I now believe it’s possible that the current rules governing business method and software patents could end up harming all of us—including Amazon.com and its many shareholders.… —Jeff Bezos, CEO of Amazon.com (2000) [I]t is a fact, as far as I am informed, that England was, until we copied her, the only country on earth which ever, by a general law, gave a legal right to the exclusive use of an idea. In some other countries it is sometimes done, in a great case, and by a special and personal act, but, generally speaking, other nations have thought that these monopolies produce more embarrassment than advantage to society; and it may be observed that the nations which refuse monopolies of invention, are as fruitful as England in new and useful devices. —Thomas Jefferson (1813) LEARNING OUTCOMES After you have read this chapter, you should be able to: • Explain what a patent is and how it differs from a copyright • Explain the five major requirements for patentability • Describe the different types of patents • Provide examples of Internet-related technologies or business methods that may be protected by patents Introduction The dramatic growth of the Internet over the last two decades has been accompanied by an equally dramatic increase in the number of Internet-related patents. Although not a statutorily defined category, Internet patents provide exclusive rights to the hardware, software, and business methods that enable the Internet to function. For example, patents have issued on methods of: conducting online auctions, connecting people in online social networks, ranking search results, responding automatically to email messages, and gaming via the use of avatars. Software companies and other online businesses have amassed large portfolios of patents, which they acquired in the hope of achieving or maintaining an advantageous market position. For example, International Business Machines (IBM), which owns more U.S. patents than any other organization, obtained an average of more than 3,000 patents per year between 2001 and 2008. Also topping the list is Microsoft, which increased its annual patent acquisitions from 97 in 1996 to 2,026 in 2008. To put these impressive figures in perspective, renowned innovators such as 3M and Apple obtained 394 patents and 185 patents in 2008, respectively. Although patents have long been sought after by the most innovative companies, the race to acquire Internet-related patents has been especially intense since the landmark 1998 opinion of State Street v. Signature Corp. opened the floodgates to an important and controversial category of patents known as business method patents, which will be explored below. During the last five years, both Congress and the Supreme Court have been particularly active in considering some of the pressing issues in patent law in the context of a rapidly changing business environment. In this chapter you will explore the basic features of patents, study examples of how patents are being asserted in the online context, and examine the Supreme Court’s recent pronouncement in the ongoing controversy over business method patents. What Is a Patent? A patent is a government-granted, temporary right to exclude, awarded in return for an individual’s disclosure of a new and useful invention. In the United States, patents are granted by the U.S. Patent & Trademark Office (USPTO) and last for a nonrenewable term of 20 years. Exclusive Rights: Independent Invention Is Not an Excuse Patent owners enjoy the temporary right to exclude others from engaging in any of the following activities with respect to their inventions: • Making • Using • Selling (or offering to sell) • Importing into the United States Although the patent term is much shorter than the copyright term, the exclusive rights of the patent owner are broader than those of the copyright owner. For example, while copyrights are limited by a generous fair use doctrine, patents are not subject to any comparably broad exception. Similarly, while independent creation of a copyrighted work is a defense to a copyright infringement action, independent invention is not a defense to a patent infringement action. The right to exclude others from making a patented invention can be used to block even those who later independently invent the same invention without knowledge of the earlier patent. The fact that independent invention is not a valid defense to a patent infringement action means that online businesses must be aware of others’ patent rights even if they do not copy other companies’ business methods or technologies. For example, in 2002 Friendster launched its pioneering social networking website and became an early leader in a phenomenon that accelerated throughout the decade. Although it was eventually eclipsed in popularity by MySpace and later by Facebook, Friendster nevertheless succeeded in obtaining four patents between 2006 and 2009, including one entitled “System, Method and Apparatus for Connecting Users in an Online Computer System Based on Their Relationships Within Social Networks.”1 Even if MySpace and Facebook did not copy Friendster in designing and implementing their online businesses, they might nevertheless inadvertently infringe one or more of these patents by independently inventing the same patented subject matter. Although Friendster has declined in prominence in the United States, it continues to be a substantial social networking presence in Asia and could potentially assert its U.S. patents against other social networking websites in the years to come. Patents Confer “Negative Rights” It is a common misconception that patents provide patent owners with the affirmative right to make or practice an invention. In fact, patents confer only “negative rights” to exclude others from making, using, selling, offering to sell, or importing an invention. To illustrate, suppose Inventor 1 invents and patents a pencil. Inventor 2, seeking to improve the pencil, adds a rubber eraser to the tip of the pencil and patents the combination of a pencil with a rubber tip. Inventor 1 cannot make the pencil with the rubber tip because Inventor 2 has a patent on this combination. Note also, however, that Inventor 2 cannot make his own invention of a pencil with the rubber tip because Inventor 1’s patent on the pencil itself precludes him from doing so. In this scenario, illustrated in Exhibit 6.1, the improvement patent (“Pencil/Eraser Patent”) would be known as a blocking patent because it blocks the first inventor from making the improved version (pencil with a rubber tip) of that inventor’s own invention. EXHIBIT 6.1: Blocking Patents and Negative Rights Licensing and Cross-Licensing Because virtually all inventions build to some extent on earlier inventions, blocking patents are quite common. Of course, patent owners are not required to use their patents to prevent others from making, using, or selling their inventions. In some cases, the patent owner may instead wish to license (allow) another party to practice the invention, usually in return for royalties (money) or some other form of compensation. Licensing is particularly important in the software and information technology industries, where large firms may hold hundreds or even thousands of patents that could potentially block each other’s new technologies. To address this problem, industry players commonly engage in cross-licensing, whereby each party to the cross-licensing agreement is permitted to make, use, or sell the inventions protected by the other party’s patents. Cross-licensing can be viewed as a less costly alternative to litigation and provides firms with room to innovate without fear of inadvertently