Despite its innocuous title of “Improvement in Telegraphy,” US Patent 174,465 has often been described as “the single most valuable patent in history.”1 It was also the focal point of one of the most sordid dramas in patent history. Despite determinations by the US Supreme Court in the Telephone Cases of 1888 that Alexander Graham Bell deserved credit for invention of the telephone,2 allegations of conspiratorial fraud, bribery, and intellectual theft surrounding the granting of his patent still cloud that credit today.

There is no question that on Valentine’s Day in 1876 Bell and Elisha Gray, both pictured in figure 1, filed documents with the US Patent Office. Bell filed a full patent application while Gray filed a “patent caveat,” similar to a provisional patent application under modern law. Both were assigned to the same examiner, Zenas Fisk Wilber. A known alcoholic, Wilber asserted years later that in exchange for a bribe of $100, he improperly revealed the content of Gray’s caveat to Bell’s attorneys and allowed material drawn from it to be added to Bell’s disclosure.3In any event, Bell prevailed.

Figure 1. Although Alexander Graham Bell (left) and Elisha Gray (right) filed documents with the US Patent Office on the same day, Bell ultimately prevailed in solidifying credit for invention of the telephone under the “first-to-invent” system used by the US. In recent years the costs associated with that system have been significant; they are one motivation for the decision to change to a “first-to-file” system. Current legal costs for an interference—the internal patent office procedure used to identify the first inventor—are typically estimated at about $500 000. As much as the first-to-invent system of the US has been touted as beneficial for small inventors, cost has ensured that interferences usually take place between large entities and that they are rare. Currently almost 500 000 new patent applications are filed in the US each year, but only about 100 interferences are initiated.

Figure 1. Although Alexander Graham Bell (left) and Elisha Gray (right) filed documents with the US Patent Office on the same day, Bell ultimately prevailed in solidifying credit for invention of the telephone under the “first-to-invent” system used by the US. In recent years the costs associated with that system have been significant; they are one motivation for the decision to change to a “first-to-file” system. Current legal costs for an interference—the internal patent office procedure used to identify the first inventor—are typically estimated at about $500 000. As much as the first-to-invent system of the US has been touted as beneficial for small inventors, cost has ensured that interferences usually take place between large entities and that they are rare. Currently almost 500 000 new patent applications are filed in the US each year, but only about 100 interferences are initiated.

Close modal

An apocryphal anecdote has Bell’s attorney beating Gray’s to the patent office by a mere hour or two, but whether he did was ultimately irrelevant: Since its early history, the US has operated under a “first-to-invent” patent system in which the question of who wins the race to the patent office is less important than who invented something first (see the article by James Richardson and Craig Wood in PHYSICS TODAY, April 1997, page 32). That system will change on 16 March 2013.

The change implemented by the Leahy–Smith America Invents Act that has attracted the most attention is often informally described as a move from the first-to-invent system to a “first-to-file” system. Characterized that way, the concept is deceptively simple: The first person to file a patent application for an invention is the one entitled to the patent. The reality, though, is considerably more complex.

First, the act does not implement a pure first-to-file system, but rather implements a variation referred to as a “first-inventor-to-file” system. The distinction is important, particularly for scientists, because a major difference from a pure first-to-file system is that it permits, and even encourages, publication of the invention before it is filed with the patent office—provided the applicant adheres to certain critical provisos.

Second, even pure first-to-file systems must account for inventors who not only come from widely varied technology areas but also develop their inventions under a varied array of circumstances. They develop and implement concepts that may be similar rather than identical, and, like Isaac Newton standing on the shoulders of giants, their inspiration routinely relies on an existing body of knowledge. They work in environments ranging from R&D departments of large multinational corporations to university laboratories to the perhaps romanticized garage of the individual inventor. Their research is financially supported by both private and public interests, which often reflect divergent views on such things as the role of profit and competition. And it is increasingly rare that the value of inventions is seen through a parochial lens; inventors are now routinely confronted with the complex reality of global commercialization of their ideas.

Perhaps most notable of all, the Leahy–Smith Act does much more than change the US to a first-to-file system; it is the latest step in an effort that began some 130 years ago to harmonize the patent laws of different nations. The most comprehensive revision of US patent law in more than 50 years, the act was introduced against the backdrop of an acute need to address inefficiencies in the system. Some of its more important other provisions include the introduction of new defenses against patent infringement, new ways for third parties to oppose the granting of patents, and procedural mechanisms evidently designed to curtail the activities of entities sometimes pejoratively called patent trolls. That term usually describes an entity that seeks to exploit patents for licensing revenue without making its own products and was adopted as a substitute for “patent extortionist” after a 2001 libel suit against Intel.4 Undoubtedly, the various provisions reflect the perspectives and objectives of different classes of inventors—including scientists—but it is overly idealistic to suppose that those provisions flawlessly balance the various tensions among those perspectives.

