Patents: Improving Quality and Curing Defects

May 10, 2001

Rayburn House Office Building

Testimony of Dr. David Martin, CEO - MCAM Inc.

Mr. Chairman,

I would like to thank you and your distinguished colleagues on the Judiciary Committee for giving me an opportunity to address the topic of patent quality in the United States. While once deemed badges of intellectual honor, patents have become the currency of the U.S. and global economies. Echoing the debates over a currency standard 100 years ago, the United States Congress now faces an issue much more challenging than a decision on gold or silver. The basis of our economy, no longer tangible metals, is now the national product of the human mind - innovation. The protection of this precious asset is the basis for the intellectual property laws of the United States and the developed world. However, in the past decade we have seen the erosion of confidence in patents. While many conveniently dismiss this as inevitable because that which is intangible cannot be measured, I will provide evidence that confidence is evaporating because the assets are indefensible due to egregious inadequacies in patent prosecution and examination.

In January 2001, on a single trading day, one U.S. company and its investors lost over $330,000,000 as a result of misappropriated reliance on U.S. Patents. The patents that Transkaryotic Therapies, Inc. (TKT) owned had significant deficiencies in uncited prior art that were overlooked in prosecution and examination. Relying on uninformed patent advice and confidence afforded by questionable patents, TKT challenged the validity of patents held by Amgen and lost. Our company, using our web-based patent examination system (MCAM DOORS™), published a projected outcome of this case in November 2000, before anyone in the securities industry or in the courts would commit to the probable direction of the decision. Careful examination of the protected intellectual property in this case shows that the TKT deficiency was avoidable had the U.S. Patent Office thoroughly examined prior art.

The whole of the USPTO's operating budget is dwarfed by only four catastrophic days like this. In FY2000, U.S. investors and companies lost billions of dollars as they watched companies representing "proprietary rights" see those rights erode or disappear under scrutiny. This figure does not include actual damage awards, which continue to skyrocket. The losses caused by these avoidable oversights are not offset by gains on the prevailing side. In fact, the inflection for the losers was over twice the magnitude of the gain for the winners - yielding a total loss to the market of several billion dollars. We have yet to see the worst.

Prominent intellectual property law firms and the businesses and investors they advise have relied on a "needle in the haystack" defense with respect to the adequacy of research prior to patent prosecution. Purposely avoiding citing prior and concurrent innovation (including U.S. and international patents) to maximize the likelihood of patent issuance, IP lawyers and patent examiners routinely ignore relevant art that would limit or prevent successful patent issuance. The Federal Courts have held that a patent applicant has no duty to disclose prior art. However, the courts have cautioned that individuals substantially involved in the prosecution of patents may not, "cultivate ignorance" by "disregard[ing] numerous warnings that material information or prior art may exist, merely to avoid actual knowledge of that information or prior art." During recent conversations with lawyers from America's most respected law firms, partners specifically stated to us that they have no interest in knowing all relevant prior art because if that knowledge were available, they would lose the plausible deniability of "not knowing" that the patent they've prosecuted had obvious prior or concurrent art weakness. In written testimony at a USPTO Public Hearing, Gerald Mossinghoff of Oblon, Spivak, McClelland, Maier & Neustadt stated:

Whether or not a search is conducted in any given case is determined by (a) the available budget for filing an application, and (b) the degree to which the applicant is aware of the current state of the art. In many cases, corporations are very active in selective fields. Thus, they are well aware of the state of the prior art in their field, much of which may be their own material. In these situations, there is little or no economic value in conducting a search and applications may be filed without any formal search.

So there we have it. Rather than taking the time to do it right the first time, prosecutors and examiners are deciding that saving $10,000 is worth the risk of the market losing billions. This Committee and America cannot afford that gamble. While the economics at the time of prosecution are deemed by many practitioners as superfluous, the cost to our economy is titanic. As a performance-based organization, the USPTO must be held accountable to resist the temptation to default to issuance without appropriately determining novelty and non-obviousness. Using commercially available software, an examiner could make this determination in less than 1 hour.