infringing on the other party’s patents. Obtaining a Patent Like trademarks, patents can be obtained by application to the USPTO. Unlike copyright and trademark rights, however, no patent rights arise until the USPTO grants the patent. In fact, publicly using, selling, or publicizing an invention without timely submitting a patent application can serve to bar later patenting by anyone, including the inventor. Patent Prosecution Inventors will normally hire a patent attorney to prosecute (apply for) the patent on behalf of the inventor. Note that the term “prosecute” is used in a different sense in patent law than it is in the context of criminal law, where “prosecution” refers to the bringing of an action by the government against an alleged criminal. Patent prosecution may last several years while a USPTO agent known as a patent examiner ensures that the invention described in the application meets the statutorily defined standards of patentability. Patent examination is much more rigorous and therefore much more expensive and time consuming than copyright or trademark examination. Inventions are kept secret for a period of 18 months, after which time they are generally published and thereby made available to anyone. If the examiner determines the invention to be patentable, the patent will issue (be granted) and the inventor may at that point enforce the patent against others. Despite the common imagery of the lone inventor toiling away in a garage or basement, more than 80 percent of United States patents are owned by corporations. Moreover, more than 50 percent of U.S. patents are owned by foreign entities (and half of these, by Japanese entities), a figure that has been gradually increasing over the last several decades. In turn, individuals and businesses from the United States own many patents in other countries. Appeal from an Examiner’s Adverse Decision Sometimes a patent examiner will determine that an invention is not patentable. When this occurs, the inventor may choose to appeal the examiner’s adverse decision to the Board of Patent Appeals & Interferences (BPAI), and if necessary seek further review in a specialized federal appeals court known as the Court of Appeals for the Federal Circuit (CAFC), often known simply as the Federal Circuit. The Federal Circuit hears both appeals originating from the applications rejected by the USPTO as well as appeals from district court patent decisions throughout the country. Patent Searching Individuals and businesses may search both issued patents and published applications at the USPTO website, http://www.uspto.gov, as well as via Google’s patent search tool at http://www.google.com/patents. Note that, with some exceptions, patent applications pending for 18 months are published and therefore available to the public whether or not the patent eventually issues. Because publication creates a risk of piracy notwithstanding any patent rights that are eventually granted, businesses may sometimes wish to forgo patent protection in favor of trade secret protection, a topic that will be explored in the next chapter. Patent Duration Patents last for a nonrenewable term of 20 years—far shorter than copyrights, which may persist for 120 years or even longer. While the 20-year term begins to run from the date the patent application is filed with the USPTO, the patent owner may not assert the right to exclude until the patent actually issues. Because most patent applications take two or more years to undergo examination before they issue, the effective term may therefore be somewhat less than 20 years. Exhibit 6.2 illustrates how the patent term might play out for a patent application filed in December of 2010 that remained pending for two and a half years before issuing as a patent. EXHIBIT 6.2: Patent Duration for Hypothetical Application Filed in 2010 A Shorter Duration for Internet Patents? Even the relatively short term of patents has sometimes received criticism as unnecessarily long. In 2000, Amazon.com’s CEO Jeff Bezos made headlines when he published an open letter in which he recommended a much shorter 3- to 5-year term for Internet and software patents. Bezos’ comments came as Amazon was embroiled in litigation with Barnes & Noble over Amazon’s notorious “1-click” patent, which allowed Internet purchases to be made with a single click of the mouse. Critics claimed that the patented 1-click method was not an invention but instead an obvious variation of existing practices. Bezos, who was one of four named inventors on the patent in question, defended Amazon’s right to assert its patent but believed that the existing patent term was too long and might even retard development of the Internet (see the opening chapter quote, which is excerpted from Bezos’ letter). You will read an excerpt from the court opinion in Amazon v. BarnesAndNoble later in the chapter. Although Congress has considered significant patent reform legislation since 2005, it is highly unlikely that a shorter patent term for Internet patents will be adopted in light of the United States’ international treaty obligations. In particular, the United States agreed to a minimum patent term of 20 years for inventions in all fields of technology when it became a party to the international Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) (see Chapter 14). Even in the absence of the TRIPS agreement, providing a different patent term for Internet patents would require a precise definition of what constitutes an “Internet patent” and could result in additional litigation regarding how a given invention should be classified. Patent Duration, 1790 to the Present The strong public reaction to the 1-click patent and Amazon’s surprising response are indicative of the critical importance of patent duration to the development of the Internet. In fact, perhaps the most significant limitation on the exclusive rights of a patent owner, once a patent is granted, is the limited patent term. Although the patent term has lengthened since the first patent act in 1790, the increases have been modest compared to the copyright extensions explored in the last chapter. Pursuant to the Constitution’s Intellectual Property Clause, Congress enacted the first patent statute in 1790, which authorized patents for up to 14 years. In 1836, Congress provided for a single 7-year extension to the original 14-year term, but only if the inventor had “failed to obtain … a reasonable remuneration for the time, ingenuity, and expense” devoted to the invention. The term was changed to a fixed 17-year nonrenewable term in 1861, a compromise between the 21-year period favored by some and the 14-year period favored by others. The 17-year patent term endured for over 130 years, until it was changed in 1995 to its present term of 20 years, measured from the date of filing. See Exhibit 6.3. EXHIBIT 6.3: United States Patent Duration (1790-2011) Patent Fees Obtaining and maintaining a patent is expensive. The initial fee for filing an application with the USPTO is $330. If an examiner determines that an invention is patentable, a fee of $1,510 must be paid to the USPTO before the patent will issue. Since 1980, maintenance fees have been required in order to enjoy the full patent term. Fees must be paid to the USPTO at 3.5 years ($980), 7.5 years ($2,480), and 11.5 years ($4,110) after the patent issues. Because patent fees change regularly, interested parties should consult the current fee schedule at http://www.uspto.gov. The largest expense of prosecuting and maintaining a patent, however, consists of attorney fees. When attorney fees are added to the USPTO fees, the cost of obtaining a patent, not including maintenance