Written Statement of the

Coalition Against Database Piracy

on H.R. 2652, the Collections of Information Antipiracy Act of 1997





The Coalition Against Database Piracy ("CADP") welcomes the opportunity to share with the Subcommittee its views on why Congress should enact fair database protection legislation, in general, and its views on H.R. 2652, "The Collections of Information Antipiracy Act of 1997," in particular.



Mr. Chairman, CADP thanks you for your leadership in this important area. Our members are especially grateful for your recognition of the crucial role that databases play in our information society and the need for Congress to enact legislation that will fill the gap in protection afforded collections of information under current U.S. law.



CADP is an ad hoc group composed of large and small U.S. database providers who stand to suffer grievous harm--and whose thousands of employees' jobs will be at risk--if you do not promptly enact federal database legislation that is effective and fair. Its members include the American Medical Association; the Information Industry Association; The McGraw-Hill Companies; Phillips Publishing International, Inc; Reed Elsevier Inc.; Skinder-Strauss Associates; the Thomas Publishing Company; The Thomson Corporation and Warren Publishing, Inc.



CADP's members are an integral part of the U.S. database community. Today, the United States is the world leader in the creation and distribution of informational databases. Our members employ or represent many thousands of editors, researchers, and others who gather, update, verify, format, organize, index and distribute the information contained in their vast array of database products. As a result of the efforts of CADP members and others in the U.S. database community, scientists, researchers, academics, scholars, businesses, government and consumers have ready access to a wealth of user-friendly, reliable and up-to-date information they consult daily.



Mr. Chairman, your bill addresses a basic unfairness in our legal system: its failure to protect adequately the interests of those whose hard work and substantial financial investments result in the creation and dissemination of valuable databases. H.R. 2652 is about eliminating the inequity in a legal regime that allows an unscrupulous competitor to copy with impunity the contents of someone else's compilation and then destroy the first compiler's market by selling a competing, less expensive product. It is also about rectifying the injustice that takes place when a dishonest customer or a "cyberprankster"--without permission--electronically copies and makes freely available through the Internet a database compiled by a hard-working entrepreneur. In sum, it is about helping restore fairness to the database marketplace.



CADP's goal is simple and straightforward: to deter piracy that causes commercial harm to database creators while maintaining the traditional balance between the legitimate interests of the owners and users of informational products. More specifically, we seek enactment of legislation recognizing:



• the crucial part databases play in our information-intensive economy;



• the amount of financial, human, technical and creative resources that a database maker invests in its products;



• the unfairness of allowing those who do not make such investments to engage in unauthorized uses of, and thereby injure the market for, the hard work of others;



• the vulnerability of informational products to illegal copying and dissemination--especially in a digital environment where materials can be copied and distributed with a few clicks of a mouse and at a fraction of the cost of their development;



• the absence of sufficient protection accorded collections of information under current U.S. law;



• the incentive for database makers to continue to make their substantial investments will be dampened considerably unless an effective legal regime is implemented; and



• the appropriate balance between the respective interests of database creators in protecting their investments and the interests of users in a robust, unimpeded flow of information.



Mr. Chairman, we believe that both creators and users alike will benefit from a clear, well-defined statute setting out permissible and impermissible uses of protected information products. Your bill, H.R. 2652, is an important first step towards striking the correct balance between the important interests of both owners and users of collections of information. CADP remains committed to working with Congress and other interested parties in crafting such legislation.



CADP also believes that the time is ripe for congressional action. As discussed below, the risks to database creators will only increase as our society becomes more and more dependent on computers and digitized information and as technology provides new and more efficient ways to reproduce and distribute information products. Moreover, as the Subcommittee begins its review of H.R. 2652, it has the benefit of two invaluable tools--the Copyright Office Report--entitled Legal Protection for Databases (the "Copyright Office Report") and a compelling economic analysis prepared by Drs. Laura D'Andrea Tyson and Edward F. Sherry entitled Statutory Protection for Databases: Economic and Public Policy Issues (the "Tyson Report").



The Copyright Office Report provides an excellent overview of the historic scope of legal protection for databases as well as recent legal and technological changes affecting the business environment for database providers and users. A review of this objective and comprehensive study underscores the global importance of statutory database protection.



The Tyson Report complements the Copyright Office Report. It sets out the strong economic rationale for legal protection from piracy for databases created by private entities. It also describes how an appropriately crafted database law can meet the challenge of protecting the enormous human and financial investments of database producers, while at the same time ensuring that information is available to support education, scientific progress and economic growth. Both these reports merit the Subcommittee's serious scrutiny.



The need for prompt congressional action is also underscored by the recent developments in the European Union ("EU")--an obvious effort by the EU to ratchet up its share of the world-wide database market, primarily at the expense of U.S. database providers. Last year, the EU adopted a sui generis database protection directive.(1)

The EU Directive requires its members to adopt conforming database protection legislation by December 31, 1997. Under the Directive, a database company outside of the European Union--such as those in the United States--is not within the reach of the Directive's provisions unless its own country provides a level of protection that the EU deems "equivalent" to its own. Without comparable U.S. legislation, U.S. databases will suffer a significant competitive disadvantage in the huge EU market: databases from EU nations will enjoy the benefits of sui generis database protection and U.S. products will not.(2)

If the U.S. does not act promptly, existing and future databases created in this country will be free for the taking in EU member states, while EU-produced products or those pirated by EU producers from the U.S. database market will be protected in the EU.





