Protection of databases under Copyright would be detrimental to the purpose of U.S. intellectual property protection and to the benefit society derives from such protection.

"In enacting a copyright law Congress must consider... two questions: First, how much will the legislation stimulate the producer and so benefit the public, and, second, how much will the monopoly granted be detrimental to the public?" This balancing test expresses the utilitarian basis of U.S. Copyright contrary to the Author's Rights of civil law countries. U.S. copyright makes the interests of the author subservient to the benefit of the public whereas Author's Rights conveys a basically inalienable right in the author to control and benefit from their work. However the Constitution expresses it best "to 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" is the purpose of copyright and patent. The primary purpose of copyright is to provide an incentive to individuals to invest their time and energy into creating works that provide benefit to the public. How then does the protection of databases fit into this criterion?

Traditionally databases have fallen under the definition of compilation in §101 as " a work formed by the collection and assembling of preexisting materials or of data that are selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of authorship". Protection of compilations via §103 extends only to that which was added by the author of the compilation in no way does it extend to the preexisting material or facts. The seminal case interpreting this standard is Fiest Publications v. Rural Telephone. Fiest explained that "the factual copyright in a compilation is thin...the facts and ideas it exposes are free for the taking". "Facts, whether alone or as part of a compilation, are not original and therefore may not be copyrighted. A factual compilation is eligible for copyright if it features an original selection or arrangement of facts, but the copyright is limited to the particular selection or arrangement. In no event may copyright extend to the facts themselves." The requirement for originality is a facet of the requirement for creativity. Simply laying out data in alphabetical or other straight forward manner does not meet this requirement according to Fiest. But any original or creative method for organizing or selecting a group of data does. So what then would be protected by including databases specifically as copyrightable. It would seem that the only purpose would be to overrule Fiest and allow protection of even the simplest structure and most obvious selection of data. But does this promote a sound public or constitutional policy?

As mentioned the grant of constitutional power to Congress for providing copyright is to "promote the progress of science" but extended protection of databases may impede scientific progress. A database being a collection of data organized so as to make finding and access to the data as easy as possible they are use extensively by scientists and researchers. The scientists and researchers extract the data and use it to develop their ideas and analysis which are at the heart of scientific progress. The data extracted from the databases is often reproduced in theses scientific works so as to explain the concept and allow for peer review and independent testing which is the cornerstone of the scientific process. But what if the data could not be reproduced and distributed? A scientific paper could assert propositions and conclusions but not the data that is the basis of the analysis. Instead there would likely be citations to the data in whatever database the researcher used. Other scientists whether wishing to test the concept or to expand upon it with scientific developments would be forced to access the cited database to discover the underlying data (unless equivalent data is on another database and it can be found with a low search cost). This could greatly increase not only the financial cost but also the amount of time required to do scientific work. Perhaps this use of the data might be considered a fair use and thus sidestep the problem but several factors weigh against fair use.

Determination of fair use is predicated upon several factors laid out in §107. As a whole these factors look at the commercial impact the copying has upon the copyright holder but they do take special consideration if the purpose of the copy "is for nonprofit educational purposes". But much research is of a commercial nature and scientific papers and concepts are often disseminated in for profit publications. So in many cases the nonprofit educational purpose may not apply. Looking then at the commercial impact it appears that each dissemination of the data in an article provides free access to that portion of the database for the reader. It there were only a citation to the source of the data then each reader wanting to look at the data would have to pay that database for the privilege. The inclusion of a set of data from a database in an article would then have a significant negative commercial impact on the database and a positive economic effect on the copier because the article would be more complete and easier for the reader to use it would be of greater value and the consumer would be willing to pay more for it. Thus in many cases copying of data may not be a fair use. The result if databases are made copyrightable would be a chilling effect on the free transfer of ideas and facts. Scientific progress would not be served unless the actual problem is that their is too little incentive to create the proper level of databases unless they are protected by copyright.

Copyright creates an incentive to create works. An economically efficient level of incentive would provide for maximum benefit (social and private) of "creating new works minus both the losses from limiting access and the costs of administering copyright protection". The proper level of incentive is not easily quantifiable but with database the incentive is a two edged sword. Again the crucial element that makes databases different from other subjects of copyright is that they are primarily facts and data that are not original creations of the database. The database simply collected and organized the data either from original sources or from other databases. But if databases become the subject of copyright then any new database will have to either seek out the original source of the fact or negotiate with another database for the right to use the data. Either way the cost of creating a new database will be greatly increased. With higher cost there will be fewer new databases and the database market will become dominated by a few large providers. As there are fewer companies providing databases there organizational structure will become industry standards and increase the barrier to entry.

