The Seybold Report on Internet Publishing

December 1998


Publications Home

Full-text search

Customer service




Site Map

Browse back issues





Up Front

U.S. Digital Copyright Act Gets the Nod from the President

By Matt McKenzie

Is the Digital Millenium Copyright Act a well-crafted compromise, or a threat to legitimate research? The law, passed in October, sets the rules for enforcing the international copyright treaties signed two years ago. But it left unresolved the debate over stronger enforcement of commercial databases.

On Oct. 28, President Clinton signed the Digital Millennium Copyright Act (DMCA) into U.S. law. The final version of the DMCA (, which bans the production, sale and, in most cases, use of devices intended to thwart digital copyright protection technology, is the product of a yearlong battle between publishers and a loose alliance of civil rights groups, academics, scientists and librarians. Both the publishing industry and some former opponents have described the law as a compromise, protecting both the interests of copyright owners and the rights of individual consumers.

Yet whether the law lives up to its goal of benefiting both copyright holders and end users, and whether it really settles the debate over digital copyright, are questions that don't have any clear answers. Groups that opposed earlier versions of the law, such as the Electronic Frontier Foundation (EFF) and the Association for Computing Machinery (ACM), have grudgingly accepted the DMCA in its current form, although they still worry that it could erode the concept of "fair use" in cyberspace and hinder legitimate research. And some of the law's most vocal supporters, such as the Software Publishers Association and the Association of American Publishers, have already vowed to push for a law granting new copyright protections to databases—a controversial provision that didn't make it into the final version of the DMCA but that some major players in the publishing industry desperately want passed (see sidebar).

Read a book, go to jail. The DMCA, which brings U.S. law into line with World Intellectual Property Organization treaties signed in 1996, doesn't actually tinker with the basic principles of copyright law. Instead, the law aims to safeguard the technologies, such as encryption or copy-protection systems, that protect and manage digital content. Specifically, it outlaws attempts to circumvent or disable these systems, and it also bans devices whose primary purpose is to defeat copyright protection technology. The DMCA would make it illegal, for example, to tamper with the watermark on a digital image, or to remove encrypted digital "wrappers" that control access to electronic documents and software.

In most cases, courts will treat people who break the law as civil offenders and impose fines up to $2,500 per violation. The DMCA also includes criminal penalties, however, subjecting big-time violators to penalties as severe as a $1 million fine and ten years in prison—a hefty price to pay for lifting someone else's bytes.

In its earlier incarnations, the DMCA laid down what would have been a blanket ban on a wide range of technology and activities, including many that are currently legal and widely accepted. Most network security and anti-virus research would have become illegal, for example, since these projects often require researchers to reverse-engineer computer software or to circumvent security systems—activities that the DMCA would have outlawed. Similar bans would have applied to some forms of encryption research, compatibility testing and many other common research activities.

Unintended consequences? The sheer breadth of the law stunned scientists, academics and researchers, many of whom literally saw their livelihoods going down the drain. In a letter to the U.S. Congress sent earlier this summer, 50 of the nation's leading computer scientists and technology industry executives stated that the law would "imperil computer systems and networks throughout the United States, criminalize many current university courses . . . and severely disrupt a growing American industry in information security technology."

The law aims to safeguard the technologies, such as encryption or copy-protection systems, that protect and manage digital content.

(To read the letter, see .)

The American Library Association also cried foul, noting that while the DMCA wouldn't change existing fair use doctrine per se, it would effectively gut fair use by allowing publishers to lock up their content and then prosecute anyone who tried to peek inside. Other critics pointed out that publishers had frequently vilified manufacturers of legitimate consumer devices, such as videotape recorders and cassette decks, for encouraging piracy, and that the DMCA would only encourage efforts to stifle new technology. One of the law's most prominent opponents, ACM president Barbara Simons, noted that had the VCR been invented after the DMCA became law, it probably wouldn't have passed legal muster. "How could one have proven that VCRs have other [legitmate] uses," Simons wrote, "when some of those uses were not yet known?"

A better deal

The final version of the DMCA was considerably less draconian, owing to a number of last-minute exceptions. The DMCA now explicitly allows encryption and security research, and it will remain legal to reverse-engineer software for compatibility reasons or to test security or anti-virus technology. The law also allows individuals to circumvent copyright protection in order to protect personal information—such as the data that many Web sites collect with cookies or other tracking technology. Additional exceptions apply to libraries, archives and academic researchers, allowing them to circumvent copyright protection in order to evaluate and create backup copies of digital documents. The DMCA also spells out "safe harbor" rules that protect ISPs and network providers from prosecution—as long as they take action once they're notified that something illegal is afoot on their systems.

One of the law's unique provisions is an explicit statement that it won't supersede existing fair use principles, such as the consumer's right to resell, loan or give away a copyrighted work once he's paid for it or the right to excerpt copyrighted works for research and criticism.

To back up this promise, the DMCA does two things. First, it won't take effect for 24 months, ostensibly to give everyone time to consider the implications of the law; and, second, it authorizes the Librarian of Congress to study the law's effect on fair use and legitimate research and, if necessary, to recommend changes in how the law is defined and enforced.