Overview of Database Industry



The types of databases that abound in this Information Age are a far cry from just the traditional printed compilations that have existed for centuries. Today, information products come in a variety of formats--print, CD-ROM and online. They provide a vast array of comprehensive data vital to the successful operation of our economy--information about medicine, health, intellectual property, communications, finance, banking, business, news, travel and defense. U.S. databases provide the world everything from antidotes to poison, to vital information about prescription drugs, to the keys to building safer cars, to comprehensive compilations of patents and related information.



Database creators play a crucial role in our information-driven society. They add immense value to a mass of otherwise unintelligible, disparate data. As you indicated, Mr. Chairman, at the time you introduced H.R. 2652, collections of information enable users "to retrieve from this haystack of information the specific actual needle that they need to solve a particular economic, research or educational problem."



Moreover, these products greatly reduce the time and effort needed to conduct important research and assure the reliability of all the facts included. Without access to these products, vast amounts of valuable and systematically organized information would be unavailable to many users who would not be able to replicate easily the financial and human investments made by the database compiler.



The creation, maintenance, and distribution of these information products do not come cheaply or without great human effort. Many American jobs depend on a healthy, vibrant U.S. database industry. As noted above, CADP's members employ or represent thousands of editors, researchers, and others. They also invest millions of dollars in hardware and software to manage these large bodies of information.



A few examples of the databases produced by CADP members--large and small--help to illustrate the importance of these products to our society.



1. Warren Publishing, Inc.'s Television and Cable Factbook. The Factbook is a unique directory containing business profiles of all U.S. cable TV systems, licensed broadcast video facilities (i.e., full-power TV stations) and related industries and services (program suppliers, equipment manufacturers, regulatory agencies, et al.). The Factbook is comprised of 3 volumes totaling more than 5,000 pages annually and also is available on CD-ROM.



In addition to the print products, the databases used to produce the Factbook are widely used by various sectors of the U.S. telecommunications business and academic communities. Warren Publishing makes electronic sales of the Factbook databases to clients for analyses on their own computers; Warren Publishing also performs customized analyses upon commission by clients.



Warren Publishing assigns 18-20 full-time employees to the Factbook. They gather, verify, edit and format data for use in both the print and electronic versions. Two other people are employed full-time for sales and fulfillment of customized reports and databases, representing more than one-third of Warren's total workforce. In addition, Warren Publishing annually hires an average of 10 independent contractors to input data from returned Factbook questionnaires. Warren spends tens of thousands of dollars and man-hours each year on original research conducted through mailed questionnaires and telephone surveys.



2. Reed Elsevier's MDL Information Systems, Inc. MDL is a U.S. based company with more than 330 employees worldwide that creates, produces and distributes databases and computer programs used around the globe by, among others, the pharmaceutical and chemical industries, as well as by government and education organizations involved in basic scientific research. For example, MDL produces a range of databases that, taken together, offer chemists an electronic library for ready access to data selectively obtained from more than 400 chemistry journals, 250 chemical supplier catalogs, pharmaceutical patents, company and regulatory reports, as well as conference proceedings. These databases cover chemical suppliers and pricing, handling and safety information for 100,000 chemical products, organic chemistry syntheses and preparative methods, xenobiotic transformations and compounds, and structure and biological activity data for 70,000 drugs.



3. The Thomson Corporation's POISINDEX. This invaluable database provides medical professionals--usually an emergency room physician or poison control specialist--with immediate access to comprehensive listings of toxicological information--a crucial tool to complement their years of experience and training. Authorized users have unlimited access to this information at their own facilities 24 hours a day, 365 days a year. POISINDEX enables them, for example, to identify a substance that a child may have ingested and then to provide instructions for critical, immediate care. Treatments guided by this specialized database have helped save thousands of lives since POISINDEX was created over twenty-three years ago.



POISINDEX contains about 1,000,000 entries describing substances such as drugs, chemicals, commercial and household products and biological materials. More than 30 professionals with training in nursing, pharmacy, toxicology and medicine are responsible for reviewing these substances and obtaining pertinent information on them. In addition, more than 200 practicing clinicians from over 20 countries participate in the editorial process as members of the POISINDEX editorial board. The database lists each substance and up to four full-text documents detailing its clinical effects, treatment measures, degree of toxicity and other relevant information. Software engineers develop computer software to store, edit, sort and retrieve the data and to maintain, test, produce and support the database.



4. Skinder-Strauss Associates' Lawyers Diary and Manual. Attorneys in New Jersey, New York, Massachusetts, Florida and New Hampshire routinely use the Lawyers Diary, or Red Book, as their daily reference and directory for information regarding courts, judges, government agencies, and the members of the bar. Practicing lawyers rely upon its comprehensive and accurate databases to assist them with their day-to-day communications, and many regard the Red Book as their most essential source for this needed information. A third-generation, family-owned business, Skinder-Strauss has more than 40 full-time employees who are actively engaged in the daily activities of data collection, verification, editorial compilation, research and data entry. The various databases managed by the company require contact with more than 400,000 individuals and entities at least once a year. All contact and verification research is initiated by the company through extensive direct mail, telemarketing and other proactive efforts. These initiatives involve the expenditure of significant sums, thousands of man-hours and the pride and dedication of those so engaged.



