially those with life cycles shorter than the term of the patent. Such patented technologies never provide a public benefit, because by time the patent expires, the technology is no longer useful.

One thing the Public Patent Foundation is doing is compiling the data and performing the analysis I mentioned above, so that all reasonable persons can be presented with evidence supporting or condemning the policy decision made by the courts that "anything under the sun made by man" is patent eligible.

What is a "wrongly issued patent"? Should patents only be issued for a demonstrable, produced invention?

A patent can be "wrongly issued" for several reasons, including that the patent office was not aware of significant prior art during the examination process or that the patent office simply made the wrong conclusion regarding whether or not the patented technology was new and unobvious. I'm unsure what you mean by "demonstrable, produced invention", but the current standards of novelty, non-obviousness, and reduction to practice are good standards. The problem arises from either a lack of evidence on which to base a judgment as to whether something is new, unobvious, and reduced to practice, or a lack of competency in making those judgments.

Should the inventor state/swear that they intend to use the patent?

Many countries have patent laws that force a patentee to exploit her invention, else it becomes subject to a compulsory license at a minimum royalty rate. Such a rule is better than what we have in the United States, which does not require exploitation of patented technology. At the same time, however, such a shift may penalize small businesses who may not have access to the resources necessary to exploit a certain technology. Such small player patentees would have their leverage in negotiating a license with a larger competitor undercut by the statutory compulsory license.

It seems like many patents these days involve "good ideas" which are never implemented by the patent holder. Should "inventors" of software and/or business methods be required to provide evidence that they've made the system work before a patent is granted?

Patent law requires a patent applicant to reduce the patented technology to practice prior to applying for the patent; else any patent resulting from the application is invalid. To reduce a technology to practice, the patent applicant must either actually create the technology or describe it in such detail that one of ordinary skill in the art with the requisite resources could create the technology without undue experimentation. For instance, if you invent a time machine, but can't afford to make it, you can still get a patent so long as you tell others how to make it with sufficient detail such that they can successfully make the time machine at least 70-75% of the time. If, however, your instructions are insufficient for one of ordinary skill in the art with requisite resources to create the patented technology at least about 2/3rds of the time, then your patent is invalid for what is called "lack of enablement."

What about patents granted for obvious methods and technology? Should a patent be more than a unique design of a commonplace item such as a document or file?

The law requires a patented technology to be both new and unobvious. The crux of your question resides in defining the term "unique." If something is "unique" enough that ones of ordinary skill in the relevant art recognize it as being a new and unobvious technology, then current patent policy suggest rewarding the publication of that technology with a patent. Otherwise, the developer will keep the technology secret and other members of society will not be able to learn from and improve upon it.

What is the international impact of American patent law on world business?

First, half of the world's economy takes place in the U.S.. That fact alone means that U.S. patent law directly regulates half of all the world's business. Second, through international treaties, many of the policies of U.S. patent law have been adopted and implemented by other countries. This results in regulation of business wholly outside the U.S. closely mimicking the regulation of business within the U.S..

Computers are extensions of the human brain; computer storage is an extension of human reading and writing; electronic communication is an extension of the human voice. How do you feel about patents which use computers to do things that humans have been doing for millennia?

A patent cannot cover pure functionality; else it is invalid for indefiniteness. Rather, a patent can only cover specific structure used to accomplish a particular function. As such, it is only the structure that is patented, not the resulting function. Many people misunderstand this very important facet of patent law because sometimes, especially for the most publicized patents, the structure covered by the patent is the only known structure for accomplishing the particular function. This leads people to assume that the function itself is patented, which is not the case. Designing around patents is highly encouraged in patent law, and someone else is free to learn from the patent and come up with different structure for accomplishing the same, or a substitutable, function.

If a patented technology accomplishes a very old function, but with structure that is new and unobvious, then that satisfies the requirements for patentability. Further, one may need to recognize that functions are not necessarily the same simply because their result is the same. For instance, few humans who can do in a day (week, year) the complex calculations machines do today in mere nanoseconds. The function, in that case, is not getting the answer; it is getting the answer in virtual real time, which is something that humans have never done.

Do you feel that public discussion should be allowed before a patent is granted?

Public comment on patent applications prior to issue is an idea with some merit. Such is the law in many foreign countries, and recently the patent office abolished its prohibition on receiving third party correspondence regarding patent applications. However, if the process of pre-issuance public discussion includes a mechanism for third parties to delay the patent application from issuing, that mechanism might become unjustifiably abused and manipulated, particularly by larger corporations who can afford to "hold-up" a smaller companies "crown jewel patent."

(Log in to post comments)