Using advanced linguistic analysis, MCAM has conducted over 1000 studies indicating that possibly over 30% of the patents issued in the United States share one or more claims with other patents. If one considers the patent standard of non-obviousness, the percentage worsens remarkably. What we're saying is that approximately 1/3 of U.S. patents not only have no value but may contribute to inadvertent negligent business practices.

More subtle and destructive is the impact that multiple patents issued on the same or similar content at or about the same time. During pendency (the time it takes for the USPTO to actually make an allowance decision), many patents covering identical content are issued - leaving companies, investors, and courts at a total loss as to who truly owns the limited monopoly supposedly afforded by a patent. While often unwilling to conduct prior art searches, attorneys could be required by the Securities and Exchange Commission to draft opinion letters prior to equity offerings that thoroughly disclose the threats to claims of a proprietary nature in a company's offering. If accountability can't exist on the prosecution side, it should at least be deployed to protect the investing public.

In the fall of 2000, the United States Patent and Trademark Office requested use of the MCAM DOORS™ patent analysis system to evaluate its use to detect uncited prior art. In the interest of improving patent quality, we provided access at no charge for a trial period. We have drawn from that log to show the Committee precisely the quality of the patents that the USPTO evaluated using our system. It is helpful to note that, according to one of the senior members of USPTO's Information Search and Retrieval Division charged with evaluating our technology, the system could not detect prior art that the USPTO examiners missed. The USPTO has evaluated over 900 patents using the MCAM DOORS™ system during a period extending from November 2000 until April 2001. In these 900 patents, over 150,000 cases of uncited prior art occur. To put this into perspective, if the patent office expert in charge of the evaluation truly evaluated the relevance of the prior art found using our system, it would have taken 6 trained examiners 24 years simply to read the documents. So, did the system really not work or did we so dramatically highlight the profound ineptitude of the current systems currently paralyzing the hard-working examiners that the Search and Retrieval group decided to bury the evidence? Their use log is reproduced as an Appendix if you'd like to answer that question yourself.

Oddly enough, at about the same time that this PTO employee was defending the status quo, the Patent Office Professional Association (POPA) published a report stating: "examiners often lack the time and resources to find all the relevant prior art relevant to a patent application." The report goes on,: "The hurdles for examiners to find prior art are often impossible to surmount."

The following example is taken directly from the USPTO MCAM DOORS™ utilization log - an audit function that we built into our system to ensure accountability.

United States Patent 5,487,682 (hereinafter '682) is a patent for a "Shielded Data Connector." When prosecuted and examined, this patent referred to nine patents as prior art and cited no other trade references that could threaten a claim to novelty. Filed two years prior to '682 was U.S. Patent 5,593,311 (hereinafter '311), which asserts significantly equivalent claims. While subsequent innovations have found '311 relevant, no examination of the uniqueness between '311 and '682 was done at USPTO. More problematic is the uncited prior art embodied in U.S. Patent 4,602,833 (hereinafter '833), which was never cited by '682 but was cited by '311. In fact, over 300 patents exist as uncited prior art to '682 that have been deemed relevant by USPTO examiners in the prosecution of concurrent and subsequent innovations. Our analysts took less than 90 seconds to find three overlapping patents (excerpted below) that threaten the enforcement of '682 - less time than it takes the USPTO to make a classification decision on a patent application. On average, most U.S. patents cite less than 10% of the relevant prior art. The term "relevant prior art" in MCAM nomenclature is simply defined as identical or near identical claims that have already been examined on one or more occasions by USPTO examiners and been found to be relevant. Our standard involves an integration of semantic and examination associations - a standard consistent with the letter of the patent law.

5,487,682

22. A shielded electrical connector comprising:

an insulative outer shell carrying latching structure thereupon for connection to a mating component;

an insulative housing member receivable within the outer shell, said insulative housing including a terminal support platform having an open upper face with contact channels spaced laterally therealong and having a rib located between adjacent channels;

a plurality of electrical contacts laterally positioned along said platform where each contact is received within one of said channels;

where one of said ribs between two of the plurality of contacts includes a longitudinally extending slot therein; and

a cross talk shield between said two contacts and that is positioned there by said slot in said rib.