payments, ranges from approximately $10,000 to $50,000. The exact cost will depend on a number of variables such as the type of technology involved, the number of variations to the invention sought to be patented, and the hourly rate and efficiency of the attorney prosecuting the patent application. Maintenance Fees and Patent Duration If maintenance fees are not timely paid, the patent may lapse and the invention will fall into the public domain. Research indicates that the second maintenance fee is paid for approximately two-thirds of patents, while the third maintenance fee is paid for only about half of patents. Thus, the average effective patent duration may be several years shorter than the nominal 20 year term. By comparison, copyrighted works created after January 1, 1978 do not need to be renewed and therefore enjoy their full statutory term of 70 years or more. Trademarks, although potentially perpetual if renewed every 10 years, have been calculated to have an average expected life of around 15 years due to nonrenewal—roughly the same as the average expected life of a patent. Other Limitations to the Exclusive Rights Once a patent expires, it becomes part of the public domain and may be used freely by anyone without restriction. During the term of the patent, the patent owner’s rights are limited in a number of ways. As with copyright law, the first sale doctrine permits individuals to resell a product embodying a patented invention, so long as the initial sale of that product was authorized by the patent owner. Unlike copyright law, which has a broad fair use doctrine, patent law’s analogous experimental use exception (35 U.S.C. § 271(e)(1)) is extremely narrow and does not meaningfully impact the Internet. Finally, Section 1498 of Title 28 provides an eminent-domain-like power to the federal government to practice any invention without the permission of the patent owner, though doing so will give the patent owner a right to seek “reasonable and entire compensation” from the federal government in court. The Limited Territorial Reach of Patents U.S. patents provide rights only within the United States and its territories. Thus, an owner of a U.S. patent seeking to prevent those in other countries from copying the invention would have to seek patent protection in each country. Although businesses might find it more convenient to obtain a single patent that can be enforced throughout the world, no such patent regime exists. However, responding to the practicalities of a global economy, an international agreement known as the Patent Cooperation Treaty was concluded in 1970 to streamline the process of obtaining patents in multiple countries. Nevertheless, businesses tend to be selective when determining in which countries to seek patent protection in light of the costs and complexity of obtaining and enforcing rights in multiple countries. The estimated costs of obtaining patents presented previously ($10,000–$50,000) apply only to the acquisition of U.S. patents. Acquiring patent rights in other countries requires significant additional investments. Exporting Patented Inventions: Microsoft v. AT&T Given that the law allows patent holders to exclude others from making an invention within the United States, what happens when a business makes components of an invention—but not the completed invention—in the United States and then exports those components for assembly abroad? In 1984, Congress answered this question by enacting 35 U.S.C. Section 271(f), which prohibits exporting “all or a substantial portion” of a patented invention’s components for assembly abroad. In other words, a competitor cannot get around the prohibition against “making” by leaving out the final step of assembly which can be easily completed once outside the United States. However, in a twist on this theme, the Supreme Court had to decide whether Section 271(f) prohibited the exportation of software to a recipient located abroad who would then copy the software for installation on computers made and sold abroad. In effect, the court had to determine whether exported software constituted a “component” under the statute. The case, Microsoft v. AT&T,2 involved an assertion by AT&T of a patent covering a particular aspect of Microsoft’s Windows software. Microsoft conceded that Windows, when installed on a computer, infringed the AT&T patent if the relevant code was executed in the United States but argued that Windows did not infringe when exported from the United States and installed in another country, notwithstanding Section 271(f). The Court agreed, thus finding no infringement and limiting the territorial reach of AT&T’s patent. Central to the Court’s decision was the fact that the foreign recipient made and installed copies of the software exported by Microsoft, not the exported software itself. Although critics charged that this holding exalted form over substance, the Court emphasized that Congress was free to alter the result if it wished but that in the absence of Congressional action the territorial boundaries of the patent law should be respected. Patents versus Copyrights Patents and copyrights are often studied and considered together, and for good reasons. Each is a type of government-sanctioned monopoly that lasts for a limited period of time. Each provides compensation for an individual’s intellectual contribution with the ultimate aim of promoting the public good. Each is governed almost exclusively by federal law, the authority for which is provided by the Constitution’s Intellectual Property Clause (simultaneously known as the “Copyright Clause” or the “Patent Clause,” according to the context). Each has had an enormous impact on the development of the Internet. Yet patents and copyrights are distinct rights and are quite different in a number of important respects. For example, patents last for a much shorter period of time than copyrights, are far more expensive to obtain, and can only be secured by application to the USPTO. Several of the more important differences are summarized in Exhibit 6.4. EXHIBIT 6.4: Comparing Copyrights and Patents COPYRIGHTS PATENTS1 Subject Matter Creative works Inventions Duration • 120 years (works for hire) • life of author + 70 years 20 years Duration in 1790 14 years (28 if renewed) 14 years Registration Not required Required Ease of Obtaining Right Easy Difficult Cost None ($35 if registered) $10,000 $50,000 per patent Governing Statute Copyright Act of 1976 Patent Act of 1952 Federal Agency Copyright Office USPTO Public Right to Use Broad, due to fair use Extremely narrow The Anatomy of a Patent Exhibit 6.5 illustrates the structure of a typical patent document, using as an example excerpted portions of the Amazon “1-click” patent that underlies the case you will read later in the chapter. The patent specification is the portion of the patent that describes the invention and the manner and process of making and using it. The patent concludes with one or more claims that particularly point out and distinctly claim the subject matter which the inventor regards as his invention. Although the specification is helpful in understanding the invention, it is the claims that define the scope of a patent owner’s legal rights and therefore constitute a critically important part of the patent. The 1-click patent, for example, contains 26 claims, 5 of which are visible in Exhibit 6.5. EXHIBIT 6.5: The Amazon “1-Click” Patent (excerpts) What Can Be Patented? Requirements for Patentability The modern framework governing utility patents is set forth in the federal Patent Act of 1952, as amended, which provides that patents may only be granted for inventions that are: 1. Within the