November 22, 1996
EXPRESS MAIL
Commissioner of Patents and Trademarks
Box 4
Patent and Trademark Office
Washington, D.C. 20231
Attention: Ms. Carmen Guzman Lowrey
Associate Commissioner for Governmental
and International Affairs
Re: Basic Proposal for the Substantive Provisions of the
Treaty on Intellectual Property in Respect of Databases
Dear Ms. Lowrey
I am president of Sports Team Analysis and Tracking Systems, Inc., which does business as STATS, Inc. We are writing in response to the request for comments contained at 61 Fed. Reg. 54159 (Oct. 17, 1996) concerning the Basic Proposal for the Substantive Provisions of the Treaty on Intellectual Property in Respect of Databases, commonly known as the "Proposed Treaty for the Sui Generis Protection of Databases" (the "Proposed Treaty"), which is to be considered at the upcoming Diplomatic Conference sponsored by the World Intellectual Property Organization.
The Proposed Treaty presents a very serious threat to the news reporting and analytical activities of STATS and other publishers and media organizations. It defines the central term "database" far too broadly and lacks the traditional protections for public uses of fact developed under copyright law. The Proposed Treaty, if implemented by United States legislation, will restrict the production of new creative works which are based on analyzing statistics and other facts traditionally in the public domain. We believe that the United States delegation to the Diplomatic Conference should not support the Proposed Treaty.
As a glance at any newspaper will prove, there is a tremendous public thirst for knowledge about sports. STATS is in the business of collecting and providing sports fans and media organizations with analysis and information about the sports, teams, and players they follow. The following are some of STATS's current business activities:
a. STATS publishes 11 annual books on the major sports, some of which are best-sellers in their field. These books include the Major League Handbook and The Scouting Notebook for major league baseball, the Pro Football Handbook and Pro Football Revealed for NFL football, and the Pro Basketball Handbook for NBA basketball. These books contain detailed analysis of statistics on teams and players as well as articles discussing performance and trends for the upcoming season.
b. STATS provides detailed factual information (including real- time scores and statistics) and analysis of sports statistics to many of the most important sports news organizations, including Turner Sports, NBC Sports, the Associated Press, ESPN and Fox Sports. These organizations in turn use STATS' information and analysis in on-air broadcasts to the public.
c. STATS provides its STATS On-Line Service to sports fans who want detailed statistics and current statistical news on professional teams. In addition, we provide a similar service for professionals in the sports industry called STATS Pro-Line. We also provide real-time scores and statistics on major league baseball and NBA basketball games to SportsTrax, a paging device developed by Motorola, Inc.
d. STATS is a content provider to the America Online ("AOL") on-line service. We currently provide analysis and statistical information on NBA basketball, major league baseball, NFL football, and NHL hockey to a site on AOL. We also provide score and statistics updates for games in progress for all major sports on the STATS Scoreboard, which is a "location" within the STATS AOL site. AOL recently issued an award to STATS' Scoreboard as the best application of technology by an AOL content provider and "Member's Choice" award as one of the top sites on AOL.
The "raw material" for STATS' analyses are sports statistics -- the at-bats, shooting percentages, times sacked, goals scored type of figures familiar to every sports fan who reads the box scores. STATS regularly accesses and uses the body of sports statistics which are compiled by the major sports leagues or their outside "official statisticians." STATS also compiles its own body of statistical information through STATS reporters who attend games or who observe public sources such as television and radio broadcasts. Both current and historical sports statistics have been a traditional feature of sports reporting and analysis for decades and have uniformly been considered to be within the public domain. Without free access to this data, companies such as STATS could not provide sports fans with the creative analyses they desire.
Current United States copyright law protects the public's right to access and use sports statistics and other purely factual information, even when contained within an otherwise copyrightable work. Many recent court decisions have held that this information is and should remain in the public domain. STATS' business is dependent on continued free access to that information. Indeed, STATS is currently engaged in litigation with the National Basketball Association in order to defend the principle that once made public, sports statistics (including those revealed within broadcasts of sports games) are in the public domain and may be redisseminated at any time without permission of the sports leagues.
