D-Lib Magazine
January 1999

Volume 5 Number 1

ISSN 1082-9873

Armageddon on the Potomac

The Collections of Information Antipiracy Act

blue line

Jonathan Band
Morrison & Foerster, Washington, D.C.


One of the more popular movies released during the summer of 1998 was Armageddon. In this sci-fi/action flick, Bruce Willis and his team of oil drilling rough-necks try to save the world from an asteroid that is on a collision course with the Earth. At literally the last moment, Bruce detonates a nuclear warhead his team implanted deep within the asteroid, splitting the asteroid into two halves which pass safely on either side of the Earth.

The information policy of this country experienced a similar close brush with death this fall. The asteroid was the Collections of Information Antipiracy Act, H.R. 2652, and the Bruce Willis character was played by none other than Senator Orrin Hatch, chairman of the Senate Judiciary Committee. Unlike Bruce Willis, however, Senator Hatch survived the destruction of the asteroid. This article tells the story of Armageddon on the Potomac.

Background. Since the birth of the Republic, facts have resided in the public domain. To be sure, a compilation of facts could receive copyright protection, to the extent the compilation reflected selection, coordination, and arrangement of the facts. But all that copyright protected was the "expressive" aspect of the compilation -- the selection, coordination, and arrangement -- and not the facts themselves.

Thus, a second generation publisher typically could not copy an entire compilation and resell it, because in doing so he probably would copy some degree of selection, coordination, and arrangement -- expression. The second generation publisher could, however, extract facts from the first compilation for use in his compilation. Indeed, he could conceivably extract all the facts from the first compilation, so long as his compilation did not reflect any of the selection or arrangement of the first compilation. This would occur if he presented the facts in a different manner, or if he added enough facts of his own. Facts were viewed as building blocks of knowledge that everyone was free to use and reuse.

The only exception to this basic principle was the "sweat of the brow" doctrine, which arose in a few jurisdictions to deal with a very narrow category of compilations: directories like telephone books in which there was no protectable expression. Courts in these jurisdictions thought it was unfair and unwise to afford no protection to the efforts of people who assembled these plain vanilla directories. Accordingly, they interpreted the copyright laws as prohibiting the wholesale copying of these expressionless directories.

The sweat of the brow doctrine fell into disfavor after the adoption of the 1976 Copyright Act. In 1991, the Supreme Court drove the nail into the coffin with its unanimous decision in Feist v. Rural Telephone, 499 U.S. 340 (1991). The Court found that under the copyright clause in the Constitution, copyright protection could extend only to expressive elements in compilations, and that effort without creativity could not convert facts into expression.

The Feist decision had no discernable economic impact on the database industry. The number of published databases continued to increase, and the size of the databases continued to grow. Significantly, in the years immediately after the Feist decision, the U.S. database industry did not actively lobby the U.S. Congress for legislation to resurrect the sweat of the brow doctrine.

The perturbation which knocked the asteroid out of a safe orbit and sent it hurtling towards the Earth occurred in Europe in 1992. The bureaucrats of the European Commission decided to create a new form of protection for databases in an effort to jumpstart the European database industry, which lagged far behind the US database industry. Under this new regime, a second generation publisher could not extract a substantial part of a first generation database, even if the second publisher did not extract any protectable expression. The European Union ultimately adopted this database directive in 1996.

The European Database Directive prompted the introduction of similar legislation in Congress in the Spring of 1996. This legislation, H.R. 3531, met with sharp criticism from numerous quarters, and died with the close of the 104th Congress.

Act I. The action of Armageddon on the Potomac begins against this background on October 9, 1997, when Howard Coble, chairman of the House Intellectual Property Subcommittee, introduced the Collections of Information Antipiracy Act, H.R. 2652. H.R. 2652 contained the same basic prohibition found in the EU Database Directive and H.R. 3531: a person could not 1) extract, or use in commerce, 2) a quantitatively or qualitatively substantial part of 3) a collection of information gathered or maintained by another person through the investment of substantial resources 4) so as to harm the actual or potential market for a product or service containing that collection of information. This legislation would give a database publisher unprecedented control over the reuse of the information contained in the database.