As we move toward an information economy more people are using electronic databases as not only a part of their profession but in organizing their daily affairs. With a limited number of database providers most individuals will learn and become familiar with the organizational structure of those databases. Since it is the very organizational structure that is being protected any new database would have to have a different organizational structure. But there is a cost to the consumer to learning a new system which inhibits them from learning and using a new database provider. Since databases today enjoy protection for any original organization and selection this cost is currently faced to some extent by new databases but database copyright would exacerbate the problem. Particularly if the elements found to be unoriginal in Fiest are allowed to be copyrighted.

Fiest found that mere organization by alphabetical order was not original enough to warrant copyright. In the same vein would be organizing by date, size, kind, etc. These are the most basic of organizational elements. They are at the very heart of the way we organize our society in large part due to their natural efficacy but also due to the fact that most individuals understand how to use these organizational structures. Thus if a list is alphabetized many more individuals can use it and thus it has greater value. But imagine a world where the first person to organize all the species of plants by alphabetical order gained a monopoly on that style of listing plants. Any subsequent database would have to use another method. But there are no close substitutes for alphabetizing. Alphabetizing is a method that we have all been taught and use to access data nearly every day, it has become a standard of civilization. The cost to society and to the creator of learning multiple new methods of organizing is prohibitively high. Indeed each time a group of data were organized in any manner then for the term of copyright that data could not be organized in the same manner by anyone else, even if they independently gathered the information.

Something is not organized until someone understands the methodology and thus sees it as being organized. This basic tenet militates toward independent creation of databases. A basic element of copyright infringement is that an infringing copy must actually be a copy of the original. Independent creation of an exact duplicate of a copyrighted work is not an infringement. The problem arises with proving that the second creator actually copied the first. The test utilized by the courts is often one of access. If the alleged infringer had very little chance of knowing of the other work then copying from it becomes unlikely. But when a work is quite famous or well known then the alleged infringer is deemed to have likely had access to the original and the second creation may be held to be an infringing copy. Since the very value in a database is not just its useful organization but the understanding by the public of the organization then any accepted and hence valuable method would by definition be widely known. In reality there are but a few basic methods of organizing that make sense to us and these are all well known. There just are not many viable alternatives for methods of expressing information in a database.

A database is not just an organizational structure but it is also the selection and collection of data. Some databases are collections of otherwise copyrighted material but often the data are simply facts which are not copyrightable. As in Fiest the data was names and phone numbers of persons in a geographical area which are non copyrightable facts. But if such a database were given a full copyright (as opposed to the "thin" copyright currently provided for original aspects) then would not the data itself in effect be copyrighted and hence owned by the compiler? What database copyright would do is to give full force to the 'sweat of the brow' doctrine which was specifically excluded from copyright by Fiest . Sweat of the brow reflects the effort put into finding and compiling the data in a database. The overall effect would be to create a new form of intellectual property protection existing in the overlapping penumbras of copyright and patent law.

Imagine a researcher who invents a new process or method of operation that either is described by use of a database structure or operates in the form of a database structure. Which is patentable. Patenting this new invention will provide exclusive rights to the invention for 20 years whereas copyright provides no protection for it at all. All that could be copyrighted would be a particular expression of the structure of the invention but not the invention itself. But if databases were copyrightable then the actual relational structure and method of operating the database would be protected for the term of copyright, the life of the creator plus fifty years. The real significance in this outcome is the prevalence of such database methods of operation they are called computer programs.

What a computer program does is to access a collection of data and manipulate it through various instructions in a functional relationship which is exactly how a database operates. One of these database type operations that is visible to the user is the operational structure of the program as expressed through the interface. In Lotus v. Borland the interface was found to be a non copyrightable method of operation. If the interface were to become copyrightable then different software companies could not provide interfaces which were similar to or emulated the dominant software in a market segment. There is a tremendous cost to a consumer of learning new software so the first they learn will tend to be the one they continue to use. Thus it will be much more difficult for competing software to gain a foothold against a dominant application if they can not provide an interface that allows consumers to minimize their cost of switching to the new software. Furthermore the overall copyright of a software program's database method of operation would provide patent like protection for a program but without the review process a patent is subject o and for the much longer copyright term.

As discussed above the primary purpose of protecting intellectual property (both patent and copyright) in the U.S. is to provide an incentive to the creator or inventor to produce new works and inventions for the benefit of society. But too much incentive can create a cost to society greater than the benefit. Protecting databases through copyright produces just such an inordinate cost. The software and database industries have flourished under existing protections and in a relatively short time span have become some of the largest industries in the U.S. and the world. From this fact alone it appears that no further legal incentive is needed to spur on new software and database creation, the market incentives are working just fine. But taking into account the perverse effects described above the true effect of database protection would be to secure the future of those companies that are currently the dominant players through decreased competition. Accordingly it would be poor policy to protect database under copyright.

© Copyright Jay McKinsey, 1997

http://www.pleasurepoint.com/~jay