5. The American Medical Association's (AMA) Physician Masterfile. This comprehensive database contains information regarding approximately 800,000 physicians, including both AMA members and non-members. Its physician demographic data--including state medical licensing and educational information--helps protect the public from fraud and abuse by enabling the ready confirmation of the credentials of those holding themselves out as physicians. The Physician Masterfile's unique physician identifiers allow many industries to bring up-to-date information to physicians regarding the availability of new drugs and their side effects, and to protect the public in the event of drug recalls by the Food and Drug Administration.



6. Phillips Business Information, Inc. provides a broad range of information products for distinct business markets, including more than 35 directory and directory-related products. For example, Phillips' EDI Yellow Pages and Electronic Commerce Directory, formerly Who's Who in Electronic Commerce, is just the type of informational product that, given the ever-increasing role of electronic commerce, is vital to our maintaining our leadership in the global marketplace. It links the reader to more than 34,000 potential Electronic Commerce ("EC") and Electronic Data Interchange ("EDI") business partners. This year's edition will show a user how to implement electronic commerce into their business; select the right order processing payment software for their business; and secure their business information from outside interference. Sections in the directory include: organizations active in EC and EDI; business partners by industry; market data; value added networks; software providers; value-added banks; business and technical services; bar coding equipment providers; associations and user groups; standards organizations; and more. The directories will keep the reader up-to-date with a competitive and constantly changing industry and give access to key decision makers who move within companies and from company to company. Compiling this directory requires two full-time editors to gather, inspect, and update more than 34,000 names, addresses and telephone numbers, plus independent contractors to assist with programming.



7. Thomas Publishing Co., located in New York City, has published industry information products for a century. Its 400 employees, with a payroll of more than $21,000,000 per year, publish 24 major buying guides, 29 product news magazines, two product information exchange services, a magazine on factory automation, three software comparison guides, and a publication to help buyers select the most cost-efficient transportation modes for their inbound freight. Its Register of American Manufacturers compiles purchasing information about 155,000 companies, classified under more than 60,000 product and service headings.



The Thomas directories are primarily supported by advertising. In that connection, independent entities throughout the United States solicit advertising orders, provide advertising related material, as well as editorial information to the company. Those organizations are paid in excess of $50,000,000 for their services.





Vulnerability of Databases.



While creating, verifying, maintaining and disseminating databases is expensive and time-consuming, copying and distributing databases without permission is cheap and easy.



Today, database pirates can use widely available technologies to copy or print electronic databases and distribute them around the world. The advent of digital, high-speed computer networks adds greatly to this threat of piracy. Internet users can copy and distribute large collections of information with the click of a mouse and at a fraction of the enormous costs required to develop these products. These risks will only increase as our society becomes more dependent on computers and digitized information, and as technologies provide new and even more efficient ways to copy and distribute informational products.



Without effective legal protection, databases are easy prey for parasitic competitors who are free to take the fruits of the creator's hard work to produce competing products. These risks are not limited to competitors' market-destructive acts. For example, LaMacchia v. United States, 871 F. Supp. 535 (D. Mass. 1994), demonstrates that non-competitors can engage in activities that inflict serious commercial harm on publishers. In LaMacchia, an MIT student uploaded commercial software (such as WordPerfect and Microsoft Excel) onto an electronic bulletin board. 871 F. Supp. at 536. He encouraged others to download these applications free of charge. Id. Although unmotivated by any desire for pecuniary gain, his actions cost the affected software developers over $1 million in losses. Id. at 537. The indictment against LaMacchia was dismissed because he acted without the commercial motive required in cases of criminal copyright infringement.



Regrettably, data pirates of all stripes have little to fear because existing U.S. law does not effectively deter such blatantly unfair practices. It is time for Congress to fill this gap in U.S. law.





The Gap In U.S. Law.



Although existing legal doctrines--including copyright, contract, and misappropriation law--all offer important protections, they are insufficient, particularly in today's digital world, to deter database piracy effectively. Even some of those most skeptical of past domestic and international database protection proposals acknowledge that current law leaves organizations that market their products to the public "vulnerable to market-destructive appropriations" and that "this state of chronic underprotection tends to keep the production of information goods at suboptimal levels." J.H. Reichman and Pamela Samuelson, Intellectual Property Rights in Data?, 50 Vand. L. Rev. 51, 54, 55 (1997). New legal protection must be added to U.S. law to complement existing doctrines so that database creators will have the incentive to continue making the enormous expenditures necessary to produce, update and market informative and innovative databases.





Copyright Law



For many years, database makers could take solace in the fact that some federal courts of appeals recognized the so-called "sweat of the brow" doctrine, under which copyright protection was based on the compiler's significant hard work and investment in developing its compilation. In those circuits, "sweat of the brow" afforded compilers an important tool against the unauthorized takings of "free riders." That is no longer the case.



In its landmark decision, Feist v. Rural Telephone Co., Inc., 499 U.S. 340 (1991), the Supreme Court discarded the sweat of the brow approach under copyright law and made it clear that a compilation will enjoy copyright protection only if it evinces sufficient "originality" in the manner in which its facts are arranged, selected or coordinated. After Feist, the amount of time, effort and money expended by a compiler is irrelevant to a determination of whether or not a work qualifies for copyright protection. Today, copyright protection for compilations turns on whether or not they satisfy the originality test--"that the work was independently created by the author (as opposed to copied from other works) and that it possesses at least some minimal degree of creativity." 499 U.S. at 345.