A relatively small group of database publishers actively supported the legislation. They included Reed-Elsevier (the world’s largest publisher of scientific journals), Thomson (which recently had purchased West, the largest legal publisher), the American Medical Association (which publishes the Physician’s Desktop Reference as well as membership lists), and the stock exchanges, particularly the New York Stock Exchange. What all these publishers had in common is that they either controlled or had special access to the sources of the information which they published. Because of this unique market position, the legislation only benefited them; they could prevent others from taking information from their databases, without having to worry about their own access to information.

Most database producers, however, do not have special access to information. The legislation might have increased their ability to protect their work product, but this benefit would have been offset by the increased cost of the information input. Moreover, many database producers create databases only for internal use, and thus have no fears of others copying their information. For these producers, the legislation imposed only costs and no benefits. Accordingly, value-added publishers such as Dun & Bradstreet and Bloomberg; companies which rely on data for many of their operations, such as AT&T and MCI; and scientific researchers who build on existing data to push forward the frontiers of human knowledge; voiced concerns with the legislation.

The House Intellectual Property Subcommittee held two hearings on H.R. 2652, one in October, 1997, the other in February, 1998. The hearings focused on whether additional database protection was needed at all, with little reference to the specific language of H.R. 2652. The proponents made two arguments. First, they argued that since the Feist decision, the federal courts had been depriving databases of meaningful copyright protection. The poster child for this contention was the Eleventh Circuit’s en banc decision in Warren Publishing v. Microdos, 115 F.3d 1509 (11th Cir. 1997). In that case, the court determined that the defendant's copying of large portions of Warren’s book of statistics concerning the cable industry did not constitute of copyright infringement. The absence of meaningful protection, asserted the proponents, would chill investment in the development of new databases.

Second, the proponents argued that legislation was needed to meet the reciprocity requirements of the EU Database Directive. The EU had provided that foreign nationals could receive the special protection of the Database Directive only if their countries of domicile afforded similar levels of protections to European database publishers.

The opponents of the legislation readily conceded that the Eleventh Circuit had wrongly decided Warren Publishing, but contended that one wrong decision did not justify a major statutory amendment. They pointed to many other post-Feist decisions where courts had extended adequate copyright protection to databases. The opponents listed other forms of protection available to database publishers, including licensing, trade secrecy, state common law misappropriation, trademark, and technological measures. While no one form was perfect, all these methods taken together with copyright provided publishers with adequate protection. Additionally, opponents indicated that there was absolutely no concrete evidence that publishers were not investing in the creation of new databases out of fear of piracy. Indeed, all economic indicators demonstrated the health and vitality of the database industry.

With respect to the EU Database Directive, the opponents noted that a U.S. publisher could receive protection if it located a subsidiary in Europe. Further, the opponents argued that the U.S. should oppose reciprocity provisions in intellectual property laws, and seek relief from the Database Directive before the World Trade Organization.

Act 2. Notwithstanding the contentions of the science, library, and educational communities and the communications and financial services industries that additional database legislation was not needed, the House IP subcommittee decided to proceed expeditiously with H.R. 2652. With a mark-up scheduled for March 24, 1998, attention shifted from the theoretical threshold issue of whether legislation was needed at all to a more detailed critique of the specific provisions of H.R. 2652.

During Act 2, the following pattern emerged: an industry would explain the bill’s negative impact on its way of doing business, and the House would respond by granting that industry a specific exemption which addressed some of its concerns. The basic structure of the legislation, however, remained the same. This meant that industries not sufficiently organized to lobby effectively for an exception felt the full force of the legislation. According to the opponents, in the absence of an exception, H.R. 2652 gave database publishers perpetual monopoly control over transformative downstream uses of the information in the databases. The opponents explained their reasoning as follows:

Why perpetual? The bill allows the publisher to prevent extraction or use for 15 years after making the investment in the collection of information. The proponents state that the protection in the collection can be renewed by making additional investments in the collection in the form of verification, maintenance, and routine updating. Thus, an online database which is periodically checked or maintained can receive perpetual protection, even if no new information is ever added.