5,593,311

9. An hermaphroditic electrical connector comprising:

a connector housing configured for mating engagement with a like housing; and

at least one electrical contact supported in said housing and configured for mating electrical engagement with a like contact;

said electrical contact having a mating end and an opposed termination end;

said mating end of said contact having an elongate deflectable contact beam including a central apex, an inclined front facing mating surface on one side of said apex and an inclined rear facing engagement surface on the other side of said beam;

wherein upon said hermaphroditic interconnection of said electrical contact with said like contact said front facing mating surfaces make initial engagement, said apices pass over one another deflecting said contact beams and said rear facing engagement surfaces contact one another locking said contact to said like contact.

4,602,833

1. A hermaphroditic electrical connector of the type comprising an insulating housing having a plurality of conductive terminals and electrical shunt means therein, said terminals having resilient contact tongues which engage like tongues in a like connector, said shunt means being in shunted relation with said terminals when said connector is in an unmated condition, said shunt means being in unshunted relation with said terminals when said connector is in mating engagement with a complementary connector, characterized in that said shunt means are fixed to a dielectric carrier which moves relative to said housing in response to mating with a complementary hermaphroditic connector, said shunt means engaging portions of said terminals remote from said contact tongues, whereby said shunt means moves relative to said housing and out of shunted relation with said terminal portions.

It is troubling to continue hearing USPTO's insistence that it doesn't have the resources to deploy tools for examiners to do a better job. It is equally troubling to hear that, if fully funded, the USPTO would add more bodies rather than overhauling a broken infrastructure. Please understand, I fully support the allocation of sufficient funding for USPTO. However, I strongly object to pouring more resources into leaking buckets. Funding must be linked to quality control. Using a web browser, MCAM DOORS™ can clearly rule out patentability for most inventions in fewer than 2 minutes. The average pendency at USPTO currently is almost 2 years. Using information technology tools to triage examination decisions would immediately address the infrastructure demands on the USPTO and would improve the quality of examinations done by them.

Finally, former Commissioner Q. Todd Dickinson, before this very Committee, made the statement last year that the quality of examination couldn't be that major a problem as the number of petitions for re-examination were not significant. What he failed to articulate is that there are significant procedural and financial impediments to filing such petitions - making it prohibitive for companies of individuals to provide this critical check-and-balance function on a runaway quality deficiency at PTO. This Committee's members can immediately act to make re-examination a democratic process so that quality can be assured.

Conclusions and Specific Recommendations

The world isn't coming to an end. Our simple call is for accountability and responsibility echoing voices of ages past.

"The humblest citizen in all the land, when clad in the armor of a righteous cause, is stronger than all the hosts of error." William Jennings Bryan - July 9, 1896

Add to the Committee's righteous cause the data provided herein and I am confident that the hosts of error and omission will be vanquished. We can effect immediate change in the following ways.

1. While a duty to search cannot be imposed without modification of the law (which we would recommend), 37 CFR §1.104(a) obligates examiners to conduct "a thorough investigation of the available prior art relating to the subject matter of the claimed invention." This Committee and Congress should enforce the standard and advise the USPTO that examination time must not be restricted by capricious time limits if an examiner has not satisfactorily searched not only the "field in which the invention is classified, but also analogous arts." This can be done with commercially-available software systems and should be immediately deployed.

2. Given the legitimate claim by patent examiners of the inaccessibility of significant prior art repositories, the public should be allowed to provide examiners with information prior to the final publication of letters of patent. Once issued, the patent could then be afforded greater limitation to challenge as the public comment period would likely improve the quality of issued patents.

3. Realizing the importance of patents for domestic and international protection of United States business interests, this Committee should authorize an audit of issued U.S. patents and foreign patents registered in the U.S. and identify those patents that are redundant. In so doing, this Committee could significantly curtail the frivolous legal proceedings debating the merits of unenforceable claims allowing the courts to focus on meaningful challenges and infringement issues.

Appendices:

USPTO Utilization Log

Patently Obvious™ Reports