No more games. The publishing community is eager to describe these changes as a "compromise" with opponents, and most of the groups that lobbied on behalf of the bill claim that they're content with the status quo on fair use exceptions and research activities. This largesse may spring from genuine fairness, but it seems more likely that the bill's supporters threw their opponents a bone in order to get what they really wanted: a strong, versatile legal weapon in the war against intellectual property theft.

In any case, publishers are certainly happy with the outcome: AAP president Pat Schroeder described the DMCA as a "big win for America's creative artists, for America's copyright industries, and for American jobs," and groups such as the Software Publishers' Association ( and the Business Software Alliance have been equally liberal with their praise both for the law and for Congress. Lobbyists for the recording and film industries have also joined in the love-fest, predicting that the DMCA will finally pave the way for artists who want to distribute and sell their works in digital form.

A hollow victory? The law's opponents, including EFF, the ALA and representatives from the academic and scientific communities, no longer talk about the DMCA in apocalyptic terms. Some of them, in fact, seem genuinely satisfied with the outcome. Peter Jaszi, a spokesman for the Digital Future Coalition—an umbrella group that includes both the ALA and EFF, along with 40 other organizations—describes the law as "a substantial victory for both the creators and consumers of intellectual property."

Yet most of these groups remain convinced that the final version of the law is a major setback. Shari Steele, staff attorney for EFF, stated that "none of us [among the groups that opposed the DMCA] are happy. We're all accepting that we got as many concessions as we could." More specifically, Steele said, "EFF is unhappy that the exceptions for the uses of encryption are not broader. The libraries are unhappy that there aren't better exceptions for first publication [an element of fair use doctrine]." And, in a statement released after Congress passed the DMCA, the Electronic Privacy Information Center expressed concern that "the exemptions are too limited and will prevent the development and use of necessary research and security tools."

A vocal opponent. ACM president Barbara Simons has been an active opponent of the DMCA.

Association of Computing Machinery,

During the next 18 months both sides will be closely watching one another—and the Librarian of Congress—for hints about how the DMCA will actually affect public policy.

Barbara Simons is just as pessimistic about the impact that law will have on scientific and academic research. Because the DMCA still requires researchers to make a "good faith effort" to contact copyright holders before circumventing copyright protection systems, she said, they can still use the law to block legitimate research—such as cases when reverse-engineering a product might reveal security flaws that will hurt sales. Even if the law is enforced liberally, Simons added, it will have a chilling effect among researchers who don't want to risk huge fines and prison time. Simons also challenges many of the terms used in the DMCA, some of which "are so vague as to be almost meaningless" and which will encourage publishers to prosecute offenders selectively.

Ironically, Simons, concluded, the law may actually hurt intellectual property holders by sweeping flaws in their protection systems under the rug—that is, until a thief comes along and exploits the weakness. "Computer security is something that everyone should be concerned about," she said. "Especially the owners of large amounts of intellectual property."

What now?

In spite of the crowing among publishers and hand-wringing among academics and civil libertarians, it's not at all clear how the DMCA will affect anyone involved. More copyrighted works may, for example, appear in encrypted form, requiring users to pay before they can access the content and restricting their ability to reuse or redistribute it. On the other hand, existing systems of this type, such as IBM's Cryptolopes, never hit the mainstream because this sort of protection annoys consumers, not because pirates have the means to bust IBM's encryption algorithms. The software industry ran into the same dilemma in the 1980s, toying with and ultimately discarding copy-protection systems because so many users complained about them.

Nor is it clear what impact, if any, the Librarian of Congress will have as a fair use watchdog. The DMCA is disturbingly fuzzy on exactly what kinds of studies are required or what the Library of Congress can do to correct perceived abuses. And Simons is uncertain whether the LOC will ever be able to apply the new rules effectively: "How can you anticipate the effect that your actions will have on future development?"

It's also unlikely that the publishing industry will accept a liberal interpretation of the law—there's too much money at stake, and too much advantage to gain from taking a hard-line position on issues such as how to define fair use or whether a particular piece of hardware really has no legitimate, legal purpose. "There's going to be enormous political pressure applied," Simons said, "and the Library of Congress is going to get caught in the middle."

In fact, we've already had a glimpse of how this process may unfold. There's currently a court fight raging between the Recording Industry Association of America (RIAA) and Diamond Multimedia Systems, involving Diamond's Rio MP3 audio player—a Walkman-like device that can download high-quality (and frequently illegal) audio files directly from the Internet and replay them. This particular case actually deals with the RIAA's claim that another law, the American Home Recording Act, requires Diamond to equip the Rio player with copy protection technology. But the case raises an interesting—and troubling—point: How can any law deal effectively with future technology, when it's impossible to know exactly what kinds of devices will reach the market and how consumers will use them?

Stay tuned. The next phase of the battle over digital copyright won't happen quickly, and it certainly won't be dramatic. Over the next 18 months, both sides will use the law's waiting period to launch public relations offensives, to argue a potentially endless series of hypothetical cases and to watch one another—and the Librarian of Congress—closely for hints about how the DMCA will actually affect public policy.