scope of patentable subject matter 2. Useful 3. Novel (new) 4. Nonobvious 5. Enabled The reason patents are more difficult, time consuming, and expensive to obtain than copyrights or trademarks is that patent attorneys must draft patent applications such that they comply with these statutory requirements, personnel at the USPTO must examine patent applications to ensure compliance, and, when the attorney and examiner disagree, both must work to determine whether the patent application can be revised so that it does comply. The five requirements are explored below. The Broad Scope of Patents: Section 101 Subject Matter Congress set forth four broad categories of patentable subject matter in Section 101 of Title 35: • Processes • Machines • Manufactures • Compositions of matter Under Section 101, patents have been issued on inventions as diverse as computer software, image compression technology, pharmaceutical ingredients, diapers, an isotope of the element Americium, a method of combing hair to conceal partial baldness, and a method of organizing and managing mutual funds using a computer. From these examples it can be seen that the scope of patentable subject matter is extremely broad, encompassing virtually any invention that is novel, useful, and nonobvious. In fact, in the legislative history leading up to the Patent Act of 1952, Congress famously declared its intent that patents be granted on “anything under the sun that is made by man.” This phrase gained renown when the Supreme Court quoted this portion of the legislative history in support of its decision in Diamond v. Chakrabarty3 that a genetically modified bacterium constitutes patentable subject matter. Limits to Patentable Subject Matter Although the scope of patentable subject matter is extremely broad, it is not unlimited. The Supreme Court in Chakrabarty noted that the following have been held not patentable: • Laws of nature • Physical phenomena • Abstract ideas Thus, neither the law of gravity nor E = mc2 would constitute patentable subject matter. Another example of unpatentable subject matter is provided by the digital watermarking technology that you studied in Chapter 5. Recall that copyright owners have developed a means for embedding watermarks in digital audio or video files to help detect and prevent unauthorized copying. Unfortunately, embedding watermarks can cause distortions when the file is played, which can detract from the listener’s or viewer’s experience. When an inventor named Petrus Nuijten developed a technique for reducing these distortions and tried to patent it, the Federal Circuit held that the electromagnetic signal itself, in which the watermark was embedded, did not constitute patentable subject matter because it was not a process, machine, manufacture, or composition of matter.4 However, the Court did not upset the USPTO’s determination that the process used to create the signal was patentable. Thus, Nuijten was able to obtain a patent on his invention, though one that was somewhat narrower than the one he initially sought. Utility The utility (usefulness) requirement is a very low bar and rarely precludes patentability in the Internet context. To meet the minimal utility requirement, an invention must merely operate as described. Furthermore, there is no need for the invention to be any better than existing technologies or methods. Thus, patents have issued on: (Ferrera 169-180) Ferrera, Gerald R., Margo E. Reder, Stephen Lichtenstein, Robert Bird, Jonathan Darrow, Jeff. CyberLaw: Text and Cases. Cengage Learning, 01/2011. VitalBook file. LEARNING OUTCOMES After you have read this chapter, you should be able to: • Describe the essential characteristics of a trade secret • Provide examples of trade secrets • Explain the significance of the Uniform Trade Secrets Act (UTSA) and the Economic Espionage Act (EEA) • Describe the types of contracts businesses can use to protect secret information Introduction The Information Age It is often said that we live in the Information Age, a period roughly coinciding with the revolution in computing and telecommunications technologies that took shape in the 1960s and that has accelerated ever since. In fact, so much information is now available that several commentators have recently compared it to pollution and emphasized the need for effective information-sorting mechanisms. An Information Explosion To provide a sense of the amount of information being generated, the U.S. Copyright Office registered 526,378 creative works in a recent year, while the number of U.S. patents issued since 1975 exceeds the number granted during the preceding 185 years combined. In the online environment, the U.S. Library of Congress reports that its web archive contains approximately 160 terabytes (1 terabyte — 1,000 gigabytes) of data as of 2010—a lot of data, but nothing compared to the more than 250 exabytes (1 exabyte = 1 million terabytes) of data estimated to have been transferred over the Internet in that year alone. Information, which constitutes the raw material on which many businesses depend for their vitality, is clearly critical to the Information Economy. Patents and Copyrights Cannot Protect All Data Although information advantages may often be a key component of the competitive strategy of a business and a decisive factor in its success, a great deal of information cannot be protected by either copyright or patent law. For example, a secret recipe such as the formula for Coke may not exhibit a sufficient degree of creative expression to be copyrightable, while the public sale of Coke for more than 100 years can prevent the Coca-Cola Company from patenting the recipe under the patent law novelty provisions (recall the one-year grace period from Chapter 6). Even where information could be patented or registered as a copyright, a business may prefer trade secrecy for its potentially longer period of monopoly, out of concern that disclosed information could be pirated or for other reasons. This chapter explores the clandestine world of trade secrets, discussing what they are, how businesses can develop and protect them, and what a business can do to ensure against their loss. What Is a Trade Secret? By far the most famous example of a trade secret is the Coke formula mentioned above, which has remained secret since its creation by an Atlanta pharmacist in 1886. Similar examples include KFC’s secret chicken recipe, McDonald’s special sauce recipe for the Big Mac, the formula for the popular lubricant WD-40, and the formula for Listerine (which began as a trade secret in the 1880s but was disclosed in the Journal of the American Medical Association in 1931). In addition to formulas and recipes, many other types of information can also qualify for trade secret status. Examples include: • Computer software • Customer lists that indicate purchasing histories and contact information • Technical specifications for building or operating a device or service • Marketing plans, including dates for new product launches Virtually any information that provides economic value to a business can potentially be protected as a trade secret, and so the scope of subject matter covered by trade secret law tends to be broader and less precisely defined than it is for, say, patent law. Trade Secrets: Governing Law Unlike patents and copyrights, which are governed almost exclusively by federal law, trade secrets are primarily governed by state law and the precise definition may therefore vary somewhat from one state to another. State Law: The Uniform Trade Secrets Act The Uniform Trade Secrets Act (UTSA) has helped to harmonize trade secret law throughout the United States. The UTSA is one of