For hundreds of years, patent systems have encouraged inventors to disclose their inventions to the public through the incentive of a time-limited, government-backed monopoly to control use of their inventions by others. The bargain was well expressed by Galileo Galilei in December 1593 when he petitioned the doge of Venice for a patent on a machine “for raising water and irrigating land with small expense and great convenience. . . . It is not . . . fit that this invention, which is my own, discovered by me with great labor and much expense, be made the common property of everybody.” If he were granted a temporary monopoly over its use, he said, “I shall the more attentively apply myself to new inventions for universal benefit.”5 

The Leahy–Smith Act fundamentally recalibrates the interaction of patent laws with the timing and form of publications. At its most basic level, the move to any form of first-to-file system encourages inventors to file patent applications at earlier stages of an invention’s development. Applications that are more conceptual and more incremental are likely to be filed, as inventors—particularly those working in competitive areas—seek the benefit of obtaining the first filing date. Ironically, physicists who have been repeatedly discouraged by research journals from engaging in serial publication of incremental results will find considerably more acceptance and advantage to serial patenting.

Interestingly, a shift in focus to filing applications on technologies earlier in development is likely to work better for physicists than for some other scientists. The US Patent and Trademark Office has not routinely required a working model of an invention in many years, and physics is generally considered to be a predictable art because the developed laws of mechanics, electromagnetics, acoustics, and so forth enable reliable extrapolation of results. The requirements for demonstrating functionality of an invention are accordingly far more easily satisfied—and are therefore more amenable to earlier filings—than in the unpredictable arts that mostly encompass chemistry and biology.

The change to a first-inventor-to-file system may also have consequences on the approach physicists take to publication of their results because of the most significant deviation from a pure first-to-file system provided by the Leahy–Smith Act: the retention in US law of a form of grace period for nonpatent disclosures (such as journal or other publications) by the inventor. An invention is patentable when it is both novel and not obvious over a body of knowledge referred to as prior art, as outlined in figure 2. Under a first-to-invent system, the definition of prior art is complex because it must account for the often difficult-to-determine date when something was invented. Under a pure first-to-file system like that used in most of the world, the definition of prior art is simple: It is everything that was publicly known before the filing date of an application. But in choosing to retain a grace period for disclosures, the US has introduced a critical wrinkle. Under its first-inventor-to-file system, prior art is everything publicly known before the filing date of an application, except for disclosures made by or derived from the inventor within the preceding year.

Figure 2. The Leahy–Smith Act implements a “first-inventor-to-file” patent system, which is a modified form of a “first-to-file” system. Disclosures made within a grace period of one year prior to filing a patent application are considered in determining who is entitled to the patent. To be granted a patent on an invention, an inventor must file an application with the US Patent and Trademark Office and must be either the first to have so filed or the first to have disclosed the invention during the grace period. Whether a reference is “prior art” defines whether it may be used in evaluating an invention’s novelty and nonobviousness.

Figure 2. The Leahy–Smith Act implements a “first-inventor-to-file” patent system, which is a modified form of a “first-to-file” system. Disclosures made within a grace period of one year prior to filing a patent application are considered in determining who is entitled to the patent. To be granted a patent on an invention, an inventor must file an application with the US Patent and Trademark Office and must be either the first to have so filed or the first to have disclosed the invention during the grace period. Whether a reference is “prior art” defines whether it may be used in evaluating an invention’s novelty and nonobviousness.

Close modal

It is important to recognize that the body of knowledge that forms the prior art is not limited by the language in which it may be known nor by the fact that portions of it may be obscure. It does not even matter whether any portion of a particular reference was ever read, as long as it was publicly accessible. One recent example that strikingly illustrates the point is the patent dispute involving Blackberry services. With hundreds of millions of dollars at stake, a little-known publication discovered in the Norwegian University of Science and Technology Library in Trondheim, Norway, proved to be critically important. Other examples of material considered as prior art are eprints, doctoral theses deposited and indexed in libraries, and poster papers displayed at scientific conferences.

