Sign In

Communications of the ACM

Communications of the ACM

Good News and Bad News on the Intellectual Property Front

Microsoft Research's Bill Buxton

"I am prone to think of a future where, if you are aware of the computer, that is an indication of a failure of design," says Microsoft Principal Researcher Bill Buxton.

Credit: Engadget

For the past four years, my "Legally Speaking" columns have warned about overbroad and imbalanced proposals for digital copyright, database protection, and licensing of information legislation. Since I wrote my last column on these issues there have been some important developments. As to each, there is some good news and some bad news.

One important development was the passing of the Digital Millennium Copyright Act (DMCA) by the U.S. Congress in September 1998. The good news is that the bill enacted was considerably more balanced than the legislation initially proposed by the Clinton administration. The bad news is that the bill still goes too far in regulating legitimate activities related to bypassing technical protection systems as well as technologies that can be used for such purposes.

A second development was the passage of the Collections of Information Antipiracy Act by the U.S. House of Representatives. This bill would have created a new form of intellectual property protection for the contents of databases along lines similar to the European Directive on the Legal Protection of Databases. The best news about this development is that the U.S. Senate came to be aware of substantial broad-based opposition to the approach of the House bill and chose not to move forward with that legislation. As a result, freedom of information will have at least a short reprieve. The bad news is that the bill will be back in the 106th Congress.

A third development has been the making of significant changes to proposed Article 2B of the Uniform Commercial Code. No longer will this model law aim to regulate all information transactions. Because of fierce opposition from some major copyright industry groups, Article 2B has been trimmed in scope so that it will now regulate commerce only as to "computer information transactions."

The good news for the public and for these copyright industry opponents of Article 2B is that the contraction in scope means this model law will be less disruptive of current practices and assumptions about information transactions than it would otherwise have been.

More good news comes from the fact the drafters of Article 2B have, of necessity, bowed to the will of the National Conference of Commissioners of Uniform State Laws (NCCUSL), the group that convened the Article 2B project in the first place and that must approve the model law before it goes to state legislatures for enactment. In July 1998, NCCUSL insisted the final version of Article 2B contain a provision explicitly recognizing that courts could refuse to enforce contracts or contract terms in violation of public policy. The previous draft had only recognized that "unconscionable" contracts or contract terms (such as those that are "shockingly oppressive") would be unenforceable.

The bad news is that despite the fact that a wide array of organizations have called for a "tabling" (read: killing off ) of the Article 2B project because of its impenetrable prose, problematical distinctions, and obvious tilt in favor of licensor interests, NCCUSL still intends to consider Article 2B for final approval at its July 1999 meeting. The American Law Institute (ALI), the other supervisory body for the Article 2B project, has sufficient continuing reservations about Article 2B that it has decided not to consider it for final approval at its May 1999 meeting.

All of these developments have international implications. The U.S. digital copyright legislation will be the standard by which the U.S. will judge other nations' legislation aimed at implementing the copyright treaty negotiated in Geneva in December 1996.

The hesitation of the U.S. Congress about adopting a broad exclusive property rights approach to protecting investments in the contents of databases will provide other countries with reason to be cautious about this approach. This may create some space in which an unfair competition-based approach to protecting databases can emerge as a satisfactory way to ensure prosperity for database industries without smothering prospects for follow-on innovation and scientific work that requires reuse of information.

Had the Article 2B project been warmly received by U.S. industries and been as widely enacted in the U.S. as its drafters hoped, the U.S. would undoubtedly have exerted considerable pressure for international adoption of its main features. The rest of the world may find instructive lessons in the U.S. debate over Article 2B. However this debate is resolved, the issues it raises about how information license contracts can be formed, what default rules should apply, and what public policy limitations should be imposed on the freedom of contract will continue to need resolution. If global e-commerce in information is to thrive, there must be some basic international consensus on these important issues. If Article 2B is not the right answer, another answer must be found.

Back to Top

Digital Millennium Copyright Act

Until March 1998, legislation to implement the copyright treaty negotiated at the headquarters of the World Intellectual Property Organization (WIPO) in Geneva in December 1996 had been stalled in the U.S. Congress because of contention amongst major industry groups about the level of responsibility that online service providers should have for user infringements. The Clinton administration's White Paper on Intellectual Property and the National Information Infrastructure ("White Paper") asserted that under existing law, online service providers were strictly liable for all infringing activities in their systems. It perceived no viable rationale for reducing the level of this responsibility, given that online service providers (OSPs) were in the best position to guard against infringement. Major copyright industry lobbyists agreed with this position.

Major OSPs, including not only companies such as America Online, but also telephone companies, argued, that it was technologically and economically infeasible for them to monitor all user activities, as a strict liability standard would require. They wanted Congress to exempt OSPs from liability unless they knew about infringing activity and took no steps to redress the problem.