more than 250 uniform laws that have been developed by a nonprofit organization known as the National Conference of Commissioners on Uniform State Laws (NCCUSL). The NCCUSL, although not itself a legislative body, drafts and proposes uniform laws that are then submitted to state legislatures. States may in their discretion decline to adopt the uniform laws, adopt them in their entirety as originally drafted, or adopt a partial or modified version as desired. The UTSA was originally submitted to the states for adoption in 1979. As of 2010, 46 states and the District of Columbia have adopted some version of the UTSA. Three of the four remaining states—New York, New Jersey, and Massachusetts—are currently considering its adoption. These three states as well as the fourth state, Texas, continue to rely primarily on judicially created common law rather than a legislatively enacted statute as the basis for their trade secret law. Federal Law: Economic Espionage Act of 1996 As the Cold War ended and large-scale electronic commerce began to take root, Congress observed that classic military espionage was giving way to economic espionage—the deliberate theft of proprietary business information for the benefit of a foreign entity or government. The ease with which business information could be surreptitiously downloaded to a diskette or even accessed at a distance through computer hacking was thought to have contributed to a quadrupling of economic espionage incidents over a four-year period, resulting in related losses to businesses of $63 billion annually. During the 1990s, the FBI investigated allegations of economic espionage by individuals or organizations from at least 23 different countries, and considerable evidence suggested that foreign governments were supporting espionage against American companies. In order to provide federal law enforcement with the tools to combat both foreign and domestic trade secret theft, Congress enacted the Economic Espionage Act of 1996 (EEA), criminalizing the theft of trade secrets and providing for penalties of up to $10 million as well as imprisonment for up to 15 years. In 2006, the first EEA convictions were obtained when Fei Ye and Ming Zhong pled guilty to charges brought under the Economic Espionage Act in connection with a scheme to steal microprocessor secrets from Sun Microsystems and three other companies. The two men had been caught at the San Francisco airport while attempting to board a plane to China. Although the UTSA and the EEA contain similar prohibitions against the theft of trade secrets, only the government can bring an action under the EEA. Trade Secrets Defined The UTSA defines a trade secret as: [I]nformation, including a formula, pattern, compilation, program, device, method, technique, or process, that: (i) derives independent economic value … from not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use, and (ii) is the subject of efforts that are reasonable under the circumstances to maintain its secrecy. The Economic Espionage Act provides a very similar definition, specifying that trade secrets include: [A]ll forms and types of financial, business, scientific, technical, economic, or engineering information, including patterns, plans, compilations, program devices, formulas, designs, prototypes, methods, techniques, processes, procedures, programs, or codes … Like the UTSA, the Economic Espionage Act provides that information will only qualify as a trade secret if (1) the information is valuable because it is not generally known and would be difficult to lawfully acquire and (2) reasonable measures are taken to preserve secrecy. These two principal requirements for trade secret status will be explored in turn. Valuable Because Not Generally Known Information may be considered generally known if it is widely available on the Internet, published in a corporation’s annual report, discernable via a public factory tour, or disclosed in marketing materials or professional journals. For example, when purported trade secrets regarding the Church of Scientology were posted to an Internet newsgroup where “competitors” of the Church were likely to look, the posted information became generally known.1 However, the mere fact that information has been published on the Internet does not necessarily make it generally known, as this chapter’s opening quote suggests. For example, where allegedly secret information regarding semiconductor chips was posted on a Chinese-language website but no evidence was proffered that any competitor was aware of the posted information, the court held that trade secret status had not been destroyed.2 Even if information is not generally known to the public, it will still be ineligible for trade secret protection if it is clearly understood in the relevant industry. Under the UTSA definition provided above, the key is whether the information is generally known to “persons who can obtain economic value from its disclosure or use.” The determination of whether information is generally known can be compared to the patent law concept of novelty, but is not identical with it. For example, although patentability may be precluded based upon a relatively obscure but publicly available document, trade secret status might not be precluded so long as the information contained in the document is not generally known to competitors. Not Readily Ascertainable In addition to not being generally known, a trade secret must not be readily ascertainable by those who can obtain economic value from its disclosure or use. Information is readily ascertainable if it is available in trade journals, reference books, or published materials, or if it can be duplicated quickly and inexpensively or reverse engineered. Although there are no bright line rules, court decisions suggest that information is more likely to be readily ascertainable if it can be derived within a few minutes or hours and less likely if derivation would require months or years. Whether or not information is readily ascertainable might be compared to patent law’s obviousness standard, in that the easier the information is to duplicate or acquire the less likely it is that the information will be eligible for protection. Simultaneous Protection by Patent and Trade Secret Law? Information published in a domestic patent or patent application will ordinarily be considered generally known. This is important because it means that a business cannot simultaneously protect the same information as a patented invention and a trade secret. The business must choose one or the other. Nevertheless, it is a common practice for businesses to obtain patents on inventions without disclosing certain information useful in making the invention commercially practicable, such as cost information, supplier information, or information about failures during the course of the invention’s development. This additional information, not described within the patent itself, is instead kept as a trade secret and may be separately licensed as know-how. Know-how may also arise after the application for patent is filed. For example, when Larry Page applied for his PageRank search engine patent in 1998, the Patent Act required him to describe the invention in sufficient detail so as to enable others to make and use it. However, because the Patent Act does not require inventors to update the patent disclosure if better techniques are later developed, any improvements by Page or others to the search engine algorithm, subsequent to the time of filing the patent application, could be kept as trade secrets. (Those improvements would not, however, be protected by the patent.) Because businesses that in-license (receive) intellectual property generally desire the most commercially effective means of offering a product or service, intellectual property owners will often out-license (permit others to use) both patent and trade secret information. Reasonable Measures to Maintain Secrecy Not only must the information not be generally known, but businesses must take measures that are reasonable under the circumstances in order to maintain secrecy. Although the circumstances may vary from case to case, reasonable measures might include: • Restricting access to databases, such as by requiring the use of a password • Restricting access to sensitive areas, such as by maintaining sensitive documents in locked cabinets or rooms • Limiting sensitive information to employees that have a “need to know” • Marking proprietary documents as “secret” or “confidential” (see Exhibit 7.1) • Embedding watermarks in electronic documents to help identify the source of the disclosure • Conducting “exit interviews” when employees leave the company to remind them about confidentiality obligations and the proper treatment of proprietary information EXHIBIT 7.1: Sample Document Marked as “Top Secret” In addition to these operational and technological measures that businesses can take to maintain trade secret status, a number of contractual measures are available. These include requiring employees or others to sign nonsolicitation agreements, nondisclosure agreements, and covenants not to compete. Nonsolicitation Agreements A trade secret owner wishing to decrease the risk that its employees will be tempted to quit their jobs to work for one of the trade secret owner’s business partners may ask those business partners to sign nonsolicitation agreements. These agreements prevent the partner businesses from attempting to lure away the trade secret owner’s employees, who may possess valuable trade secrets. Similarly, a trade secret owner worried that its employees may try to start a competing enterprise may include a nonsolicitation clause as part of the employment contract in order to prevent the employee from soliciting former colleagues to join the new enterprise. Nondisclosure Agreements (NDAs) As the name implies, a nondisclosure agreement (NDA) is a contract that requires a party not to disclose certain specified information that a business considers important to its competitive advantage. Businesses may routinely enter into NDAs with employees, contractors, consultants, clients, business partners, and potential investors. The NDA should indicate what specific information is subject to nondisclosure and how long the nondisclosure obligation will persist. Businesses should be aware that an NDA that imposes nondisclosure obligations for too short a duration can undermine a plaintiff’s claim that “reasonable measures” have been taken. For example, in 2008 a court concluded that plaintiff Silicon Image was unlikely to prevail in its trade secret misappropriation claim against competitor Analogix, noting that the plaintiff’s NDAs imposed confidentiality obligations for only two to four years and that many of them had expired. The proprietary information in question related to the HDMI technology common to popular consumer electronics such as High Definition Televisions (HDTVs) and some cell phones. Although it was unclear to what extent the ruling was based on the NDA duration, the court cited earlier opinions where NDA durations of one to ten years had been held too short. Covenants Not to Compete A covenant not to compete, also called a noncompete agreement, is a provision in an employment contract in which an employee promises not to compete with the employer should the employment relationship end. Covenants not to compete will usually be enforced when they are reasonable, which requires courts to balance a number of factors. Courts must weigh the legitimate interests of the employer who has invested resources and effort in training an employee against the public interest in receiving the benefit of the employee’s services as well as the employee’s interest in earning a living. In order to be enforced, covenants not to compete must generally be reasonable across two dimensions: (1) duration and (2) geographic scope. As a rule of thumb, noncompetition clauses that prohibit former employees from competing with the employer for a term of more than two years will tend to be unreasonable. However, the standard of reasonableness is highly dependent on the circumstances and tends to be shorter within the Internet context. For example, one early Internet trade secret case famously held that a one-year covenant was too long to be enforced because “in the Internet environment, a one-year hiatus from the workforce is several generations, if not an eternity.”3 Similarly, in DoubleClick v. Henderson,4 the court granted an injunction enforcing a noncompete agreement for only six months rather than for the one-year period sought by plaintiff DoubleClick, an Internet advertising company. Covenants Not to Compete in California Although states generally disfavor covenants not to compete, California is one of a small number of states that uniformly refuse to enforce such covenants absent a specific statutory exemption. Because of the large number of Internet-related businesses located in California, its committed policy against covenants not to compete is notable. Google In some cases, businesses seeking to hire employees who have signed noncompete agreements will make great efforts to have the dispute resolved by a California court. For example, in 2005 Google sought to hire Kai-Fu Lee as its Vice President of Engineering. Unfortunately for Google and Lee, who was reportedly offered $10 million by Google to accept the position, Lee had signed a one-year noncompete agreement with Microsoft, where he had been working since 1998. Microsoft, seeking to enjoin Lee’s employment, filed suit in Washington State. Google, seeking the benefit of California law in its efforts to overcome the covenant not to compete, then filed a separate suit in California. However, the California judge declined to rule in the case until the Washington court issued its decision, and the parties eventually settled the dispute on confidential terms with Lee heading up Google’s China operations. (See Chapter 10 for additional treatment of this case.) Government Disclosure Under the Freedom of Information Act (FOIA) Under the Freedom of Information Act (FOIA, pronounced “foy-uh”), the public may obtain government documents from federal agencies upon request. Because businesses must frequently disclose information to the government as part of the ordinary regulatory process, secret business information may potentially be placed at risk. Anticipating this problem, FOIA provides a number of exceptions to its disclosure provisions for trade secrets and other sensitive information. In some cases, businesses may find it advantageous to file requests for confidentiality with the relevant government agency in order to ensure nondisclosure and to help demonstrate that reasonable measures have been taken to maintain secrecy. For example, shortly before releasing the iPhone 4 in 2010, Apple submitted a formal request to the Federal Communications Commission to maintain in confidence certain iPhone photographs and schematics as well as other documents containing technical and design information. Justifying Trade Secret Law Trade secret law and the contractual measures employed to protect trade secrets impose a number of costs on society. They can restrict employee mobility and limit employees, at least temporarily, from utilizing their most valuable skills. Conversely, they can create challenges