The retention of a grace period is of particular importance to scientists because they frequently place a stronger emphasis on journal publication than do nonscientist inventors. At least in the US, prior publication will not preclude the ability to obtain a patent, provided the patent application follows publication by no more than a year. In that respect, too, physicists may find the system more conducive to their practices than will other scientists. Most journals of interest to physicists do not impose publication embargoes of the sort that are common among journals in other fields, notably medical journals. Publication of an invention as an eprint, as part of a doctoral thesis, or even on a private webpage will not immediately jeopardize the ability to obtain a US patent.

More widely, the act of publishing an invention may even be advantageous. As long as a patent application is filed and the time constraints are satisfied, the first-inventor-to-file system grants the patent to the first person who discloses the invention to the public through either a publication or the patent filing. For inventors who are unconcerned about patent rights outside the US, early publication of an invention is sound strategy. It is the most effective way of preserving the right to the patent in the US, is less costly than filing a patent application, and provides a year to fully consider whether a patent is even desirable.

Such a publication strategy is limited to the US and the small number of other countries that afford a grace period. While early forms of patent-reform bills conditioned the US’s adoption of a first-to-file system on acceptance of grace periods by Europe and Japan, such conditioning is absent from the enacted version and requires a modification of the strategy when foreign patent rights are of concern. For example, a provisional patent application may be filed just before publication. Provisional patent applications are cheaper, easier, and quicker to file; and although they will not themselves ever be examined or mature into a patent, they provide a one-year placeholder that preserves the right to file foreign applications even with an intervening publication.

While the grace period promotes the early disclosure of inventions, other provisions of the Leahy–Smith Act do the opposite. One eternal question inventors face is whether to seek a patent on their invention or to maintain it as a trade secret. Secrecy can be attractive, especially in those instances where disclosure of an invention’s details makes it relatively easy to design alternatives. The risk of maintaining an invention as a trade secret has been that someone else might independently develop and patent it, which would suddenly make its continued use an infringement. The Leahy–Smith Act fundamentally changes the calculus in making that determination.

Expanding a provision that was introduced a decade ago to address the perceived special case of business-method patents, the Leahy–Smith Act immediately affords a prior-user defense to patent infringement in all technology areas. Provided that the invention was in generally continuous use for at least a year prior to the filing date of a patent that covers it, such use may continue. While the use needs to have been commercial, that term is defined broadly: It covers internal uses by companies, sales between independent agents, premarketing regulatory review, and—rather counterintuitively—use by nonprofit laboratories.

The consequences of that defense are significant. One principal motivation that encouraged filing a patent rather than maintaining an invention as a trade secret has been removed. The risks of maintaining it in secrecy no longer include the possibility that its use can later be prevented through independent invention by someone else. Other considerations for the decision accordingly take on more immediate importance. How easy is it for a competitor to discover the invention? How likely is independent invention of the same thing? How important is publicity to one’s strategic goals? Because the change in law makes secrecy a more valuable protection strategy, the expectation is that the number of inventions that are kept secret will increase.

Although the prior-user defense is intentionally broad, institutions of higher learning are immune to it: An exception prohibits the defense from being applied against inventions developed by university researchers with federal funding. The exemption is specifically structured with the academic-publication concerns of university researchers in mind and provides an incentive for them to continue their traditional approach to publication. Patents originally filed by universities are now stronger than patents owned by others because they may be applied against anyone, even someone who may previously have used the invention in secret.

The favoring of university researchers who use government funds is consistent with a general effort that has been under way for three decades to increase the involvement of universities in the patent system. When former commissioner of patents and trademarks C. Marshall Dann wrote for PHYSICS TODAY in March 1978 (page 23), he adopted an almost cajoling tone in encouraging physicists to see the patent system as an alternative to journal publishing. Although he recognized that the way of thinking about patents on federally funded inventions was evolving, it was still two years before greater control over such patents would be extended to universities and other nonprofits through the Bayh–Dole Act. Considerable changes have occurred since he observed that “some scientists, including physicists, think that there is something crassly materialistic about being interested in patents, and that patents are usually given for clever trivialities.”