Why monopoly control? In many instances, there is no feasible way for another person to collect the information independently, for at least three reasons. First, for certain historical information (the number of cows in Vermont in 1950), there is no way to go back in time to gather the information. Second, the publisher is often the original producer of the information. The New York Stock Exchange, for example, is the only complete source for the quotes of all the trades that take place on its floor. A firm could theoretically call all the companies trading on the Exchange every minute to find out what they were trading for, but that would be economically infeasible; the firm’s costs for collecting the information would be much higher than the Exchange’s. Third, the publisher often builds up its database incrementally over many years, incurring modest costs along the way. A new entrant, in contrast, would have to assemble all the information at one time. The cost of this massive collection effort would be prohibitive.

Why no transformative uses? Almost any kind of transformative use, such as abstracting a database or combining some of the data from the database with information from other sources to create a new and useful database, harms a potential market for the database publisher. This is particularly the case given that even one lost sale can constitute "harm," and that the potential market includes any market the publisher has "current and demonstrable plans to exploit," even if the publisher has not begun to implement these plans nor has revealed these plans to the market. By allowing a publisher to stop "harm" to a "potential market," the bill enables the publisher to stop innovative uses of the database by others, which of course also diminishes the publisher's own incentive to innovate.

Why information? Although the bill is entitled the Collections of Information Antipiracy Act, liability attaches even if a person extracts or uses a "qualitatively substantial" part of the database which may be quantitatively insubstantial. The proponents have asserted that taking even two pieces of information could trigger liability. The problem is that the user or new entrant has no way of knowing what information is "qualitatively substantial."

The opponents prepared an alternative bill which federalized state common law misappropriation. This narrower approach avoided the critical problems of H.R. 2652 identified above, but at the same time provided database publishers additional protection against wholesale piracy. But by then, H.R. 2652 had too much momentum. At the Intellectual Property Subcommittee mark-up on March 24, 1998, virtually no opposition to the legislation was registered. The only skepticism was voiced by Congresswoman Zoe Lofgren, who questioned the bill’s constitutionality. Lofgren stated that H.R. 2652 may run afoul of the Constitution’s copyright clause because it created copyright-like protection for databases without insisting upon the creativity required by the Supreme Court in Feist. The subcommittee ignored Lofgren’s concerns, and reported out the legislation.

One week later, the full Judiciary Committee reported out the legislation with little discussion. The Commerce Committee voiced concerns with the legislation, and had H.R. 2652 referred to it for one week. Although the Commerce Committee failed to take any action on the bill during that week, Commerce Committee Chairman Bliley wrote a letter to Chairman Hyde listing his concerns and expressing hope that they would be addressed before the legislation was enacted.

After the Commerce Committee failed to act on the bill, Chairman Hyde requested the House leadership to place H.R. 2652 on the suspension calendar, a procedure generally reserved for non-controversial bills. Twice the chorus of nonprofit and commercial opponents was sufficient to get H.R. 2652 removed from the suspension calendar. The third time, however, the opponents could not keep the bill off the suspension calendar. The opponents lined up sympathetic Congressmen to speak against the bill when it came to the floor for a vote on May 19, 1998. The House leadership then moved up by several hours the time the bill came up for a vote. The only Congressmen in the chamber were a handful of supporters of the bill, who knew of the time change, and one opponent, Congressman George Brown. Congressman Brown spoke against the bill, but did not request a roll call vote. The bill quickly passed on a voice vote in the near empty chamber. (A roll call vote would have demonstrated the absence of a quorum, requiring the bill to be withdrawn for later consideration.)

Although the bill had passed the House, the Senate Judiciary Committee made it clear that it had no intention of considering database legislation in the 105th Congress. Its intellectual property priorities were the patent reform bill and the Digital Millennium Copyright Act. From its perspective, H.R. 2652 as passed by the House remained too controversial and problematic, and had the potential of interfering with the enactment of the priority legislation.

Chairman Coble, however, was determined to enact H.R. 2652 in the 105th Congress. Accordingly, when the Judiciary and Commerce Committees were negotiating over competing versions of the Digital Millennium Copyright Act, Chairman Coble threw H.R. 2652 into the mix. He agreed to the Commerce Committee’s amendments to the DMCA in exchange for the Commerce Committee agreeing to the attachment of H.R. 2652 to the DMCA.

On August 4, 1998, the House passed the DMCA, with H.R. 2652 comprising Title V of the legislation. Now the Senate could no longer avoid the issue; the Collections of Information Antipiracy Act would be before the conference committee when it tried to reconcile the conflicts between the House and Senate versions of the DMCA.