Feist did much more than retire the sweat of the brow doctrine and clarify that "originality" formed the linchpin of copyright protection in compilations. It also noted both that facts were not copyrightable and even where protection exists for compilations, its scope is thin because it extends only to the original selection, arrangement and coordination of the database. The message given to the database community by Feist was clear: The factual contents of the database are not protected by copyright; they may be copied with impunity.



Notwithstanding a valid copyright, a subsequent compiler remains free to use the facts contained in another's publication to aid in preparing a competing work, so long as the competing work does not feature the same selection and arrangement. As one commentator explains it: "[N]o matter how much original authorship the work displays, the facts and ideas it exposes are free for the taking... . [T]he very same facts and ideas may be divorced from the context imposed by the author, and restated or reshuffled by second comers... . [without infringing the first compiler's copyright]. 499 U.S. at 349. [Internal citations omitted.]



As one commentator succinctly put it, "[u]nder the Supreme Court's analysis [in Feist], a competitor would be infringing no copyright if it simply stole the data and left the base." J. Litman, After Feist, 17 Dayton L. Rev. 607, 609 (1992).



Lower court interpretations of Feist have caused additional reasons for consternation in the database community.



First, inconsistent decisions have caused database owners to have to guess at whether a federal court will afford a particular compilation any copyright protection at all. Initially, some database creators thought that they could find solace in Feist's statement that "the requisite level of creativity is extremely low; even a slight amount will suffice. The vast majority of works will make the grade quite easily. . . ." 499 U.S. at 345. Unfortunately, that has not always been the case. For example, two federal appellate courts have reached profoundly different results with respect to whether yellow page directories contain the necessary degree of originality to warrant copyright protection. Compare Key Publications, Inc. v. Chinatown Today Publishing Enter., Inc., 945 F.2d 509 (2d Cir. 1991) (copyright protection held to exist) with Bellsouth Adver. & Pub. Corp. v. Donnelley Info. Pub., Inc., 999 F.2d 1436 (11th Cir.) (en banc) (coming to the opposite conclusion), cert. denied, 114 U.S. 943 (1994).



More recently, the United States Court of Appeals for the Eleventh Circuit set off alarm bells in the database community when it ruled against Warren Publishing, a CADP member, in Warren Publishing, Inc. v. Microdos Data Corp., 115 F.3d 1509 (11th Cir. 1997). Despite the fact that the selection of cable systems in Warren Publishing's Factbook was created using a unique and original definition of "cable system" that resulted in fewer listings for a given region than would a selection using the standard definition of "cable system" used by the Federal Communications Commission, the court refused copyright protection. According to the Court, Warren "did not exercise any creativity or judgment in 'selecting' cable systems to include in its Factbook, but rather included the entire relevant universe known to it." 115 F.3d at 1518. This decision raises grave concerns that the level of originality required for copyright protection may be far higher than was reasonably believed to be the case given the Supreme Court's language in Feist.



Second, post-Feist cases underscore the fact that U.S. copyright law does not today provide meaningful coverage even to those compilations that meet an initial test of copyrightability. Courts have consistently reiterated Feist's conclusion about the thinness of the copyright protection for factual compilations, making absolutely clear that in no event does copyright law inhibit a defendant's ability to exploit the contents of a database.(3)

As the Copyright Office Report states, "most of the post-Feist appellate cases have found wholesale takings from copyrightable compilations to be non-infringing. The trend is carrying through in district courts as well."(4)



Third, post-Feist cases give short shrift to two key characteristics typical of many valuable databases--their thoroughness and the human and financial resources expended in creating and marketing them.



The greatest irony of all is that the more thorough the database, the more time, money and effort that goes into making it--and the more valuable it may be to a user--the more likely it is that a court will find that it lacks the requisite degree of originality to qualify for copyright protection.(5)

This result is inconsistent with sound public policy.



"A database of meteorological, environmental, or medical information, for example, must be comprehensive, accurate, and up-to-date, or the results could be injurious to health or safety.... Subjective selection or a unique arrangement may impede the database's utility or ease of access."(6)



In sum, after Feist and the demise of the sweat-of-the-brow doctrine, it has become increasingly clear that the copyright law is ill-equipped to protect informational products that are the result of substantial human, technical and financial resources.





Contracts



Although private contracts are very valuable in protecting the works of database creators, they do not provide protection at a level sufficient to induce the creation and distribution of databases of the diversity that are increasingly in demand today. The most obvious infirmity of contract law is its general inability to provide legal relief against malfeasors who have not entered into a binding contract with a person seeking to prevent piracy of database material. A specific undertaking, on the part of X, not to reproduce or distribute database material created or distributed by Y is, of course, valuable to Y but in no way protects Y in the event that Z, without authorization, obtains access to the database material and then reproduces it and distributes it for profit.(7)



Another problem with contract law is that while its contours are roughly equivalent across the 50 states, there are circumstances under which the contract laws of two sister states may provide different results if applied to the same legal problem.(8)

Current efforts by many parties to draft Article 2B of the Uniform Commercial Code demonstrate the widely felt desire for uniformity in information contract law. It should be noted, however, that the history of the entire UCC is marked with the exercise of state legislatures' power to enact "special" provisions in respect of their individual states, thus rendering the UCC less than perfectly "uniform."(9)



Additionally, while the Seventh Circuit spoke forcefully--and correctly, we believe--in ProCD v. Zeidenberg(10)

when it held that contracts are not routinely preempted by the copyright law, the case represents the law of but one of the federal appellate circuits. As such, its instructions need not necessarily be followed elsewhere in the United States, and the extent to which contracts are preempted remains arguably open elsewhere.