Before the Bayh–Dole Act, the government took the view that allowing an inventor to retain ownership of a patent on an invention developed with federal funds was contrary to public policy, and it required the public to pay twice: first to support the initial research and second in higher costs for technology subject to a private-ownership interest. There is a logic to such a view, but the demonstrable fact is that without the incentive of private ownership, little publicly supported research was effectively commercialized. Allowing federally supported inventors to retain patent ownership has resulted in markedly increased commercialization of some technologies to the direct benefit of the public.6 Technology-transfer offices of universities have consequently become much more aware of the financial value of patenting, to the benefit of both the universities and university researchers. The Leahy–Smith Act continues that trend beyond the university by increasing the share of patent-licensing royalties that may be retained by government contractors who are nonprofit operators of government-owned facilities.

In addition to the first-inventor-to-file system with a grace period and the broadening of the prior-user defense, the Leahy–Smith Act includes other provisions intended to address specific recent criticisms of the patent system. For example, the large backlog of unexamined applications, plotted in figure 3, has increasingly been a source of frustration for inventors and has prompted concerns that delays are straining the system’s ability to spur technological innovation. Much of the backlog is attributable to a congressional practice of taking patent fees paid by inventors and diverting them to other government programs. The Leahy–Smith Act increases fee levels and gives the patent office greater financial control over their use, so that the office can examine applications more effectively and expeditiously.

Figure 3. The backlog of patent applications awaiting examination by the patent office has risen sharply over the past 15 years. The average interval is now about 36 months between the time of filing and the time a patent is granted. The patent office hopes the Leahy–Smith Act, which gives the office greater control over its own finances, will aid in addressing the backlog.

Figure 3. The backlog of patent applications awaiting examination by the patent office has risen sharply over the past 15 years. The average interval is now about 36 months between the time of filing and the time a patent is granted. The patent office hopes the Leahy–Smith Act, which gives the office greater control over its own finances, will aid in addressing the backlog.

Close modal

Some critics go further and assert that the patent system not only fails to spur innovation but actually stifles it. Litigation by patent trolls is frequently identified as an exploitation of the system that abuses the limited monopoly granted to inventors in exchange for their disclosure. The Leahy–Smith Act sharply curtails one of the primary strategies used in that type of litigation, namely the practice of joining many defendants—there exist instances of more than 100 defendants spread across the US and otherwise unrelated—in a single infringement action. The expectation is that removal of that strategy will allow legitimate infringement actions to proceed while greatly reducing the ability to engage in exploitive behavior.

Others contend that some patents are obvious and trivial; they question the quality of examination. In addition to improving funding for the patent office to address examination quality, the act greatly expands the ability of third parties to submit relevant prior art for consideration by the examiner and to initiate proceedings to oppose the granting of specific patents. Generally modeled after similar opposition proceedings in other countries, that additional level of review may dovetail with investigations into the use of peer review in the evaluation of patent applications (see the box ).

The Leahy–Smith Act is not merely a tweaking of the US patent system. It alters many of the most basic precepts of the system and will dramatically affect many basic patenting strategies developed over the past decades as the intricacies of the former system were refined and understood. Much remains unclear about the full scope of the changes that will occur as a result of the system’s overhaul. Many terms in the legislation are ambiguous and will ultimately require judicial decisions to resolve.

The only US president to have been granted a patent is Abraham Lincoln. The idea for his patent came from a mishap in which Lincoln found his boat stranded on a sandbar, and the boat’s hands used planks, crates, and barrels to buoy it enough to free it. His invention, illustrated in figure 4, was never actually used with any vessel, but it is easy to imagine his excitement at the process and at the promise of being granted a patent. His poetic expression of that excitement, that the “patent system . . . added the fuel of interest to the fire of genius,”7 is one that still captures the system’s ultimate goal.

Figure 4. Abraham Lincoln, inspired by a mishap in which he found himself stranded on a sandbar, became the only US president to be granted a patent. Illustrated by the two figures shown here, the text of US Patent 6,469 begins: “Be it known that I, Abraham Lincoln, of Springfield, in the County of Sangamon, in the State of Illinois, have invented a new and improved manner of combining adjustable buoyant air chambers with a steamboat or other vessel for the purpose of enabling their draught of water to be readily lessened to enable them to pass over bars, or through shallow water, without discharging their cargoes.”

Figure 4. Abraham Lincoln, inspired by a mishap in which he found himself stranded on a sandbar, became the only US president to be granted a patent. Illustrated by the two figures shown here, the text of US Patent 6,469 begins: “Be it known that I, Abraham Lincoln, of Springfield, in the County of Sangamon, in the State of Illinois, have invented a new and improved manner of combining adjustable buoyant air chambers with a steamboat or other vessel for the purpose of enabling their draught of water to be readily lessened to enable them to pass over bars, or through shallow water, without discharging their cargoes.”