for employers seeking to hire employees with relevant skills. Trade secret law can restrict the free flow of ideas, reduce competition, and slow technological and business development. Moreover, trade secret litigation and the desire to avoid litigation impose significant transaction costs in the form of uncertainty, attorney fees, and paperwork. Given these disadvantages, it might be wondered why trade secret law exists at all! Two principal rationales have been offered in justification of the legal protection of trade secrets. First, trade secret law can promote innovation, particularly with respect to subject matter that may not satisfy the requirements for patentability. Second, trade secret law reinforces standards of commercial ethics. These two rationales are explored below. Innovation and Efficiency Like trademark, patent and copyright law, trade secret law is thought to promote innovation by ensuring that those who have invested in developing information can benefit from their efforts without the fear of free riding. In addition, trade secret law can help to promote business efficiency. For example, without trade secret law employers might hesitate to hire strangers of unknown loyalty and instead engage in nepotism (hiring of relatives) or cronyism (hiring of friends) even if other more qualified workers are available. By providing a right of action for misappropriation, trade secret law encourages employers to select their employees from a broad range of applicants, fully train them, and share information with them as required for efficient business operations. More generally, by providing a legal structure within which information may be shared yet maintained in confidence, trade secrets encourage the formation of stable business relationships. Business Ethics Second, trade secret law is justified on the basis of a number of ethical or moral dimensions. Trade secret law provides a legal basis for reinforcing loyalty and confidentiality obligations that many believe are inherently part of the relationship between employer and employee. The definition of misappropriation includes within its reach the acquisition of a trade secret by bribery or misrepresentation, both of which have obvious ethical dimensions. In addition, trade secret law prohibits economic espionage, which can be characterized as a form of theft that may sometimes involve deceit. In short, trade secret laws can be justified on principles of good faith and fair dealing in addition to principles of market efficiency. “Obtaining” and Maintaining a Trade Secret Unlike trademarks, copyrights, and patents, there is no process by which a trade secret can be registered, and in fact no generally accepted procedure by which particular information becomes a trade secret. Instead, trade secrets come into existence when valuable information that is not generally known to competitors is developed or acquired and reasonable efforts are made to maintain the information as a trade secret. Trade Secret as Alternative to Patent and Copyright Trade secret protection is sometimes considered an alterative to patent or copyright protection, though the various types of intellectual property may provide overlapping rights in some cases. For example, different aspects of a single computer program may be simultaneously protected by patent law, copyright law, and trade secret law. When determining which type or types of intellectual property to pursue, a number of important differences should be borne in mind. For example, copyrighted works reflect creative expression and therefore tend to derive value from being publicly observable, while information may be protected as a trade secret only if it is not generally known to the public. Other differences relate to the duration of protection and costs associated with maintaining information in confidence. Costs of Trade Secrets Trade secrets are often described as a less expensive alternative to patents, in part because they do not require registration or maintenance fees. Yet trade secrets do impose costs on businesses. For example, designing and maintaining secure computer systems that have appropriate password, encryption, and firewall controls is not costless. In addition, there are transaction costs associated with limiting access to information and requiring business partners, suppliers, and clients to enter into nondisclosure and other agreements. Finally, some employees may find the security measures inconvenient or distasteful, particularly where contractual provisions frustrate their ability to accept future employment. Duration of Trade Secrets In light of the fact that trade secrets are not registered with the government, they also do not need to be “renewed” nor do they expire after a statutorily prescribed period of time. So long as the trade secret remains secret and continues to confer economic value, the owner’s rights in the trade secret under state and federal law will persist. Trade secrets are therefore potentially perpetual, in contrast to the limited terms of copyrights and patents. In the online environment, the rapid pace of development may mean that information will lose much of its value before the 20-year patent term expires. The potentially perpetual duration of trade secrets may therefore be of only nominal added value. Businesses may nevertheless prefer trade secret protection over patent protection, however, because the patent law disclosure requirement increases the chance of unauthorized copying. Although patents confer the right to prevent others from making, using, or selling an invention, infringement of patented processes or business methods may be difficult to detect. Moreover, copying may occur in other countries where patent rights have not been secured at all. Maintaining information as a trade secret avoids the need to obtain patent protection in multiple countries or to engage in expensive and time-consuming litigation to enforce those patents. Unsuccessful Patenting Places Secret Information at Risk Although trade secrets are potentially perpetual, the public disclosure of a secret could bring a premature and abrupt end to trade secret status. For example, if a business were to file a patent application, the application including its detailed description of the invention would typically be made public after 18 months under applicable law. However, if the USPTO examiner ultimately determined that the invention described in the application was unpatentable, the business would have destroyed its trade secret without securing the benefit of a patent. A similar result would occur if a patent was issued by the USPTO but later invalidated by a court. Third Party Patents May Limit Trade Secret Duration In addition, if a business decides to protect information as a trade secret and a third party independently develops and patents an invention based on the same information, the business can find itself unable to continue using its own trade secret information. Although the special exemption for prior users discussed in the previous chapter provides some protection to those who used business methods before they were patented by others, this exemption does not apply to hardware or any category of invention other than methods of doing business. A business’s own partners may even attempt to patent the former’s trade secrets. This was the case in Corbis Corp. v. Stone,5 which pitted a small start-up company that had developed a digital image, video, and music tracking system against Corbis Corp., a photograph licensing corporation owned by Microsoft’s Bill Gates. Originally business partners, the two companies found themselves embroiled in litigation with each party