> home
> About
> Contact Us
> Editorial Info
> IEEE-USA

04.10

What Should an Engineer Know About First to File?

By Daniel Fisher

I was recently asked by a colleague to sit in on a meeting with staff of his U.S. Senator on Capitol Hill. The topic was Senate bill S.515, a patent reform bill. This bill has been years in the making. It was recently reported out of the Judiciary Committee with a “Manager’s Amendment” and may be taken up by the full Senate when it can be scheduled for “floor time.”

Most of the discussion centered around one provision of the bill, the transition from “first-to-invent” to “first-to-file.” The United States has had a first-to-invent patent system since the first patent statute. However, the rest of the world has adopted a first-to-file patent system. Now, S.515 is poised to change the United States to a first-to-file patent system.

At this meeting, there was much discussion about what this will mean to inventors, small and large U.S. companies, and to the U.S. economy. This topic has been discussed at length in the press, particularly in journals related to patent law, for many years. As the reality of such a transition seems to become more real, I began to think about what it would mean to me as a practicing patent attorney and former practicing engineer.

Under the current first-to-invent system, most research departments require their researchers to keep notes in a lab notebook, and many of these require the pages of the notebook be signed and dated by two witnesses. The notebook is retained and owned by the company. When inventors from two different companies file patent applications for substantially the same invention, the Patent Office declares an interference proceeding aimed at identifying which inventor was the first inventor. Then, each company culls through their own file cabinets to find the notebook that has the first drawing of the invention, the first written description of the invention and other information related to the reduction to practice of the invention. Copies of these documents must be filed in the Patent Office shortly after the interference proceeding is declared. Luckily, the interested companies have the information in their possession.

If Congress were to change to a first-to-file patent system, things would work differently. The Patent Office would no longer declare an interference proceeding because identifying the first inventor would be unnecessary. Companies would likely place new emphasis on rushing to place invention descriptions on file at the Patent Office. Fortunately, this can be easily and inexpensively done with a provisional patent application.

When inventors from two different companies file patent applications for substantially the same invention, a question may arise as to whether the separate applications were based on independent invention or perhaps one invention was derived from the other. With independent invention, the first inventor to file would be awarded the patent in a first-to-file patent system. If invention in one application was derived from the other, then the copyist would be denied the patent and the original inventor awarded the patent, even if he/she were second to file.

As an example, consider a couple of researchers in the same field but from different companies attending the same conferences and presenting their research. It is reasonably likely that some, maybe not a lot, of the patent applications filed in the Patent Office will claim inventions that were derived from someone else’s research, possibly learned at a technical conference.

Technical conferences are important marketing opportunities for many companies. However, it is equally important that companies protect the fruits of their own research. In a first-to-file patent system, companies should mandate that all papers being presented at a public conference be placed on file as a provisional application before the conference. Even so, a company may worry about question and answer sessions or even casual conversations at a reception. Some hints of future or past research activity not covered by the provisional patent application may still leak out.

Under a first-to-file system, the original inventor, who happens to be second-to-file, will have a right to argue that the inventor who filed first derived the invention from the original inventor’s research paper presented at the XYZ conference. Even a generous grace period will not save the original inventor if a copyist returns from the conference and quickly files a patent application for an invention derived from the original inventor’s conference paper. The grace period applies only to qualification of prior art, not the date of filing. Of course, this only applies under a first-to-file system. Some companies may even refuse to allow their employees to present papers at conferences until all patent applications related to the invention are on file at the Patent Office.

After the original inventor foolishly presents a paper unprotected by a provisional application or speaks too liberally during a cocktail hour, he or she might become frustrated and angry, but where is the proof that the inventor who filed first actually copied the idea? The original inventor can easily prove the date of his/her original conference paper. However to even suggest derivation, the original inventor must also show that the copyist had no description of the invention before the conference and a blizzard of inventive activity immediately following the conference. Such records exist only within the company that employs the copyist. The original inventor does not have access to such records.

Even if both companies are ethical and preserve these records, they are under no obligation to give the records to the other company. In fact, they are financially motivated to withhold access to the records. Companies are bound by fiduciary duties to their shareholders to keep such records as company secrets.

To obtain robust discovery of the records of the other company, it is best to be in a U.S. District Court. The Patent Office has only weak methods to request a company to divulge its company records. Methods involving mild coercion such as invalidating the claims of the involved patent application are used today in interference proceedings. Such methods might be used in a derivation proceeding. However, it takes a U.S. District Court judge to hold the company in contempt and confine the CEO in jail until the company complies with a discovery order.

Under first-to-invent, the evidence of invention is in the inventor’s own hands. Under first-to-file, the evidence of derivation is in the other guy’s hands.

Companies may enforce a policy of “never say anything” until the patent application is on file. That way, the invention will not “leak out” at a conference. Cautious counsel may even advise companies to never say anything until the patent grants and the research area has been fully exploited. After all, a competitor could jump on a promising new research area and obtain its own blocking patents. However, the “never say anything” policy is easier for large companies than small companies.

