Thursday, April 28, 2005

Levin to testify on patent reform before House on April 28

Talk of patent reform intensifies. It's not clear that the discussion is well thought out, but it is well orchestrated.

from Yale Daily:

Levin to testify before Congress
Levin to push for reforms in patent system

Staff Reporter

Today [April 28, 2005] in the nation's capital, Yale President Richard Levin will be speaking not as a representative of higher education but rather as a specialist in intellectual property rights when he goes before the Intellectual Property Subcommittee of the House Judiciary Committee. He delivered similar testimony to the Senate Judiciary Committee on Monday, April 25.

Levin honed his expertise in intellectual property rights and the patent system as a graduate student and professor at Yale. Joining University of Pennsylvania management professor Mark Myers, who co-chaired the National Research Council's comprehensive review of the U.S. patent system, Levin will call for reforms that he said will improve the quality and timeliness of patent decisions. Levin said the current patent system is overwhelmed by applications, especially in areas of new technology.

"We concluded that on the whole the patent system is working well and does not need fundamental revision," Levin said in his prepared statement before the Senate Monday. "Yet we did note some causes for concern."

To combat burgeoning patent litigation costs, the report recommended creating a proceeding for parties to directly contest patents granted by the federal government for up to nine months after the patent is issued [that is, an opposition proceeding], instead of delving into lengthy and expensive court proceedings. The procedure is designed to weed out poor quality patents early on.

"That's the big part of the reforms," said Jim Crowne, a spokesman for the American Intellectual Property Law Association, which endorsed most of the NRC's reforms. "If you had to wait until there was a hard dispute between the parties, the litigation costs would be enormous."

The report also called for the U.S. government to grant patents to the first party which files a patent, as opposed to the first-to-invent basis that is currently used. Levin said moving to the first-to-file system, which the rest of the world uses, would eliminate years of time-consuming legal discovery, depositions, and testimony that can stifle innovation. [only in the context of interferences, which happen infrequently.] Such a move would also put the United States in step with the rest of the world.

But Andrew Krauss, president of the Inventor's Alliance, an association of independent inventors based in Silicon Valley, said legislation that would move the U.S. patent system to a first-to-file standard would benefit large companies at the expense of individuals who cannot afford to patent every idea they might have.

"Innovation and the independent inventor are a lot of what America is founded on," Krauss said. "To change to a first-to-file system is to really suppress independent innovation, which is already suppressed enough in my opinion."

Levin's testimony today follows more than four years of research and collaboration he and Myers undertook with lawyers, economists, business leaders and inventors from around the world. The group began its deliberations with several divergent views but gradually reached consensus on several recommendations, Levin said.

"It's very gratifying to have pulled together a diverse group and see the work endorsed," he said.

Levin said his proposals, if passed into law, would not have a direct impact on Yale, but might eventually bode well for the University because it would create more investment opportunities for research companies.

"If the economy is more dynamic, then this is good for us," Levin said.

U.S. Rep. Lamar Smith '69, a Republican from Texas, has already drafted a bill that incorporates most of the NRC's reforms, Levin said.

from my talk at Stevenson on April 22:

In the AIPLA summary of the first "town meeting" on patent reform co-sponsored by the FTC, NAS/STEP, and AIPLA, there was no mention of evidence of a "patent quality" issue, but there was discussion of the use of opposition proceedings to correct the patent quality problem. From the summary:

--As introduced by AIPLA President Bill Rooklidge, the goal of the meetings is to find reforms that will improve the efficiency, promptness and quality of the patent system…. According to Bronwyn Hall, some argue that questionable patents have proliferated because of an overburdened PTO, a lack of experienced examiners in certain technologies, an insufficient prior art database, and an obviousness standard weakened by court decisions. The advantage of an open post-grant opposition system is that competitors are best able to demonstrate obviousness if it exists, Hall maintained. The disadvantage is that it could be too costly, too lengthy, and too burdensome on individual inventors, she added… The emphatic caveat from [Don] Martens, however, is that no such system could go forward without adequate PTO funding for new hires and added administration. He also stressed that the proposed opposition process is not intended as a substitute for litigation, but is only meant to supplement examination to address concerns about patent quality.--

Not everyone believes in the value of post-grant oppositions. In an article "Post-Grant Opposition: Building on Sand," Joseph Hosteny wrote:

--But one recent paper I have read, Prospects for Improving U.S. Patent Quality via Post-Grant Opposition, says that we should have a post-grant opposition proceeding, and offers as a justification the experience of European countries. But none of these countries has a constitution, which provides for resolution of civil disputes by jury trial. Their preference for using government agencies and bureaucrats is not consistent with the our Constitution and Bill of Rights, which restrain the government's role. Thus, the desire to emulate Europe is not sound reasoning.--

Hosteny suggested three difficulties if a post-grant opposition proceeding were adopted:

1. One is trusting an administrative agency with a new task when it is abundantly clear that that same agency is hobbled in completing the tasks it has now. The PTO just hasn't got the funds to do its first job, much less take on a second one.