Close modal
Patents and peer review

People who manage peer-review systems love to quote Winston Churchill: “No one pretends that [it] is perfect or all-wise. Indeed, it has been said that [it] is the worst form . . . except for all those other forms that have been tried from time to time.” Churchill was, of course, applying his characteristic wit in describing democracy as a form of government. But the analogy with peer review resonates strongly with the editors of scholarly journals, who recognize the central role of peer review and yet are all too aware of its weaknesses.8 (See also the letter by Steven Rothman in PHYSICS TODAY, September 1995, page 124.)

Still, peer review has not traditionally been used for evaluating patent applications, even though they can be as technical as journal articles and patent examiners have at least as many limitations on the breadth of their personal expertise as journal editors. One justification for the difference in approach has been a concern that technical experts consulted as referees are insufficiently knowledgeable of the legal considerations that govern decisions for patent applications. The justification is a weak one, and nations around the world, including the US, are beginning to experiment with peer-review systems for patents. The Peer to Patent pilot program in the US, which concluded at the end of 2011, is limited to small, defined technology areas and differs in important respects from tightly controlled, anonymous journal peer review. It more closely resembles online social forums. Any member of the public can participate as a referee and can provide comments through a Web interface that has no specific structure for participants to maintain anonymity, although they are permitted to use pseudonyms. The information gained from such pilot programs will ideally be informed by the extensive experiences of journals and funding agencies in determining whether and how to implement more widespread peer review of patent applications.

The anonymous peer-review procedure used by journals has sometimes also been the target of patent disputes. A 1988 case involving Physical Review Letters is instructive.9 A paper the journal rejected became a point of focus when one party to patent litigation sought to compel disclosure of the identity of one of the referees; the party argued that transmittal of an otherwise unpublished paper to a referee might result in circulation sufficient to qualify as disclosure that would bar the granting of a patent. The court, refusing to compel the journal to identify the referee, noted that “even if the referee consulted a few colleagues about the merits of the manuscript (as the [journal] procedures . . . permit), this would not, as a matter of law, suffice to constitute a ‘printed publication.’ ” Two points are noteworthy. First, submission of a paper to a journal, even when referees are consulted, is generally insufficient to count as a disclosure that would affect patent rights, at least until the journal formally publishes the paper. Second, journals’ confidentiality policies regarding the identity of referees will generally be respected and enforced by the courts.

1.
See, for example,
J. W.
Wright
 et al, eds.,
The New York Times Guide to Essential Knowledge: A Desk Reference for the Curious Mind
, 3rd ed.,
St. Martin’s Press
,
New York
(
2011
), p.
1233
.
2.
Telephone Cases, 126 US 1 (1888).
3.
For a full account of the Bell–Gray controversy, see
A. E.
Evenson
,
The Telephone Patent Conspiracy of 1876: The Elisha Gray–Alexander Bell Controversy and Its Many Players
,
McFarland & Co
,
Jefferson, NC
(
2000
).
4.
See, for example,
P.
Groves
,
A Dictionary of Intellectual Property Law
,
Edward Elgar
,
Northampton, MA
(
2011
), p.
314
.
An exhaustive discussion of the etymology of “patent troll” can be found in
R. P.
Niro
,
J. Marshall Rev. Intell. Prop. L.
6
,
185
(
2007
).
5.
See
B. W.
Bugbee
,
Genesis of American Patent and Copyright Law
,
Public Affairs Press
,
Washington, DC
(
1967
), p.
24
.
6.
See
D. C.
Mowery
 et al,
Ivory Tower and Industrial Innovation: University–Industry Technology Transfer Before and After the Bayh–Dole Act
Stanford Business Books
,
Stanford, CA
(
2004
).
7.
A.
Lincoln
,
Speeches and Presidential Addresses: 1859–1865
The Current Literature
,
New York
(
1907
), p.
13
.
8.
The origin of the analogy is difficult to identify. Some early examples include
E. D.
Robin
,
C. M.
Burke
,
Chest
91
,
252
(
1987
) ;
9.
Solarex Corp v. Arco Solar Inc, 121 FRD 163 (EDNY 1988).

Patrick Boucher is currently a patent attorney at the firm of Marsh Fischmann and Breyfogle, and was previously an associate editor of Physical Review B. He regularly comments on issues involving law and science at JuriScientia, http://juriscientia.com.