Finally, the global nature of electronic commerce makes state law solutions of less value than in a purely domestic context. Nations--not their constituent members--will shape the international regime that will determine how trans-border transactions are treated.





Misappropriation



The common law tort of "misappropriation," derived from the Supreme Court's decision in International News Service v. Associated Press, 248 U.S. 215 (1918), has had a checkered history, at best, with respect to protecting copyrightable and uncopyrightable works from behavior that might fall under the general rubric of "copying." Although both Congress and the Supreme Court have taken some pains to argue that the tort continues to exist,(11)

in point of fact it may be available only in the narrowest and most special of circumstances, particularly in light of the Second Circuit's recent decision in National Basketball Association v. Motorola, Inc., 105 F.3d 841 (2d Cir. 1997). That case, while characterized by the Copyright Office as providing "renewed clarity and authority"(12)

to the doctrine of misappropriation, held that the National Basketball Association was not entitled to relief against an enterprise providing "real-time" score services via pagers and the Internet. Indeed, the "modern" doctrine under the NBA case appears to require, as an element of the tort, that both "the information [be] time-sensitive, [and] the defendant [be] in direct competition with a product or service offered by the plaintiff. . . ."(13)

As a consequence, its value to database creators is quite limited, given that the contents of many databases do not amount to "hot news," but instead provide long-term reference points with respect to a particular industry or phenomenon. Likewise, as we have noted earlier in this testimony, it is quite possible for someone other than a traditional "competitor" to inflict grievous harm upon a database publisher while not acting in the clothes of competition.



State misappropriation doctrine is even less uniform than state contract law. Nothing akin to Uniform Commercial Code drives the several states toward uniformity, and, as a result, it is almost impossible for a national firm to base investment decisions upon its best guess of whether and where its rights against piracy may be enforced to a particular extent.





The Time is Now



Mr. Chairman, the time for congressional action is now. Without appropriate legislative relief, the accumulated effects of domestically sanctioned piracy will cause new firms to refuse to enter the database market, some current firms to leave it, the loss of thousands of American jobs and the end of our worldwide preeminence in this area. Ultimately, everyone loses as the availability of invaluable information products to the public decreases.



The dawn of the Information Age has begun radically to change the way people do business. In "the old days," commercial customs developed over appreciably longer periods of time. If a user ordered a compilation from one of our members, for example, it is very likely that papers would be exchanged and goods would be shipped according to terms which both parties understood from decades of trade usage.



Imagine now that same transaction occurring at the speed of light as contract offers, acceptance and performance occur not through the mails, but over fiber-optic networks. Commercial practice--whether scrupulous or not--develops at a pace geometrically greater than that of just a decade ago. By the same token, the destructive effects of piracy that we see right now will become much, much worse in a short period of time as the holes in our current law become more and more apparent to database pirates.



Mr. Chairman, the preeminence of the U.S. in the area of database production is seriously at risk. As Congress recognized when it passed the Semiconductor Chip Protection Act, "a finding that an industry has done well in the past without legislative protection does not mean that threats to present and future investments fall outside Congressional concern."(14)

Congress has no obligation to wait until the harmful ripples created by the gaps in current law become a tidal wave. As the Supreme Court recently noted in Turner Broadcasting Sys., Inc. v. Federal Communications Comm., 117 S. Ct. 1174, 1997 U.S. LEXIS at *58 (1997), "an industry need not be in its death throes before Congress may act to protect it."



The handwriting is on the wall. Before long, this legal weakness will cause irreparable harm to the database marketplace. Creating floppy disks (or, for that matter, CD-ROMs) requires little or no overhead when compared to the cost of publishing and distributing a printed volume or assembling the data in the first instance. In Warren, for example, the Eleventh Circuit held that the copyright law allowed the defendant to appropriate the entire contents of the Factbook, from which it then made competing product. Similarly, in the ProCD case, the defendant loaded the database onto the Internet, from whence it could be downloaded by anyone with the desire to do so. In LaMacchia, the harm caused to the software owners from one pirate exceeded $1,000,000. Similarly, anyone with a scanner and a web page poses a palpable threat to a database maker. Our current legal regime does not effectively deter such piracy, and fact patterns like those in Warren will proliferate unless Congress intervenes.



The harm created by database piracy does not fall on the shoulders of producers alone; it inures to the detriment of everyone. First, scientific and academic research will be curtailed. In the current database market, many producers charge a much lower access fee (if any) to non-profit institutions such as universities, and recoup those losses in their sales to commercial entities.(15)

Price differentiation makes economic sense, however, only if the for-profit market is secure and those who can acquire the database cheaply do not provide it to those who would otherwise have paid its original developer a higher price. As the cost of piracy becomes a greater part of doing business, this tiered pricing structure will level out, forcing database owners to charge colleges and libraries the same prices they ask of for-profit-corporations. Protective legislation will preserve current pricing flexibility, to the benefit of database owners and users alike.



