Analysis of Database Legislation

From: Patricia Brennan <patricia[_at_]cni.org>
Date: Sat, 14 Sep 1996 16:30:38 -0400 (EDT)

Of possible interest to subscribers of CNI-COPYRIGHT:

        The following is a summary and analysis of H.R. 3531, the 
        database legislation introduced in May 1996 in the House, 
        prepared for ARL by Peter Jaszi. 
	
	The full text is also available at:
	http://arl.cni.org/info/frn/copy/peter.html

     	       **************************

     Some Public Interest Considerations Relating to H.R. 3531 
Database Investment and Intellectual Property Antipiracy Act of 1996

Peter Jaszi
Professor of Law
Washington College of Law,
American University
August 28, 1996

Summary:

     On May 23, 1996 Representative Carlos Moorhead, chair of the House Judiciary Subcommittee on Courts and Intellectual Property, introduced H.R. 3531. This bill would create a strong new form of sui generis intellectual property protection -- an unprecedented right "of its own kind," distinct from and additional to copyright. This would apply to compilations of "works, data, or other materials." Indeed, it proposes a novel reconfiguration of American intellectual property law, which deserves to be carefully scrutinized by libraries, educational institutions, consumers, and others interested in the maintenance of reasonable balance between the interests of proprietors and those of information users.

     As an exercise of the congressional Commerce Power, H.R. 3531 isn't governed by the important limiting language of the constitution's Intellectual Property Clause (Art. I, Sec. 8, cl. 8.) In other words, the new legislation isn't necessarily designed to "Promote the Progress of Science and Useful Arts." To the contrary, its goal appears to be to promote the economic well-being of the information industries. The result is a legislative scheme which defines the rights of owners more broadly, and subjects those rights to fewer exceptions in favor of information consumers, than do traditional intellectual property laws, such as copyright.

     It is none to soon for the scrutiny to begin; although no hearings are scheduled on the bill, and it seems unlikely to be enacted in this session of Congress, it promises to be an important item in the intellectual property agenda of the 105th Congress. Moreover, if the United States delegation to the World Intellectual Property Organization (WIPO) has its way, the approach to database protection outlined in H.R. 3531 may be the basis of major international treaty negotiations in December 1996 -- before any of its merits are debated in this country or many other countries including Canada.      

Key Facts and Findings:

o H.R. 3531, "Database Investment and Intellectual Property Antipiracy Act of 1996" was introduced on May 23, 1996 by Rep. Moorhead (R-CA).

o The bill was introduced to address concerns of the information industry regarding the 1991 Supreme Court decision in Feist v. Rural Telephone, a decision which held that comprehensive collections of facts arranged in conventional formats were not protected under copyright, and could not constitutionally be protected under copyright.

o In 1996, the European Union (EU) approved a directive on the legal protection of databases and it is now in the process of being implemented in domestic law throughout the member countries of the EU.

o H.R. 3531 incorporates many though not all of the features of the EU Directive. Key provisions of the bill include:

  1. Protection would be available for any "database" which was "the result of a qualitatively or quantitatively substantial investment of human technical, financial or other resources in the collection, assembly, verification, organization or presentation of the database contents." Most directories, anthologies, CD-ROM and on-line databases, reference works would be included under this definition.
  2. Databases "made by a governmental entity" would be excluded from protection, but those prepared by private firms or persons using governmentally-generated or collected data could be covered.
  3. The effective term of protection for databases would be potentially perpetual, at least for dynamic compilations in electronic form.

o This would be a new legal regime distinct from copyright law and these protections would "trump" copyright law. Thus the rights the legislation would create would not be subject to "fair use" nor would the first sale doctrine apply.

o H.R. 3551 includes controversial anti-circumvention provisions included in the NII Copyright Protection Act of 1995.

o If enacted and signed into law as introduced, H.R. 3531 would usher in a new era in information commerce -- one in which a "pay-per-use" model would be the norm.

o Provisions in the bill represent an unprecedented erosion of the rights of the American research and education community and would have detrimental impacts on the conduct of research and education activities in this country.

o A full and complete domestic discussion needs to occur on this legislation. If this proposal is included in the December 1996 WIPO Diplomatic Conference, this would represent an end-run around the U.S. Congress and domestic discussions. Many other nations including Canada would be in a similar predicament - no discussion of a crucial intellectual property issue to reflect "local" concerns.