It took more than a year for a compromise provision to emerge that broke the OSP liability logjam. The compromise provision is far more complicated than the simple positions with which the two sides began. Readers who run online services of any kind (even for a library, community group, or school) should familiarize themselves with the details of the statute. For other readers, the following summary should suffice to give a flavor of this part of the DMCA.

Under this law, OSPs can qualify for one of a number of exemptions from liability ("safe harbors") if they are willing to terminate service to repeat infringers, to accommodate standard technological measures adopted by copyright industries to protect works, and to abide by certain other conditions. "Safe harbors" have been created for system storage, for enabling links to other sites, for system caching, and for transmission and routing of information, as long as OSPs do not know about the presence of any infringing material and do not directly benefit from infringement.

OSPs are under further obligation to remove material from their systems upon receiving notice that the material is infringing. The statute also provides a further process for restoring information to a particular site if an allegation of infringement proves to be spurious.

As the broadcasting, cable, motion picture, recording industry, and publishing communities became aware of the Article 2B project, they began to realize it threatened to disrupt many existing practices and licensing assumptions in their industries.

Once the OSP logjam was broken, it was reasonably clear that there would be digital copyright legislation in 1998. It was also reasonably clear that this legislation would contain a provision about circumvention of technical protection systems because the predominant view was that such a provision was needed in order to implement the WIPO Copyright Treaty. It contained the following provision: "Contracting parties shall provide adequate legal protection and effective legal remedies against the circumvention of effective technological measures that are used by authors in connection with the exercise of their rights under the WIPO Treaty or the Berne Convention and that restrict acts, in respect of their works, which are not authorized by the authors concerned or permitted by law."

The Campbell-Boucher bill in the House would have outlawed acts of circumvention of technical protection measures undertaken for purposes of enabling or engaging in copyright infringement. This minimalist position did not meet with favor from lobbyists for copyright industry groups or with the Clinton administration. They preferred a bill that would outlaw all acts of circumvention (except for those legitimately undertaken for law enforcement or national security purposes).

USACM's Public Policy Committee was among the groups expressing concern that the administration's bill would impede advances in the computing field. Committee Chair Charles Brownstein, writing to the House of Representatives about this bill, made the following observations:

"The anticircumvention provision criminalizes critical techniques that are needed for computer and network security and for legitimate research and teaching purposes. Additionally, the legislation would outlaw devices and technologies that are crucial for making information processing systems trustworthy and secure. By attempting to regulate technologies rather than undesirable infringing behavior, progress of research in computer security, including cryptography, be significantly hindered ...We urge you to consider the research and publication process that is necessary to the advancement of science in any amendment meant to address security testing."

The USACM Public Policy Committee helped to make a difference in the public policy debate about regulating circumvention of technical protection systems. It should take some pride in knowing that the DMCA, as finally enacted, contains meaningful exceptions for circumvention for encryption and computer security research purposes that were missing from earlier versions of the bill.

Nevertheless, these exceptions are still too narrow to allow the full range of legitimate acts of circumvention. Congress was sufficiently concerned about the implications of the anticircumvention regulations for fair use and the public domain that it suspended these regulations for two years and directed the Librarian of Congress to study these regulations and report to Congress with recommendations about the about anticircumvention regulations. So perhaps there will be a way to improve this aspect of the DMCA.

Of greater concern is the provision outlawing technologies with circumvention-enabling uses. This provision could freeze development of valuable technologies and may make software companies vulnerable to "strike suits" by copyright industry groups fearful that new technologies will undermine their marketing strategies. In addition, this provision makes no exception for enabling development of technologies to engage in legitimate circumvention activities. Thus, numerous exceptions carefully crafted by Congress that recognize legitimate circumvention activities can be undermined if all the tools necessary to engage in circumventions are illegal under the DMCA.

Back to Top

Database Protection

Compiling information, verifying it, formatting and updating it can be a very expensive enterprise. A substantial private market has grown up to supply information to those who want it. Many people are prepared to pay a lot of money to get high-quality information in a timely manner. To enable private firms to recoup their investments in developing information resources such as databases, there is general agreement that there should be be a way for these firms to protect collections of information from market-destructive appropriations.

Copyright has historically been the intellectual property law on which data compilers have relied to protect themselves from such appropriations. Copyright does not, however, provide protection to all compilations, but only those that exhibit sufficient creativity in the selection and arrangement of their contents. White-page listings of telephone directories are, for example, ineligible for copyright protection because they are inclusive, rather than selective, and use an unoriginal, standard format. Even when copyright protection is available to a data compilation, the scope of such protection tends to be thin. This is because copyright does not protect the data in the compilation, but only creative elements in the compilation's selection or arrangement of data elements.