Act 3. With his hand forced by Coble’s stratagem, Chairman Hatch directed the proponents and opponents of H.R. 2652 to attempt to negotiate a compromise. Under the supervision of Edward Damich, the Senate Judiciary Committee’s chief counsel for intellectual property, the interested parties negotiated for three weeks during August. The opponents made it clear at the outset that they would accept a bill that prohibited wholesale database piracy, but that they would oppose a bill which prevented the development of value added databases. The proponents, for their part, were adamant that the first generation publisher should have control over downstream uses of the data it collected.

During the course of the negotiations, the Administration for the first time expressed its opinion of H.R. 2562. It agreed with the opponents that the legislation could chill the development of value-added databases, particularly by scientists. Additionally, the Office of Legal Counsel in the Department of Justice opined that the legislation was probably unconstitutional, violating both the copyright clause and the First Amendment.

The negotiations did not produce a compromise, so Chief Counsel Damich prepared his own draft based on H.R. 2652 and amendments presented by the opponents. Neither side accepted the discussion draft. At the opening of the conference on the DMCA on September 17, 1998, several conferees, including Senator Leahy and Congressman Dingell, raised questions about the wisdom of including Title V concerning databases. During the pendency of the Conference, Chief Counsel Damich circulated two more drafts which generally moved in the direction of the opponents. The Administration sent a letter raising concerns with the Damich drafts, as did the Federal Trade Commission. Further, ten Senators sent letters to Chairman Hatch, asking him to omit Title V because the Senate had not had the opportunity to consider it. At a critical juncture, the proponents rejected Chief Counsel Damich’s third draft, and Chairman Hatch pulled the plug on Title V.

Chairman Hatch had objected to both Titles V and VI of the House-passed DMCA. Chairman Coble agreed to dropping Title V in exchange for Chairman Hatch’s agreement to the retention of Title VI, which establishes special protection for boat hull designs. The boat hull provision constitutes Title V of the DMCA ultimately passed by Congress and signed by the President. Chairman Hatch also committed to considering database legislation at the beginning of the 106th Congress.

Coming Attractions of the Sequel. Both Chairmen Coble and Hatch have committed to revisiting database legislation early in the 106th Congress. Chairman Coble probably will start with H.R. 2652 as twice passed by the House in the 105th Congress. Chairman Hatch may start with one of the discussion drafts developed by Chief Counsel Damich, or a different vehicle altogether.

During the 105th Congress, most Congressmen and Senators used their limited intellectual property attention span on the DMCA, and hardly focused on H.R. 2652. This is odd from an intellectual property standpoint, because H.R. 2652 arguably had a much farther reach than the DMCA. Nonetheless, the convergence of political interests in the DMCA led to it receiving far more prominence in Congress than H.R. 2652. Conversely, in the 106th Congress database legislation should receive more attention than it did in the 105th Congress because it will not have to compete with a intellectual property bill of the breadth of the DMCA. Of course, other matters may force intellectual property off of Congress’ radar screen altogether.

Copyright © 1999 Jonathan Band

Postscript from author, Jonathan Band, dated January 21, 1999:

"On January 19,1999, Howard Coble, Chairman of the House Intellectual Property subcommittee, introduced H.R. 354. H.R. 354 is identical to H.R. 2652 with two exceptions. First, H.R. 354 appears to correct the problem with perpetual protection identified in H.R. 2652. Second, H.R. 354 contains a fair use provision which probably will be too narrow to permit most transformative uses.

On the same day, Orrin Hatch, Chairman of the Senate Judiciary Committee, placed in the Congressional Record a statement indicating his interest in addressing database protection in the 106th Congress. He attached to his statement, for discussion purposes, three alternative approaches to database legislation: Chairman Coble's H.R. 354; the final draft prepared by Chief Counsel Damich in the 105th Congress; and a narrower unfair competition bill supported by opponents of H.R. 2652.

No hearings have been scheduled in either the House or the Senate."

Top | Contents
Search | Author Index | Title Index | Monthly Issues
Previous Story | Next Story
Comments | Home| E-mail the Editor

D-Lib Magazine Access Terms and Conditions

DOI: 10.1045/january99-band