The Proposed Treaty threatens the business of sports analysts such as STATS. For example, the statistics which STATS uses as material for its analyses would acquire the status of "databases." Were the Proposed Treaty in effect, STATS' current activities with sports statistics would be very likely to violate the sui generis protection granted to database owners. The sports leagues would argue that their statistical compilations, including individual game scores and statistics, qualify for sui generis protection, either because they are compiled through their or their statisticians' "sweat of the brow", or because the leagues make substantial investments in the games that produce the statistics. Although STATS compiles its own statistics and analysis, it also makes use of large portions of seasonal and individual game statistics compiled by the leagues. Under the Proposed Treaty, this activity is likely to be considered prohibited "extraction" or "utilization" of a database. Since there is no provision for a fair use doctrine under the Proposed Treaty (see below), STATS would be unable to defend on the grounds that its use of league sports statistics is transformative," that is, used to create wholly new works not directly competing with the database itself.
Unless licensed to work with these statistics by the sports leagues, STATS (and any other entity involved in sports analysis) could well be put out of business under a sui generis regime. This is no theoretical risk. Today sports leagues make substantial sums from exclusively licensing their intellectual property for use within defined markets. Under a sui generis scheme, a league could similarly license chosen publishers, giving them exclusive rights to use the league's database of statistics in publishing books of analysis. This arrangement could effectively bar others from bringing out competitive books.
The same problems would affect financial analysts and writers. For example, a stock exchange could obtain sui generis protection for tables of trades and statistics about prices; a company could claim the same rights in its annual financial statements and reports. Here again, the power to license could easily become the power to suppress, for under the Proposed Treaty a rightsholder could simply deny a license to a financial analyst whose studies were less than flattering.
Our concern is not solely for STATS' business. The proposed sui generis protection would radically transform settled law and would cause serious erosion of the public domain. At least four aspects of the Proposed Treaty are particularly troubling.
First, the key term "database" is defined far too loosely. The definition encompasses a substantial number of works which are traditionally subject to the limited protections of copyright law. Any literary, musical, or audiovisual work will benefit from sui generis protection, as long as it is contained within a collected group of such works organized in some "systematic" manner and is capable of being individually accessed. See Memorandum of the Chairman of Committee of Experts, Note 2.02 to Proposed Treaty, Article 2. Since the sui generis protection is cumulative of copyright, see Proposed Treaty, Articles 1(3), 12, a great many copyrighted works will also be considered "databases" and protected as such. Copyright holders will naturally prefer to use broad sui generis protection, rather than limited copyright law, to advance their interests. A century or more of efforts by Congress and the courts to balance the rights of copyright holders and users could be swallowed up by a system of sui generis protection which one-sidedly favors database owners.
Second, the proposed treaty contains no parallel to the fair use doctrine in copyright.[1] Under fair use, many uses of copyrighted material are permitted where (1) the purpose and character of the use fall into traditionally permitted categories, such as comment, criticism, research and news reporting, or where the use is "transformative" in nature, (2) the work is factual rather than creative, (3) the amount and substantiality of the use is not great, and (4) the effect on the market for the original is not great. 17 U.S.C. Sec 107. Other commentators have suggested that use of a copyrighted work is fair where the second user's work performs a different function than the plaintiff's. 3 M. and D. Nimmer, Nimmer on Copyright, Sec 13.05[B][1].
In contrast, the only defense available to the user of a database under the Proposed Treaty would be the "insubstantiality" of the use. Whether the use was in the interest of the public or functionally distinct from the method in which the database is now used would be irrelevant. Almost any use of facts within a systematic compilation of facts can be deemed to take a "substantial part" of the compilation, as long it is of "qualitiative or quantitative significance to the value of the database." Proposed Treaty, Article 2(v). The sui generis proposal thus could have the paradoxical effect of providing greater protection for factual databases than copyright now affords to highly creative works. Moreover, in almost every litigation, the question of whether a use was "substantial" would be one of fact, resolvable only at trial after substantial litigation cost to the user. This expense and risk of litigation would tend to discourage even insubstantial uses of facts from compilations and databases.