Do data compilers need more protection from the law than copyright alone provides, and if so, what form should this protection take? These questions have been much debated in the international arena. The European Union decided to create a new form of intellectual property protection for those who have made substantial investments in developing databases. Makers of such databases are protected against the extraction or reuse of the whole or a substantial part of the database for a 15-year period. In order to induce other countries to adopt an equivalent law, the EU decided to make its new legal protection available only to databases of nationals of EU member states and to nationals of countries with an equivalent law.

Some of the impetus behind the Collections of Information Antipiracy Act that passed in the House of Representatives during 1998 came from concerns that this law was needed to protect U.S. database developers in Europe.

There was, however, strong opposition to this bill from the scientific and educational communities in the U.S., especially from the National Academies of Science and Engineering, as well as from industry groups such as the Information Technology Association of America. The Federal Trade Commission and ultimately the U.S. Department of Commerce on behalf of the administration opposed the House bill. The Department of Justice even opined that the database bill might be unconstitutional.

Here too the USACM Public Policy Committee made a contribution to the debate. Committee Chair Brownstein, writing to Congressman Bliley about the database legislation, declared that the database legislation "will have a chilling effect on the U.S. science research enterprise." In particular, the letter complained that the "fair use provisions of the bill fall far short of the exemptions necessary to permit researchers to verify others' results, educators to demonstrate models in classrooms, scientists to make use of government databases, and other traditionally protected uses." These issues will need to be revisited during the 106th Congress when database legislation once again emerges on the legislative agenda.

Back to Top

Article 2B of the UCC

A year ago, Article 2B of the UCC (a draft model law for regulating transactions in information) seemed to be well on its way to receiving the blessing of the ALI and of NCCUSL, its two sponsoring bodies. ALI and NCCUSL were expected to consider Article 2B at their respective May and July meetings in 1998. Upon approval of both groups, there was every reason to expect that state legislatures would begin adopting it as early as the fall of 1998. Had this occurred, U.S. policymakers probably would have begun pushing for international adoption of substantially the same rules on the theory that the Article 2B drafting committee had considered the tough issues and had come up with solutions that would be good for commerce on a global level.

A year later, the future of the Article 2B project is uncertain. This is true despite the fact that the drafting committee has made some significant changes to the text of Article 2B in response to criticisms from academic and industry groups. Although NCCUSL currently intends to go forward with a final review of Article 2B at its July 1999 meeting, ALI has enough reservations about Article 2B that it has decided not to put the draft to an approval vote in May. There is, moreover, a growing sense that Article 2B may be unadoptable in states where major information industries are based, such as New York and California. Even NCCUSL may back away from Article 2B if it becomes convinced there is sufficient opposition to Article 2B from affected industry groups to prevent it from achieving its high aspiration of becoming for the information economy what Articles 2 (regulation of sales of goods) and 2A (leases of goods) have been for the manufacturing economy.

Several factors have contributed to the declining prospects of Article 2B. One important factor was probably the overly ambitious nature of the project. Article 2B would surely have generated debate even if it applied only to computer software, a core subject matter of the Article 2B project from its inception. However, the expansion of Article 2B's scope to encompass all information transactions was, in retrospect, a mistake, particularly given that there was insufficient representation on the drafting committee of the wider, affected information licensing communities.

As the broadcasting, cable, motion picture, recording industry, and publishing communities became aware of the Article 2B project, they began to realize it threatened to disrupt many existing practices and licensing assumptions in their industries. At first, organizations representing these industries raised their concerns to the drafting committee. However, in September 1998, six organizations representing these industries, including the Motion Picture Association of America (MPAA) and the National Association of Broadcasters, took the dramatic step of writing to the chair of the drafting committee and to the director of the ALI to urge that ALI and NCCUSL table Article 2B.

Although expressing appreciation for the committee's "sincere and considerable efforts ... to fashion an acceptable document," the MPAA letter stated that these organizations had become "convinced that the current draft is fatally flawed in its fundamental premise that all transactions in 'information' may be governed by a single set of contractual rules." It further expressed doubt that the draft could be "reworked to accommodate the breadth and variety of all of our respective and diverse business practices." (This letter can be found at docs/v9-98.pdf.)

When it became clear that the Article 2B drafting committee only intended to narrow the scope of the draft to "computer information transactions," the same six organizations wrote another letter to the heads of the drafting committee, the ALI, and NCCUSL announcing, "the latest draft fails to cure the serious infirmities that caused us to oppose previous drafts." This letter contends correctly, in my view, that the distinction between "computer information transactions" and other information transactions is unworkable.