Large companies, such as Intel Corporation or Cisco Systems, generally have internal finance committees that strategize about issuing new bonds or new stock to raise capital. An inventor and his/her department head might go to the finance committee to request raising $10 million to bring the next new patented product to market and see it launched for the first 3 years. When the finance committee goes to the public bond market with a new bond issue, it seldom has to divulge the patent behind a new product line. Large companies have multiple existing product lines with diverse existing distribution channels sufficient to underwrite the new bond issue without even divulging the new product.

With small companies, financing is very much different. The invention underpins a new product line that is often the only product under development or at least one of very few products in development. There is no internal finance committee with which to meet. Instead, the inventor must go outside of his/her company to a committee of venture capitalists who ask a lot of questions and who hire outside consultants to review the proposed new product. Knowledge of the invention leaves the company.

A non-disclosure agreement may prevent knowledge of the invention from being cited as prior art against an application of the original inventor, but it does not prevent the venture capitalist’s consultant from copying the invention and filing a patent application. It is wise for the original inventor to at least file a provisional patent application before talking to a venture capitalist, but here too, the capitalists may ask a lot of questions. If the consultant to the capitalist files a patent application, the consultant may win a first-to-file contest if the inventor did not file a provisional application or the consultant may file on a blocking patent application if the inventor answers too many questions. The small company would be forced to prove that the consultant’s patent application was derived from the venture capitalists who received their knowledge of the invention from the original inventor at that meeting. Since the Patent Office has only limited powers of discovery, the patent would likely issue to the copyist. The small company that needed the venture capital financing in the first place may be either driven out of business or forced to find other financing to sue the copyist in U.S. District Court, discover evidence of derivation and invalidate the copyist’s patent.

Little by little, large companies would become larger and small companies would become fewer in number.

Today is the wrong time in the business cycle to change patent law to a first-to-file system and give large companies greater advantages over small companies. With the banks in economic crisis, bank financing for small businesses has substantially disappeared. CIT Group, Inc. was the major bank lender to small businesses, but it nearly collapsed until it emerged from bankruptcy in December of 2009. Small businesses have been forced to borrow from venture capitalists who either require higher interest rates or take ownership outright.

If a first-to-file system diminishes patent protection for small businesses, the number of small businesses that survive in today’s economy will diminish and start-up businesses will be doing less hiring.

According to the Small Business Administration [1], “Firms with fewer than 500 employees accounted for 64 percent (or 14.5 million) of the 22.5 million net new jobs (gains minus losses) between 1993 and the third quarter of 2008. Continuing firms accounted for 68 percent of net new jobs, and the other 32 percent reflect net new jobs from firm births minus those lost in firm closures (1993 to 2007).”

Former Commissioner for Patents John J. Doll said, “Small businesses account for 99.7 percent of all employer firms in America and those firms create two out of every three new jobs and account for nearly half of America's overall employment. They have played a vital role in helping our economy add millions of new jobs and have helped reduce unemployment rates across the country.” [2]

In 2003, the U.S. Small Business Administration’s Office of Advocacy reported that “small patenting firms produce 13 to14 times more patents per employee than large patenting firms.” An Advocacy study in 2004 reported on the growing technological influence of small firms, and in particular the rise in the number of highly productive inventors at smaller firms. Meanwhile, the share at large firms fell.

Since small businesses account for 99.7 percent of all employer firms in America, it follows that patent law changes that reduce the number of small businesses would also decrease jobs in America.

IEEE-USA encourages its members to make their views on important Federal legislation known to their respective Senators and Representatives. This applies to issues raised by S.515. Contact information for individual Senators is available from the Senate website at http://www.senate.gov.

References:

[1] Small Business Administration (SBA), Office of Advocacy, Frequently Asked Questions, updated September 2009, data sourced from: U.S. Dept. of Labor, Bureau of Labor Statistics, Business Employment Dynamics. Note that the methodology used for the figures above counts job gains or losses in the actual class size where they occurred.

[2] John J. Doll, Commissioner For Patents, U.S. Patent and Trademark Office, March 29, 2007 testimony before the Committee on Small Business, U.S. House of Representatives.

The opinions expressed in this piece are those of the author and do not necessary reflect an official position of IEEE, IEEE-USA or the IEEE-USA Intellectual Property Committee.

Daniel Fisher is a member and past chair of IEEE-USA's Intellectual Property Committee. He is a practicing patent attorney in Fairfax, Virginia where he prepares and prosecutes patent applications for clients in the electrical arts. A member of IEEE since the late 1960s, he is a graduate of Johns Hopkins University School of Engineering and worked as an engineer in the defense industry for 25 years. He received an MBA from the University of Maryland in 1973 and a JD from George Washington University in 1979 and has worked as a patent lawyer for 20 years. He is a registered patent attorney and a licensed Professional Engineer.

Comments may be submitted to todaysengineer@ieee.org.