Third, the manner of adoption of the Proposed Treaty ignores our country's tradition of altering intellectual property law only after careful legislative attention to the effect of changes on the balance between private and public interests. For example, the revisions to the copyright law that became the Copyright Act of 1976 were enacted only after extensive hearings and reports over a period of eleven years. In contrast, the Proposed Treaty would be submitted for ratification only to the Senate, even though matters of copyright law are traditionally within the consideration of the House of Representatives as well. Moreover, ratification would compel ratifying states to adopt national legislation implementing the underlying sui generis protection. Subsequent congressional hearings on implementing legislation could not undo the fundamental flaws of the sui generis concept.
Fourth, the proposed sui generis protection poses serious problems of constitutionality. The Copyright Clause to the Constitution, article 1, section 8, cl. 8, empowers the Congress, "[t]o promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries." A basic assumption of the Copyright Clause is that only original works of authors are protected. In contrast, facts are not original, but discovered; copyright cannot extend to facts themselves, although it may cover their arrangement and selection where original. Feist v. Rural Telephone Service Co., Inc., 499 U.S. 340, 347-48 (1991). The Copyright Clause may well limit Congress's power to grant monopoly rights in facts, a limitation which would be breached by adoption of sui generis protection.
Even more serious is the clash between the Proposed Treaty and the First Amendment. The private property rights granted by copyright law are made compatible with First Amendment protections of freedom of speech and press through two concepts: the "fact-expression" doctrine, which permits free use of any facts contained within a work of authorship (see Harper & Row Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 556 (1985)); and the fair use doctrine, which allows even protected expression to be taken in certain circumstances. In contrast, the Proposed Treaty contains neither of these limitations. It would permit nearly unlimited protection for facts (as long as these are arranged in a "database") and contains no exception for such socially desirable uses of factual information as comment, criticism, news reporting, or research.
As noted above, adoption of the Proposed Treaty would seriously erode the public's traditional right to access and use facts. It would limit or destroy socially useful activities such as the statistical analysis which is the heart of STATS' business. Before considering such a radical change, one should demand irrefutable evidence of a compelling need for broad sui generis protection for databases. No such evidence has been offered. In particular, there is no proof that database publishers have been imperiled by the Supreme Court's rejection of the "sweat of the brow" theory for copyright protection of facts in Feist. Later case law shows that the copyright law still provides substantial protection to database owners who put even minimal originality into their selection or organization of facts. Nor is there any empirical evidence that the absence of sui generis protection under United States law has depressed the value of databases or diminished the incentive of database compilers to pursue their business.
We also have personal experience with this issue, because STATS is itself a "database" proprietor. Our detailed statistical collections and analyses, which go well beyond any league-compiled statistics, are considered authoritative among sports fans and within the sports industry. We spend substantial sums to develop these compilations and analyses, and they form the core of our business. Nonetheless, we have always believed that current United States and state law fully protects our rights in those statistical collections and analyses against misappropriation, and we have never felt the need for a new form of protection. The absence of sui generis protection has not affected our incentive to continue and expand our database business. In short, the Proposed Treaty provides an unwise -- indeed, dangerous -- solution to a non- existent problem.
Databases are adequately protected under existing United States law. In contrast, adoption of the Proposed Treaty would disturb the traditional balance between private and public interests developed over more than a century of copyright law. In our view, the many arguments against adoption of sui generis protection for databases are compelling. For the reasons set forth in this letter, we urge that the United States delegation to the Diplomatic Conference withhold support for the Proposed Treaty.
Very truly yours,
SPORTS TEAM ANALYSIS AND TRACKING SYSTEMS, INC.
By:_________________________________
John Dewan
President
[1] Some limitations on sui generis protection may be adopted in national legislation, but only where this does not "unreasonably prejudice the legitimate interests of the rightholder." Article 5(1). However, H.R. 3531, "The Database Investment and Intellectual Property Antipiracy Act of 1996," introduced in anticipation of the Proposed Treaty, contained no fair use exception.