Background

International Dimensions:

     The information industry would argue that the "need" for legislation on this topic can be traced to the 1991 Supreme Court decision in Feist v. Rural Telephone. This decision held that comprehensive collections of facts arranged in conventional formats were not protected under copyright, and could not constitutionally be protected under copyright -- squarely rejecting the notion that the compiler's "sweat of the brow" ever can substitute for the "original authorship" that the statute and the constitutional Copyright Clause require as the condition of copyrightability. While confirming the view of the law that most copyright experts (and a number of courts) had held for some time, and bringing the law of the U.S. into conformity with that of most of the rest of the world, Feist also created some consternation among businesses with large capital investments in massive data compilations, especially (although not exclusively) those in electronic form. Shortly after the decision, there was some discussion of the possibility of enacting "database legislation" as an exercise of the congressional Commerce Power, but many questions were raised about the advisability -- and even the constitutionality -- of such an approach. Rather than pushing for immediate legislation, therefore, the information industry took a "wait and see" approach.

     What they were watching was the progress of a 1992 proposal by the Commission of the European Communities for a "directive" -- that is, a community-wide legislative initiative binding on member countries -- on the legal protection of databases. Such an initiative had been under general consideration in the European Union since 1988 but the Commission's proposal was the first tangible step toward its realization. Extensive interventions by governments and private companies -- both European and American after 1992 -- brought about crucial modifications in the proposal, mainly in the direction of intensifying the protection afforded and curtailing exceptions to it. The directive became final on February 26, 1996. It is now in the process of being implemented in domestic law throughout the member countries of the EU. H.R. 3531 incorporates many -- though not all -- of the features of the EU Directive.

Analysis of H.R. 3531:

         Protection would be available under H.R. 3531 for any "database" (broadly defined) which was "the result of a qualitatively or quantitatively substantial investment of human technical, financial or other resources in the collection, assembly, verification, organization or presentation of the database contents." This would include most (if not all) directories, anthologies, CD-ROM and on-line databases, reference works, and much more. Although databases "made by a governmental entity" would be excluded from protection, those prepared by private firms or persons using governmentally-generated or collected data could be covered. Given the nature of the processes by which data is commercialized, the requirement of the investment of "qualitatively or quantitatively substantial... resources" would not appear to be a significant limitation on the availability of protection for these -- or other -- compilations.

     The effective term of protection for databases under H.R. 3531 would be potentially perpetual, at least for dynamic compilations in electronic form. Technically, protection for any given database would last 25 years from the moment when it was first published or otherwise commercialized; however, any "change of commercial significance" to an existing database (including, for example, current updating of its contents) would constitute the making of a new database, subject to a new 25 year term. Significantly, the rights in such a "new database" would extend not only to the material added or changed, but to its contents as a whole.

     Where protected "databases" were concerned, H.R. 3531 would prohibit any unauthorized person to:

o extract, use or reuse all or a substantial part qualitatively or quantitatively, of the contents of a database if there is a conflict with the database owner's normal exploitation of the database or if it adversely affects the actual or potential market for the database;

o engage systematic extraction, use or reuse of insubstantial parts, qualitatively or quantitatively, of the contents of a database subject that cumulatively conflicts with the database owner's normal exploitation of the database or adversely affects the actual or potential market for the database.

The breadth of these prohibitions creates real risks that they would be enforced not only between competitors in the database industries, but also against end-users and providers of value-added information products (such as indexes and search engines.)

     The bill does not define "substantial part," but its language points to an essentially circular understanding of this crucial term: If there would or could be measurable consumer demand for the data, then it would be unlawful to "extract, use, or reuse" it. A full range of civil penalties (including treble damages, an innovation in intellectual property law along with awards of attorneys' fees) is available against violators; those who act for "commercial advantage" or cause more than $10,000 in "harm" in a given year are liable to fines and imprisonment.

     Other key sections of the legislation provide criminal and/or civil penalties to those who:

o import, manufacture or distribute devices which have the primary purpose of circumventing any system which prevents or inhibits the extraction, use or reuse of the contents of the database; or

o knowingly provide, distribute, alter, database management information that is false.

     These prohibitions track the controversial "anti-circumvention" and "copyright management information" provisions (Sec. 1201 and 1202) of the National Information Infrastructure Copyright Protection Act of 1995. This bill is stalled in the 104th Congress in significant part because of controversy over these very provisions.

     H.R. 3531 provides for only scant exceptions. For example, provisions in the bill do not prohibit independent collection and recompilation of the data contained in a protected database -- provided, of course, that the would- be recompiler has the access and the resources necessary to undertake the project. Another provision would permit the use of "insubstantial" portions of protected databases though the significance of this provision is questionable. First, it is unlikely that any data worth extracting will be deemed "insubstantial;" second, the bill states elsewhere that serial extractions of
"insubstantial" parts of a database may be considered "substantial" in
combination; and third, unlike the EU Database Directive, H.R. 3531 makes even the privilege of "insubstantial" extraction waivable thus encouraging vendors to "contract out" of this limitation.