2. Second what will happen to the lifetime of a patent with this additional procedure?

3. Third, what will happen to litigation? Will it become simpler because a patent has gone through a post-grant proceeding? Not likely.

Hosteny followed up with suggestions for what any (hypothetical) opposition should include.

1. All parties to an opposition should have a stake in the outcome.
2. Any opposer ought to be bound by the outcome with respect to validity and enforceability.
3. There should be no oppositions where the real party in interest is not revealed.
4. The patent owner should have the option to add new claims, and broader claims.
5. The life of the patent should be extended by the time consumed by any opposition.

Hosteny later made clear that we should not have any post-grant proceeding, and further indicated deficiencies in the current AIPLA proposal relative to earlier AIPLA proposals.

1. No amendment enlarging the scope of the claims shall be permitted in the opposition.
2. Statutory grounds in 102 and 112 were limited.
3. Re-examinations would continue to co-exist with post-grant oppositions.

Steven Ludwig noted that the ABA, AIPLA, BIO, FTC, and the NAS have all publicly stated their support of the establishment of a procedure for post-grant review, and Ludwig quoted Congressman Lamar Smith: "it is amazing to me that there is general agreement on the need for post-grant opposition." Of the estoppel issue, Ludwig suggested a compromise which would protect the patent owner from extended harassment due to continuing attacks while still encouraging third parties to request post-grant proceedings would be to limit the estoppel provision to apply only to opposition requests filed more than 9 months after the grant of the patent.

II. What to do about patent quality?

A. Quality according to Deming

Much of the recent patent reform discussion refers to the word "quality." No discussion of quality is complete without reference to the work of W. Edwards Deming. Consider some of the words of his famous 14 points:

Point 3--Cease dependence on inspection to achieve quality. Eliminate the need for inspection of a mass basis by building quality into the product in the first place.
Point 5--Improve constantly and forever the system of production and service, to improve quality and productivity, and thus constantly decrease costs.
Point 6--Institute training on the job.
Point 10--Eliminate slogans, exhortations, and targets for the work force asking for zero defects and new levels of productivity. Such exhortations only create adversarial relationships, as the bulk of the causes of low quality and low productivity belong to the system and thus lie beyond the power of the work force.
Point 11b--Eliminate management by numbers, numerical goals.
Point 13--Institute a vigorous program of education and self-improvement.
Point 14--Put everybody in the company to work to accomplish the transformation. The transformation is everybody's job.

Of government service (such as the PTO), Deming wrote:

--In most governmental services, there is no market to capture. In place of capture of the market, a governmental agency should deliver economically the service prescribed by law or regulation. The aim should be distinction in service. Continual improvement in government service would earn appreciation of the American public and would hold jobs in the service, and help industry create more jobs.--

Specifically, of the mail service, Deming asked a manager about mail sorters:

--How do you pay them? I asked. "Sort 15,000 pieces of mail per day. That is his job." The source of his problem is obvious. With this method of payment, there will never be improvement in sorting of mail, nor will costs of sorting mail decrease.--

Of relevance to the discussion of the patent grant rate discussion, Deming wrote about measures of productivity:

--Unfortunately, however, figures on productivity in the United States do not help to improve productivity in the United States. Measures of productivity are like statistics on accidents: they tell you all about the number of accidents in the home, on the road, and at the work place, but they do not tell you how to reduce the frequency of accidents.

On the other hand, an orderly study of productivity, to inquire whether any given activity is consistent with the aim of the organization, and what it is costing, can be very helpful to management.--

Deming also recognized the need to identify intellectual resources: "Companies take inventory of physical property, but they fail in taking inventory of knowledge."

B. Applying Deming's concepts to patent quality

Deming's concepts teach us to study the manufacturing process itself (here, the examination of patents) at steady-state to develop statistics to identify potential improvements to the system itself. Introducing new levels of inspection (e.g., an opposition process) is not the right step to take. We need information on how to reduce the frequency of "bad patents," not an additional layer to remove bad patents.

The system of examination of patents is more complex than the systems studied by Deming, which typically involved repetition of the same task to make the same product. Here, each incoming patent application purports to have claims directed to inventions which are novel (never before done) and nonobvious (while perhaps structurally similar to some past work, involving some change the motivation for which was not apparent to one of ordinary skill). The examination of each application is a new undertaking, involving the analysis of different facts. Further, with the number of applications increasing each year, we are not at steady state, so that the burden on each examiner is changing in time.

There is a relatively constant system of evaluating patentability, involving tests for utility, novelty, nonobviousness, written description, and enablement. When commentators talk about "bad patents," they are typically talking about ones involving claims which are not novel or obvious. The most direct way to address this is to have examiners well-trained in a given technology area with ready access to databases containing full descriptions of knowledge in that given technology area and with enough time to make well-reasoned comparisons of the claims of the application to the previous work. To the extent there are identified failures (supported by data) with the patent system, steps should be taken to correct the system.


Post a Comment

<< Home