Prior to addressing specific concerns with H.R. 2652 and related intellectual property proposals that we share with others in the public and private sectors, let me briefly outline a concern central to our reservations with this legislation. As with H.R. 3531, there has been no demonstrated need for new intellectual property protections for "electronic collections and other collections of factual material." Many in the not-for-profit and commercial sectors believe that adequate forms of protection exist - copyright, contracts, and increasingly, technological measures. Since the testimony of Professor Jerome Reichman addresses these legal and technological measures, they are not repeated in this statement. The United States database industry is robust, indeed, the United States leads in this arena. The recently issued report by Dr. Tyson, sponsored and funded by Reed-Elsevier, Inc. and The Thomson Corporation, does not provide compelling empirical evidence that the threshold has in fact been crossed, requiring new levels of protection for owners and creators of collections.

That said, please let me make clear that the library community is committed to participation in discussions with creators of databases and related collections, if market failure or significant competitive harm is proven, and additional protections beyond the Copyright Act and the terms and conditions available via contracts and licenses are required. We regularly negotiate licenses with database proprietors and ensure that their terms are met within our institutions. We rely on their products as valuable resources. This hearing provides a forum to initiate discussions relevant issues. The Subcommittee has undertaken a useful role in bringing some of the perspectives to bear on this important issue. Additional hearings to elicit positions from other interested stakeholders will be important.

The recently issued study by the Copyright Office, "Report on the Legal Protection for Databases" notes that, "all agree that the proponents of a new form of statutory protection have the burden of establishing the need for such protection."1 As detailed in the Copyright Office report, there is a large and growing segment of the commercial sector that joins the not-for-profit sector in the position that such a need for additional protection has yet to be demonstrated. The following list of questions are illustrative of those requiring thorough examination prior to legislating in this arena.

1) Is there sufficient incentive for industry to invest in the creation of new collections without additional intellectual property protection? Are there areas in which there are compelling needs for databases that are not being created?

2) If it is determined that indeed, further incentive for investment is necessary, would additional investment occur under the regime proposed in H.R. 2652? What new protections would be most appropriate? Some members of the information industry believe that current legal models are sufficient, i.e., contracts and licenses. If this is true, are there other legal remedies that should be considered?

3) How would any new legislation assure that access to the "underlying data" is always possible?

4) What will assure the full and open exchange of information for non-commercial use by the scientific, education, and library communities?

5) How does one guarantee that there are adequate incentives for innovation and transformative commercial uses of protected data?

6) If acting on legislation is deemed necessary, what meaningful exemptions or limitations should be included?

I will now turn to several specific concerns of the library community with the legislation.

As detailed in the Copyright Office study, traditional misappropriation doctrine deals only with injuries resulting from unfair conduct among commercial competitors. In contrast, H.R. 2652 would penalize any "substantial" use of extraction of data that affected the actual or potential market for a product or service. Thus, the prohibitions in this bill would go beyond the sphere of competition, to reach the activities of individual consumers and not-for-profit institutions such as libraries. This casts a wide net, has potentially serious consequences to libraries as providers of information services, and could conceivably chill the legitimate use of databases and related collections.

Researchers require access to large and small amounts of data, not only insubstantial portions or an "individual item." Many databases used by researchers are specialized or serve "niche markets," so their use and client base is relatively small. Even a small extraction of information from such a database could be deemed to harm the existing or potential market, thus establishing the basis for liability under H.R. 2652 while defeating any possible claim of an exemption for research use. More generally, researchers often need access to significant amounts of data, not only the "insubstantial" parts or "individual items" which the bill would permit them to take. Ordinary research activities might not be possible if these provisions are enacted because these research activities would harm the current or potential market as envisioned by H.R. 2652. Finally, extending the protection to "potential markets" is extremely problematic. This could have a chilling effect on the development of new products and information resources.

The traditional misappropriation doctrine is limited to "hot" or "time sensitive" information. In contrast, H.R. 2652 would apply to all compiled information, including archival and historical data. In addition, H.R. 2652 as drafted could be retroactive, extending protection to existing compilations. As a consequence, it is important to consider the impact of this legislation on library budgets.

Libraries engage in licensing agreements with commercial publishers and these will continue, indeed are increasing in number and scope. The acquisition budget for this year at Johns Hopkins University is approximately $8 million with approximately $950,000 devoted to online resources. These figures do not include hardware, software, network support and equipment, and personnel. Provisions in H.R. 2652 imply significant new costs for my institution and other libraries and in the amount of data and information that we will be able to make available to students, faculty, and the public.

Moreover, no term limit on protection is proposed, protection would be perpetual, thus information would not move into the public domain. In creating new information, our system of access and publication relies upon use of both protected and public domain data. This proposal would significantly reduce the amount of information available in the public domain, with serious consequences for the creation of new knowledge. A related concern is that there are no obligations imposed on proprietors to keep their data accessible, or to notify users when they intend to abandon databases and their claims to legal protection in them. One key function of libraries, research libraries in particular, is the preservation of our cultural and historical heritage. Fulfilling this role would appear to be difficult if not impossible under this regime. Digital works are effectively destroyed when the format or medium in which they are recorded becomes obsolete. Unless of course libraries or archives are able to migrate the data to a new platform or extract data from databases they have legally acquired when proprietors cease to import it into contemporary formats. Yet the existence of a "potential" market could hinder this.