     Meager as they are, these are the only exceptions to the new right explicitly provided in H.R. 3531. Importantly, there are no exceptions or limitations of broader scope -- or of any scope whatsoever -- implicit in the proposed legislation. Since this is not copyright legislation, the rights the legislation would create would not be subject to "fair use." Nor would the "first sale" doctrine apply -- thus, the vendor of a copy of a database -- on CD- ROM or in book form -- could continue to charge the purchaser license fees for using, or permitting others (students or library patrons, for example) to use that copy for its intended purpose.

     If enacted and signed into law as introduced, H.R. 3531 would usher in a new era in information commerce -- one in which a
"pay-per-use" model would be the norm. Startlingly the bill -- which
defines "database" in the broadest possible terms and permits cumulation of copyright and sui generis protection for qualifying works -- raises the possibility that this model would be applicable to many works (like collated editions of public domain texts or collections of public domain images) which fall within the traditional subject- matter of copyright.

     Compilations of data are essential tools of scholarship, personal research, science and engineering, marketing, and new business development. The movement to a pay-per-use model which H.R. 3531 would bring about would have significant detrimental impacts on users whose access to data has previously been available at no fee or at a reasonable cost. Moreover, implementation of the model is likely to accelerate the already notable tendency toward single-sourcing and niche-marketing of data.

     A familiar example may help to show how enactment of H.R. 3531 could change the information environment. Today, the public information essential to legal practice and scholarship is in the public domain as a matter of copyright law. Some of that data is already only available through private firms which compile decisions, statutes, and regulations, of which the best known (but by no means the only) example is the West Publishing Company. Today, a lawyer or researcher (or her institution) must buy the books or CD-ROMs, or subscribe to the on-line services, in order to have full access to the law. Despite this, the copyright principle that legal protection does not attach to data (including information about the content of the law) operates to check the monopolistic tendencies inherent in this situation. Once access is established, there are no restrictions on how public domain information
"extracted" from print or electronic "databases" can be used, so long as
the "value added" by the compiler (such as annotations) is not taken.

     Under H.R. 3531, the situation would be different. Unlicensed reproduction or other use of the text of even a single judicial decision, if it were deemed "qualitatively substantial," would be enough to trigger liability on the part of the user, potentially on that of the educational institution or library through which the user accessed the database from which the text was extracted. In this new legal environment, further movement toward "pay- per-use" licensing regimes would be inevitable. In turn, the heightened control over the re-use of public domain data which the H.R. 3531 approach would support database publishers offering chronically underfunded government agencies more generous arrangements in exchange for exclusive opportunities to compile the public domain information they generate.

     While promoting the trend toward single-sourcing of government data, the model of protection proposed in H.R. 3531 would likely raise the operating costs of libraries, universities, and schools, and other institutions significantly. Moreover, enactment of H.R. 3531 -- with its sweeping conception of proprietary rights -- might actually slow the
"Progress of Science." Building legal fences around "raw" scientific
data and experimental results could mean less competition among researchers, leading to fewer new discoveries. Of course, the costs associated with the commodification of data may be worth paying, if the effect of the legislation would be to assure greater investment in the creation of new databases. It is worth noting, however, that the absence of such protection has not inhibited significant investments to date.

Conclusion:

     One thing is clear: legislation of this potentially sweeping character should not be enacted without in-depth study and full input from all interested parties. Given the fact that intellectual property legislation has sometimes been propelled through Congress on ideological bases, without full consideration, potentially affected communities should lose no time in focusing on this bill.

     The urgency is all the greater because, as part of an ongoing process to revise and update international agreements in the field of intellectual property. The United States delegation to WIPO recently proposed a new treaty on "Sui Generis Protection of Databases." This proposal contains provisions substantially similar to those of H.R. 3531 and is set to be negotiated and signed in December 1996 during a WIPO Diplomatic Conference. In the U.S., legislation is unlikely to be considered until sometime in 1997. This "end-run" around the domestic legislative process, were it to occur, would create an unfortunate dynamic in which the U.S. Congress might feel compelled to ratify and implement the new "international norm" without substantive consideration of its merits. Many other nations would be in a similar predicament - no discussion of a crucial intellectual property issue to reflect
"local" concerns. If enacted in the U.S., this would represent an
unprecedented erosion of the rights of the American research and education community with predictable, detrimental impact on the conduct of research and education activities in which this country is an international leader.

Peter Jaszi
Prue Adler
8/28/96


Patricia Brennan			   	patricia[_at_]cni.org            
Information Services Coordinator           	202/296-2296 phone       
Association of Research Libraries          	202/872-0884 fax         
21 Dupont Circle                                                    
Washington, DC  20036                                               
___________________________________________________________________
Received on Sat Sep 14 1996 - 20:32:50 GMT

This archive was generated by hypermail 2.2.0 : Mon Mar 26 2007 - 00:35:22 GMT