accusing the other of trade secret misappropriation. Ultimately, the dispute was resolved by a jury, which found that Corbis had misappropriated trade secrets by secretly filing a patent application based on Stone’s proprietary technology. The court ordered Corbis to pay more than $20 million, including $750,000 in attorney fees and costs based on its determination that the misappropriation was “willful and malicious.” Independent Invention and Trade Secrets In contrast to patents, trade secrets provide no protection against others who independently develop the same information, regardless of whether those third parties attempt to patent the information. Exhibit 7.2 summarizes some of the differences between patent and trade secret protection. EXHIBIT 7.2: Comparing Trade Secrets with Patents TRADE SECRET PATENT1 Governing Law UTSA, EEA Patent Act of 1952 State/Federal Mostly state law Federal law Subject Matter Information that is valuable and not generally known Inventions that are novel, useful, and nonobvious Duration Perpetual (potentially) 20 years Protection Against Independent Invention No Yes Third Party Patent Third party patent ends trade secret rights “Blocking patents” limit innovative freedom Registration No Yes Disclosure Undermines secrecy Required Costs No registration costs (but transaction costs) High registration costs Prevention of Copying Secrecy impedes piracy Right to sue infringer Trade Secrets and Software It was noted above that software could be simultaneously protected by both copyright and trade secret law. Given that copyright registration involves making a copy available for public inspection in the Copyright Office, it might be wondered how software code can remain secret while simultaneously fulfilling copyright registration requirements. The answer in part is that under Copyright Office regulations only the first and last 25 pages of source code need be deposited, and the portions of the source code that contain trade secrets may be redacted (blocked out). Because large software programs may be composed of millions of lines of code, the Copyright Office deposit requirements can result in only a tiny fraction of the code being made available to the public. In addition, although creative works deposited with the Copyright Office are available for public inspection, a member of the public wishing to view deposited material must sign an agreement not to copy the material pursuant to formal Copyright Office policies. Thus, when a software company deposited unredacted copies of its software instruction manuals with the Copyright Office, the court found that trade secret protection was not lost owing to the restrictions on public copying and the lack of evidence that anyone had actually viewed the deposited material.6 A Dynamic Business Environment Places Trade Secrets at Risk Trade secrets can be lost if information leaks cause once-secret information to become generally known by competitors. One of the greatest threats to the maintenance of secret information is disclosure by employees, who can be viewed as living repositories of a business’s economically valuable information, whether financial, strategic, or technological. The skills, experiences, and business knowledge acquired by an employee over the course of employment are as mobile as the employee who possesses them. And employees are indeed mobile, with some reports indicating that the average person will change jobs ten or more times over the course of the person’s career. Employment in the Internet sector is particularly fluid. Employee Mobility Although employee mobility and associated knowledge spillovers have been cited as key contributors to the success of technology centers such as Silicon Valley, every change in personnel brings with it the risk that proprietary information will fall into the hands of a competitor. In fact, the most common scenarios that give rise to trade secret litigation include: • An employee leaving an employer to begin work for a competitor • An employee leaving an employer to start a new competing company • An employee disclosing proprietary information on the Internet (either directly or via a third party) • A relationship between business partners that ends poorly Many trade secret disputes therefore arise as a result of a broken employment or business relationship, which one scholar has analogized to a “bitter divorce.” For example, in 2004 a social networking site called ConnectU sued a then-fledgling Facebook based upon allegations that Facebook’s founder Mark Zuckerberg stole ConnectU’s source code and business plan. Zuckerberg and ConnectU’s founders had originally worked together to design a social networking site while they were classmates at Harvard before the relationship soured. (See Chapter 2.) Inadvertent Disclosure Secret information may also be placed at risk inadvertently by well-intentioned but perhaps overzealous employees who, for example, tout product benefits online or at trade shows in an attempt to generate business. In some cases business partners, consultants, or law firms may inadvertently disclose confidential information. For example, although ConnectU and Facebook attempted to keep their 2008 settlement agreement confidential, ConnectU’s law firm inadvertently boasted in its newsletter that it had achieved a $65 million settlement for ConnectU. Although the settlement figure might not itself qualify as a trade secret, the incident serves to demonstrate how easily confidential information can be disclosed. Third Party Espionage Secret information may also become vulnerable as a result of intentional espionage by competing businesses or even foreign governments, a concern that resulted in the passage of the Economic Espionage Act described above. The Internet Computers and the Internet have served to magnify the risk of trade secret loss. Networked computers enable departing employees to easily email confidential information to themselves or others, or download large volumes of data to a tiny flash drive. Unrelated third parties may be able to hack into networks to view or download information from across the street or around the globe. Even loyal employees can place trade secrets at risk by mistyping an email address or by traveling with a company laptop that is then vulnerable to theft by a third party. In addition, the ability to post information on the Internet dramatically increases the speed with which secret information can be disseminated as well as the number of people who may potentially access it. An example of the risks posed by departing employees is provided by iRobot, the maker of both the well-known Roomba floor vacuuming robot as well as a military reconnaissance and bomb-disposal robot known as the PackBot. When engineer Jameel Ahed resigned from iRobot in 2002 to start competing enterprise known as Robotic FX, he allegedly sent confidential documents from his iRobot email account to a Robotic FX email account and worked to build a less expensive version of the PackBot. In 2007, Ahed’s former employer iRobot filed suit against Robotic FX and Ahed alleging misappropriation of trade secrets and breach of a nondisclosure agreement. Although the dispute settled before trial, iRobot largely prevailed. Ahed was prohibited from competing in the robotics industry for five years and his PackBot copycat product is now among the product offerings of iRobot. The domain name www.roboticfx.com now redirects to the iRobot site. (Ferrera 207-218) Ferrera, Gerald R., Margo E. Reder, Stephen Lichtenstein, Robert Bird, Jonathan Darrow, Jeff. CyberLaw: Text and Cases. Cengage Learning, 01/2011. VitalBook file.
Purchase answer to see full attachment