Although Article 2B purports to exclude core activities of the entertainment industry, the latest version of Article 2B would nonetheless apply to multimedia products and interactive services produced by entertainment companies as well as to digital sound recordings and movies. MPAA et al. regard "This change [in] scope [to be] merely cosmetic... [It] does not create a meaningful distinction between the software industries on the one hand and entertainment and the publishing industries on the other."

The six organizations also objected to a lack of clarity in rules about whether or to what extent Article 2B would apply to "mixed transactions" (for instance, a license to publish something in both the electronic and print versions of the New York Times). This, along with generous "opt-in" provisions (even real-estate transactions could opt into Article 2B's contract formation rules, if they wanted to) and the prospect of courts applying Article 2B licensing rules by analogy makes the seemingly narrower scope of Article 2B "illusory." These industry groups have declared that they "will be forced to actively oppose its enactment" unless the project is tabled or substantially revised.

These six copyright groups are not the only groups to have expressed substantial opposition to the Article 2B project. The USACM Public Policy Committee is among the other organizations that has criticized the Article 2B project. Brownstein, on behalf of this Committee, wrote to the heads of the drafting committee and ALI in October 1998, saying "we have concluded that within the constraints of the present process, it is highly unlikely that the myriad concerns raised from many industry sectors can be satisfactorily resolved by a variant on the current draft." The Brownstein letter expressed special concern about the "potential impact [of Article 2B] on legitimate acts of reverse engineering of computer systems." Although Article 2B would not outlaw reverse engineering, "it would contribute to legal uncertainty about reverse engineering by enabling an argument that shrinkwrap license agreements barring reverse engineering might be enforceable."

The Brownstein letter also expressed concern about "the growing use of 'no disclosure of flaws' and other anti-criticism clauses in licenses for software and other digital information products which can impede the free flow of information within the technical community."

The drafters of Article 2B believe these and related concerns have been satisfactorily dealt with by a provision added to Article 2B at the November 1998 drafting committee meeting. Section 2B-105 (b) now provides that: "If a term of a contract violates a fundamental public policy, the court may refuse to enforce the contract, or it may enforce the remainder of the contract without the impermissible term, or it may so limit the application of any impermissible term as to avoid any result contrary to public policy, in each case, to the extent that the interest in enforcement is clearly outweighed by a public policy against enforcement of that term." Mentioned in the reporter's notes about this section are references to public policies favoring innovation, competition, and freedom of expression.

While this provision makes Article 2B more balanced, it still sets a higher hurdle to judicial oversight of contractual terms than the Federal Trade Commission, among others, thinks desirable. Only because it was convinced that NCCUSL would not approve Article 2B unless the final draft included a provision of this sort did the committee see the necessity for adding such a provision to 2B. Its compromise solution was to require that the public policy at stake be "fundamental" and that the interest in enforcement of the contract term must be "clearly outweighed" by the public policy.

While this compromise may satisfy NCCUSL on the public policy override issue, there are so many other problems with Article 2B, including the lack of clarity in its drafting. These, I predict, will cause the Article 2B project to fail as a model law notwithstanding the vigorous support of Microsoft and certain information industry groups (NASDAQ among them).

Article 2B, however, continues to have momentum. There has been so much money expended on the project that NCCUSL and ALI will find it difficult to abandon it, however errant and incomprehensible it might have become. Regardless of the fate of Article 2B, however, substantial questions about when and to what extent mass-market licenses for software and other information products and services should be enforceable will still need to be resolved.

Back to Top


The three legal initiatives discussed in this column, the Digital Millennium Copyright Act, the Collections of Information Antipiracy Act, and Article 2B of the Uniform Commercial Code, are all responses to the challenges digital technologies have posed in recent years. Digital information is more easily copied and disseminated via global networks than print information. In view of this, it is not surprising that there has been a call for strengthening copyright laws to deal with some of these challenges, for new laws to protect digital databases, and for new laws to clarify rules about the licensing of digital information.

Although digital technologies do present some threats to information providers, policymakers should be careful not to overreact to these threats by adopting overbroad and imbalanced rules that may benefit some industries, but do so to the detriment of other industries, especially those in emerging markets and to the detriment of the public. Experience with each of the legal initiatives discussed here demonstrates that the threat of overreaction is more than hypothetical. Fortunately, USACM and other organizations understand the need to insist on balancing principles in these new laws. In the last year, some progress has been made in developing more balanced rules and in rejecting highly imbalanced rules. There is probably more good than bad news in the developments reported here but continued vigilance will be required to ensure continued progress.

Back to Top


Pamela Samuelson ( is a professor of Information Management and of Law at the University of California at Berkeley.

©1999 ACM  0002-0782/99/0300  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 1999 ACM, Inc.


No entries found