Second, as free-riders, database pirates--who have expended a fraction of the resources invested by the original compiler--cannot be expected to spend the monies necessary to update the contents they appropriated. The consequences of this failure to keep the data current could prove devastating, particularly in cases involving health, safety or environmental data. In contrast, legislative protection for these collections of information maintains an economic incentive for compilers to keep their products accurate, current, and comprehensive.



Without remedial legislation, international developments, beginning January 1, 1998, will compound the problem. On that date, the European Union Directive obligates member nations to have their own database protection laws in place. The Directive protects only databases made by nationals and habitual residents of member states, companies located in nations that the EU decides have comparable forms of database protection, companies formed under the laws of a member state and centrally administered from the EU, and companies that have a genuine economic link to an EU member state. If the Congress does not act promptly, existing and future databases created in this country will be free for the taking in Europe and our position as a world leader in this area will be compromised, perhaps fatally.(16)

The time for congressional action is now.





Key Components of a Database Protection Bill



Mr. Chairman, in May of this year, CADP submitted to the Copyright Office a document entitled "Key Components of a Fair Database Protection Bill." At that time, CADP set forth its interest in the adoption of legislation to deter database piracy that causes commercial harm, but does not result in adverse consequences for scientists, educators, news gatherers, and other database users. We also urged that any forthcoming legislation be drafted in a manner maximizing the likelihood that the European Union will find that the U.S. database law conforms to the Directive's norms--thereby ensuring that U.S. databases are protected in the European Union. The document contained a discussion of essential elements that should be included in any database protection bill.



CADP members have reviewed H.R. 2652. At the outset, we must acknowledge some initial uneasiness and uncertainty over the dramatic shift that H.R. 2652 represents from the sui generis approach contained in (a) last year's bill, H.R. 3531, (b) the EU Directive, and (c) last year's draft World Intellectual Property Organization ("WIPO") treaty. As we told the Copyright Office, our preference would be for Congress to use H.R. 3531 as its starting point, retaining its sui generis approach, but modifying that proposal to meet legitimate concerns and questions raised regarding that bill. Nonetheless, our members believe that H.R. 2652 addresses many of the points set forth in our May submission to the Copyright Office. H.R. 2652 provides a solid basis for fair and effective legislation. We are committed to working with the Subcommittee and other interested parties in perfecting the pending proposal.



We would like to share with the Subcommittee a number of specific comments with respect to H.R. 2652.



Overall Thrust of H.R. 2652. We agree with the overall purpose of H.R. 2652: to protect investment in the production and distribution of valuable collections of information by prohibiting misappropriations that cause commercial harm to their actual or potential markets without creating unintended consequences for legitimate uses by news reporters, educators, scientists, librarians, consumers and other users.



Prohibition of Misappropriations. The heart of the bill is found in §1201--"Prohibition Against Misappropriations."



Any person who extracts, or uses in commerce, all or a substantial part of a collection of information gathered, organized or maintained by another person through the investment of substantial monetary or other resources, so as to harm that other person's actual or potential market for a product or service that incorporates that collection of information and is offered by that other person in commerce, shall be liable to that person for the remedies set forth in section 1206.



While we believe that this key language will generally achieve its purpose, we have some concerns.



For example, does the use of the undefined term "substantial" in this section refer both to qualitative and quantitative parts of a collection? The bill is silent on this important point. As a result, an ambiguity exists as to whether an extraction or use of a quantitatively insubstantial, but qualitatively substantial, part of a collection of information is actionable, where such extraction or use creates harm. Because a use or extraction of a relatively small--but crucial--part of a collection can cause real harm to the collection maker's market, we urge the Subcommittee to amend the bill to cover such situations. More specifically, we suggest that the relevant language in §1201 be amended to read as follows:



"... all or a substantial part--qualitatively or quantitatively--of a collection ..."



Non-competitive Harm. As this Subcommittee is well aware from its consideration of H.R. 2265, the "No Electronic Theft Act," commercial harm can be caused by competitors and non-competitors alike, and one need not have a commercial motive to inflict such market damage. H.R. 2652 recognizes this crucial point. While the bill contains important, specific provisions with respect to not-for-profit uses and news reporting (§§ 1202 (d) and (e)), it quite properly focuses on the harm to the actual or potential market for a qualifying collection of information, rather than on the status of the offending actor.



Governmental Databases. We agree emphatically with representatives of the educational and scientific community that any forthcoming legislation should not protect government-compiled databases. Consistent with H.R. 3531, § 1203(a) of H.R. 2652 would achieve this important goal by expressly providing that no governmental entity--federal, state or local--can claim protection for its databases. Thus, databases compiled at taxpayer expense would not be protected under H.R. 2652. Section 1203(a) also expressly provides that protection under the bill would not extend to a private party who enters into an exclusive contract with a government entity with respect to government data. In sum, H.R. 2652 does not inhibit access to government information and databases by value-adding entrepreneurs, consumers, scientific educators, librarians and others. CADP strongly supports these portions of the pending bill.



Complementary Nature of Protection. We believe that it is essential that the protections afforded under the bill in no way diminish protections or restrictions that may be available under other legal regimes such as copyright and contract laws. Although it is absolutely imperative that legislation be enacted to fill the current gap in protection for databases, we agree with you, Mr. Chairman, that copyright and contract law "remain essential for protecting the enormous investments in collection[s] of information. ..." Section 1205 appears to ensure that result.