Key terms such as "substantial part" and "substantial monetary or other resources," are not defined in H.R. 2652. To fully appreciate the impact of this legislation on the library community, it will be important to include the definitions of key terms. The bill does define "information" or that subject to protection. The definition, "facts, data, works of authorship, or any other intangible material capable of being collected and organized in a systematic way," would sweep in a vast array of information products and services. The definition is applicable to print and electronic and would cover directories, anthologies, CD-ROMs, online databases, reference works, and more. It will be important to evaluate the extensive overlap between the coverage of H.R. 2652 and that of copyright - especially in light of the fact that while the term of copyright is limited, the duration of protection under the legislation is not. The effect of the proposed legislation, to whatever material it applies, could be to prevent libraries and their patrons from pursuing many important and legitimate functions permitted under copyright today.

One other important aspect of the Copyright Act is that facts are not protectable. The 1991 Supreme Court decision, Feist v. Rural Telephone held that comprehensive collections of facts arranged in conventional formats were not protected under copyright, and could not constitutionally be protected under copyright. The decision rejected the notion that a compiler's "sweat of the brow" could ever substitute for the "original authorship" that the statute and the constitutional Copyright Clause require as the condition of copyrightability. Since H.R. 2652 is designed to "protect" investment rather than creativity, it will undermine the long and critically important tradition in the United States of protecting expression but not facts. This tradition is based on an appreciation that such a policy stimulates innovations in the public and private sectors, supports the educational process, and "promotes the progress of Science and the useful arts."

Almost fourteen hundred libraries throughout the United States provide no-fee access to government information to the public. Public of course includes the public and private sectors. Activities include assisting an entrepreneur in the Boston Public Library to download Census files and create a new product based on this government information while providing comparable access to an educator. This is an example of how library provision of access to information spurs the economy and supports education. Ensuring effective and long-term public access to government information is a key function of the library community.

The library community has a long history of working to assure that Government at all levels cannot assert copyright or copyright-like protections over government-produced data. Section 1203 does not extend protection to a "governmental entity, whether Federal, State, or local" nor to "any person exclusively licensed by such an entity." This is consistent with section 105 of the Copyright Act and extends to state and local government information. This is a positive provision but amplification is required. The bill fails to address those companies which create information products and services based on government information publicly available. No protection should be provided to any private party that merely repackages government information and/or does so without significant "value added." In addition, there should be no protection for defacto sole source providers of government information.

One reading of this legislation suggests that a not insignificant portion of government information could be covered by this proposal thus effectively removing it from the public domain. The legislation would permit private parties to compile collections of non-copyrightable government information and assert protection over them. In some cases, it would be virtually impossible for other commercial or non-profit entities to compile this data as the original compiler acts as a defacto sole source. When coupled with seemingly perpetual protection, we believe that the proposal will significantly remove significant amounts and types of information from the public domain.

Mr. Chairman, these issues have been and will continue to be deliberated on the international level. As a member of the U.S. delegation to the World Intellectual Property Organization (W.I.P.O.) this past December, I can attest that a number of factors precluded consideration of additional protections for databases. Lack of time by member states to review the proposal prior the December meeting, lack of time during the December W.I.P.O meeting, and the strong opposition by others in the international community, postponed W.I.P.O. discussions. Since December, W.I.P.O has met twice to discuss and exchange information regarding additional protections for databases. The sentiment at both meetings was to garner additional information regarding country practices in this arena and the need for additional information regarding the impact of such proposals on a diverse array of constituencies. At the most recent meeting, many countries with the exception of those in the European Union, expressed significant concerns or opposition to a new regime. In addition, few European countries have fully implemented the European Union directive on databases. Thus, we have time to fully consider the consequences and needs of our communities without reacting to international efforts. There is ample time to devise solutions that will not harm those activities for which the United States is an acknowledged world leader.

In closing Mr. Chairman, we concur that print and electronic collections are crucial to the American economy. Equally important though, are those who use those collections in the process of conducting research, education, and free inquiry. Access to information, public and proprietary, is the foundation for new knowledge and economic innovation. The preservation and continuation of balanced rights and privileges in the electronic environment are essential to the free flow of information, the development of an information infrastructure that serves the public interest, and to the continuation of a viable and productive marketplace. We respectfully submit that insufficient evidence has been submitted by affected industries to warrant the kind of sweeping protections provided by H.R. 2652. This is especially important as there are others in the information industry who do not believe that additional protections are needed, and that such protections could in fact, be problematic. Thorough and complete analyses of the legislation and the potential impact of it on numerous constituencies should be undertaken prior to acting on this legislation. We look forward to working with the Subcommittee on these issues to ensure the appropriate balance between all communities and sectors. ----------------------------------------------------------------------------

1 . Copyright Office, "Report on Legal Protection for Databases," A Report of the Register of Copyrights, August 1977, page 72.