The ACM's Barbara Simons says that her group, along with other members of the opposition, is eager to consult with the government on how to implement the law without eroding fair use or hindering research and development activities. But the law's supporters, including lobbyists for the film, recording and software industries, are both well financed and well connected in Washington. If the negotiations turn on which side has more lawyers—not a very far-fetched notion—it's easy to see which side will have the advantage.

In fact, it's likely that the battle over these issues could drag on indefinitely. It will take years for the courts to decide how to define "legitimate" versus infringing technology and how to reconcile the new restrictions with established fair use principles that date back more than 200 years. In addition, one of the most controversial elements that Congress dropped from the DMCA—its protection of non-copyrighted material in private databases—will start an even bigger fight when Congress takes the issue up again next year. In this sense, the DMCA is a typical chapter in the story of the battle over intellectual property: a cliffhanger, promising an uncertain outcome and plenty of action to come.

"There's going to be enormous political pressure applied and the Library of Congress is going to get caught in the middle."

—Barbara Simons, ACM President

Still to Come: The Database Debate

When Congress reconvenes in January, another big copyright fight could be brewing. That's because, in spite of all the backslapping among publishers delighted with the Digital Millennium Copyright Act, there's one group that hasn't yet gotten what it wants: database publishers.

Some of the largest commercial databases in the world contain mostly public-domain data. This material, which generally isn't protected by copyright, includes everything from stock market data to sports statistics. The companies that collect, organize and publish these data, including giants such as Lexis-Nexis, West Publishing and Elsevier, have traditionally charged steep premiums for access to their databases—usually in a proprietary format, such as dedicated terminals or dial-up server connections.

Strong language. In recent years, however, the rules of the game have changed. Upstart competitors now gather the same data—sometimes from the databases themselves, sometimes from the original data sources—repackage them and publish them on CD-ROM and over the Web. Publishers that once held lucrative monopolies on the data in their systems are now fighting competitors that drastically undercut their premium prices.

Copyright law has traditionally protected only the presentation and the arrangement of public-domain database information—not the data themselves. One recent court decision, for example, held that phone directories are non-copyrightable because they don't show any "creativity" in the way that they arrange public-domain collections of names and phone numbers. For most database publishers, this limited protection was enough: As long as it cost plenty of time and money to gather data and to present them in a manageable form, the very act of creating a database imposed major competitive barriers. Today, that's no longer the case. In fact, it's now a fairly simple matter to take a huge collection of data and publish it coherently.

The database language in the DMCA—language that Congress also considered in a separate bill—would have changed that. The bill extended full copyright protection to any data "gathered, organized or maintained by another person" who invested significant resources in doing so. In addition, it also would have defined the taking of "significant" data in very broad terms, allowing a database publisher to argue that nearly any reuse of the same data was a serious copyright violation.

Scientists, librarians and consumer groups see the database bill as a disaster waiting to happen. One leading opponent, Association for Computing Machinery president Barbara Simons, worries that, "if you're able basically to copyright data, that could have a big impact on scientific research." The law might, for example, prevent scientists from collecting and pooling research data in ways that might harm a "potential market" for a commercial database publisher. In this sense, critics claim that the law basically carves out special protection for whoever is first to publish a particular collection of facts, even if those facts are still freely available in the public domain.

Database publishers, for their part, claim that copyright law doesn't reflect the effort that goes into gathering and publishing huge databases—products that latecomers can now duplicate and distribute with impunity. They also point out that the law includes exceptions for research, for nonprofit activities and for individuals who extract small amounts of data—although the definition of "small" in the bill is extremely vague. The real issue, publishers argue, is whether they'll be allowed to protect their investments or be forced to turn them over to anyone willing to cut and paste on a computer screen.

West case raises the stakes. Congress removed the database language from the DMCA with the understanding that it would revisit the issue next year. The publishing industry isn't likely to forget that promise, considering what happened in a major court decision on the issue. In this case, legal publishing giant West Publishing lost a copyright infringement suit against Hyperlaw and Matthew Bender, two smaller legal publishers that want to produce their own electronic compilations of federal court decisions—West's bread and butter. West had argued that, because it applies a particular page-numbering system and makes other minor changes to the text of the cases, it held a copyright on the material. A federal judge disagreed and ruled that West could copyright neither the decisions nor the numbering system, which is now a de facto standard in the legal community.

The West case raised the stakes for the publishing industry—database owners are now on notice that if they want to enjoy the same protections that they enjoyed in the past, they'll have to go through Congress rather than the courts. But opponents are just as adamant that allowing publishers to copyright collections of public-domain data would have unacceptable consequences, both for competition in the database market and for consumers who will face higher prices and onerous restrictions on how they use public-domain data.

The battle lines are drawn, and a compromise looks extremely unlikely. If you like watching a good fight, sit back and relax—this one is going to be a classic.

Association of Research Libraries, Intellectual Property Resources,

© 1998 by Seybold Publications, PO Box 644, Media, PA 19063; phone (610) 565-2480.
Reproduction in whole or in part without written permission is prohibited.