News Reporting. Unlike H.R. 3531, H.R. 2652 contains a provision addressing the status of "news reporting," but leaves that term undefined. We understand that this provision was included in response to concerns raised by press groups regarding the potential adverse impact of a database protection bill on their news gathering and reporting activities. These are important concerns and many of our members, in fact, engage in news gathering and reporting.



The Subcommittee faces a delicate task here. It must respect fundamental First Amendment rights, and should permit news gatherers to report discrete facts such as a few stock quotes or baseball scores without running afoul of the law. It must, however, also recognize that too broad an exemption for news reporting activities could lead to condoning activities that stray beyond traditional reporting and cause commercial harm to the owner of a collection of information.



As now drafted, § 1202(c) appears to fully exempt from liability those who extract or use "information for the sole purpose of news reporting," irrespective of the harm caused by such actions. We believe that the apparently blanket exemption for "news reporting" warrants further scrutiny.



Individual Items of Information and Other Insubstantial Parts. In its submission to the Copyright Office, CADP urged that the legislation specify (l) that it does not protect individual facts and (2) that uses of insubstantial parts of a collection--qualitatively or quantitatively speaking--are not infringing unless such acts occur repeatedly or systematically in a manner that causes commercial harm. It appears that § 1202(a) is an attempt to address these points. As currently drafted, however, we believe that this provision is somewhat confusing and indirect. The provision must hammer home these two important points more affirmatively.



Independent creation of databases permitted. Consistent with H.R. 3531, § 1202(b) makes it explicit that anyone is free independently to gather information as long as he does not extract it from a collection of information in which an original compiler has made a substantial investment of monetary or other resources. CADP supports this provision.



In our view, the purpose of the law is to prohibit free-riding; not to stop independent gathering of information from sources other than a collection of information that qualifies for protection under the Act. Second compilers should be free to do the work themselves and create competing collections of information; they should not be free to exploit the hard work of others. In short, our members welcome fair competition in the marketplace.



Verification. Section 1202(c) has given some of our members pause. They are concerned lest the verification privilege swallow the rule of independent creation, i.e., that someone might choose to shortcut independent creation by copying large portions of another's database and thereafter vigorously assert that he was not liable under this law by virtue of having simply used the original database for purposes of "verification." We would prefer that the bill itself, or at least its accompanying report, stress that the burden of proving use for "verification" purposes should always lie--and remain with--the copyist, for whom this narrow privilege is intended. In our view, an appropriately crafted verification provision can be helpful to owners of collections of information, as well as users, particularly scientists.



Jurisdiction and Venue. As drafted, § 1206(a)--"Civil Actions"-- appears neither to provide exclusive federal jurisdiction with respect to litigations in which rights under this new Chapter 12 of Title 17 are asserted nor to track the copyright, mask works, and other laws with respect to venue. It is clear from the last clause of Section 1206(a) that actions against state governmental entities may be brought in state court, but no language in the bill provides otherwise with respect to other defendants. It bears emphasis that there is nothing in the copyright law that makes federal jurisdiction in copyright litigations exclusive. Such exclusivity comes from the Judicial Code. The second sentence of 28 U.S.C. § 1338(a) provides that federal jurisdiction in the case of "patent, plant variety protection and copyright cases" is exclusive, and 28 U.S.C. § 1338(c) makes exclusivity clear with respect to mask works. Should it be the Subcommittee's intention--as we believe it is--to provide exclusive federal jurisdiction in respect of collection of information piracy cases except with respect to defendants who are state entities, then it appears that Section 1338 of the Judicial Code should be so modified.



With respect to venue, the phrase used in § 1206(a) "an appropriate United States district court" is not typically used to provide clear guidance with respect to where a particular litigation may be brought. Of course, 28 U.S.C. § 1391(b)--the "general" venue portion of the Judicial Code--would provide rules, but with respect to copyrights, mask works, and patents, "special" venue provisions are now set out in 28 U.S.C. § 1400. It might well be useful to create similar venue provisions with respect to actions arising under the proposed provisions of Chapter 12 of Title 17.



Federal Infringement. Finally, 28 U.S.C. § 1498 provides detailed rules with respect to the bringing of patent, copyright, and mask work infringement actions against the United States in the United States Court of Federal Claims. It would seem appropriate similarly to limit the United States' ability to assert the defense of sovereign immunity by amending 28 U.S.C. § 1498 with respect to "Chapter 12" litigations.





Congressional Power



Mr. Chairman, we believe a review of the text of the Constitution and relevant case law reveals ample authority under the Commerce Clause (Art. 1, §3, cl. 3) to support congressional enactment of legislation such as H.R. 2652. In the past, Congress has used its Commerce Clause power in virtually every area of federal legislation ranging from civil rights to environmental to trademark legislation. Applying this power in the current context is entirely consistent with past congressional exercises of that authority. For our purposes, the constitutional history of our trademark laws provides the most apt illustration of Congress's exercise of its Art.I, §3, cl 3 power.(17)



Although it is now firmly established that the Commerce Clause forms the constitutional source of power undergirding our trademark law, that was not always the case. In fact, the first federal trademark law was voided by the United States Supreme Court in its landmark 1879 decision, the Trade-Mark Cases, 100 U.S. 82. According to the Court, the first trademark law unconstitutionally premised trademark protection on the Patent/Copyright Clause of the Constitution because trademarks were neither "discoveries nor writings" as required by Art. 1, § 8, cl. 8.(18)

Significantly, all subsequent federal trademark laws have been premised on the Commerce Clause.



In our opinion, enactment of legislation such as H.R. 2652 would be an appropriate exercise of Congress' power under the Commerce Clause. Trademarks are indicia of origin that regularly travel in and affect interstate commerce. It seems equally clear that collections of information are items of commerce and it is beyond debate that the U.S. database community provides a wealth of these informational products to users both here and abroad. Just as Congress has chosen to protect trademarks under the Commerce Clause, we believe that it has the power to protect valuable compilations that are the product of substantial effort and money from harmful misappropriations.





Conclusion



Mr. Chairman, American database producers need legislation to protect them, and we need it now. Our goal is not to "lock up" data, or prevent access to information; it is to protect our work product from the commercial harm caused by free riders. Your bill represents a welcome and important first step towards accomplishing that goal. Thank you for giving us the opportunity to share our views with you.

Judiciary Homepage

1. 0 The Directive embodies a two-tiered approach to database protection. First, it requires compilations to meet an "intellectual creation" standard in order to receive copyright protection. Second, it creates sui generis protection for databases, prohibiting the unauthorized extraction of substantial aspects of a database produced as a result of substantial investment.

2. 0 It appears that the only other option for gaining protection in the EU for non-EU databases, under the somewhat ambiguous language of the Directive, is for their producers to relocate to Europe.

3. 0 For example, in Skinder Strauss v. MCLE, 914 F. Supp. 665 (D. Mass. 1995), a district court's summary judgment ruling that even if the defendant's competing legal directory infringed plaintiff's compilation as a whole, such infringement would not inhibit defendant's ability to exploit the contents of each individual database that comprises plaintiff's extensive compilation.

4. 0 Copyright Office Report at 17-18.

5. 0 Cf. Warren Publishing, Inc. v. Microdos Data Corp., 115 F.3d 1509, 1518 (11th Cir. 1997) (stating that by selecting the entire "relevant universe" known to it, Warren made its directory commercially useful and therefore forfeited the protection of the Copyright Act).

6. 0 Copyright Office Report at 75.

7. 0 As scholars have noted, "Even contract law has significant limitations when mass-marketed information products are sold to persons not in privity with the makers." Reichman & Samuelson, 50 Vanderbilt L. Rev. at 137.

8. 0 Compare Bussard v. College of St. Thomas, 200 N.W.2d 155 (Minn. 1972) (excluding evidence of prior negotiations when determining whether an agreement is integrated) with Masterson v. Sine, 456 P.2d 561 (Cal. 1968) (looking to all relevant circumstances including prior negotiations to determine whether an agreement is integrated).

9. 0 We are aware of those who argue that the anticipated adoption of the UCC-2B by the various states reduces the need for database protection legislation. But, this argument fails to take into account that contracts conforming to UCC-2B suffer from the same infirmity as other contracts or licenses--the inability to bind non-parties to the agreement.

10. 0 86 F.3d 1447 (7th Cir. 1996). In ProCD, the defendant bought a CD-ROM copy of ProCD's phone listing database in a retail outlet. 86 F.3d at 1450. The CD-ROM (as well as its packaging) contained a license, the terms of which appeared on the user's screen each time the software ran. The license precluded unauthorized distribution of the database contents. Nonetheless, the defendant pirated the data and placed it on the Internet. Id. The court, through Judge Easterbrook, found that the acts of manifesting assent by opening the package and repeatedly running the software formed a valid contract, which Zeidenberg had breached through his acts of unauthorized copying and distribution. Id. at 1452. The court also held that the contract was not preempted by the Copyright Act. Id. at 1455.

11. 0 H.R. Rep. No. 1476, 94th Cong., 2d Sess. 132 (1976); Feist, 499 U.S. at 354.

12. 0 Copyright Office Report at 83.

13. 0 105 F.3d at 845.

14. 0 The House Report on the Semiconductor Chip Act, H.R. Rep. No. 487, 92d Cong., 1st Sess. 3 n.5 (1984). The position of database producers bears a striking similarity to that of semiconductor chip manufacturers. Cf. id. at 2-3 (describing how the cost of duplicating a chip design runs at less than one thousandth of the cost of original development); id. at 4 (describing how other bodies of law, such as patent and copyright, could not protect chip design).

15. 0 Tyson Report, supra, at 15.

16. 0 Cf. H.R. Rep. No. 487. supra, at 3 (describing with approval the U.S. semiconductor chip industry's position as a world leader, as well as the industry's stress on innovation and development-friendly pricing structures).

17. 0 Congress has alluded previously to its Commerce Clause authority when enacting intellectual property legislation. For example, Congress referred to the Commerce Clause as an alternative source of authority when it enacted the Semiconductor Chip Act of 1984. At that time, Congress chose to rely on both the commerce clause and Patent/Copyright Clause because of the possibility, albeit an unlikely one, that a mask work could be found neither a writing nor a discovery as required by the Constitution (Art. 1, §8, cl. 8). See Nimmer on Copyright at §1.09.

18. 0 In the Trade-Mark Cases, the Court also made clear that Congress' Commerce Clause power could not be used to sustain the law in question because it was not restricted to marks in interstate or foreign commerce. This portion of the Court's opinion is of questionable validity today given the expansive evolution of the Commerce Clause jurisprudence over the past century.