©Copyright: Kenneth L. Horton, 1996.
II. Copyright Law Background
A. International Copyright Regimes
i. Philosophical Underpinnings to Copyright
ii. A Copyright Model
iii. Transnational Issues in Copyright
B. The E.C.: A Third Tension and the Impetus to Harmonization
III. Directive on the Legal Protection of Computer Programs
A. Works Receiving Protection
B. Determining the Appropriate Rightholder
C. The Granted Bundle of Rights
D. Exceptions to Rightholders' Exclusive Rights
E. Other Provisions--Enforcement and Protection
IV. Challenges of the Internet
V. The Software Copyright Directive under the Internet
A. Impact on the Internal Market
B. Copyright & Competition Among Manufacturers
C. Private Copying and Piracy Over the Internet
i. Formal Coverage Within the Community
ii. Formal Coverage: Minimizing Impact From Outside the E.C.
iii. The Quandary of Enforcement
iv. Addressing Enforcement: A Range of Options
VI. Conclusion: Going Forward Alone
Statutory, Legislative, and Other Primary Authority
The Internet is generating an enormous amount of "hype." Whereas just a few years ago, the networks of global research and science computers were hardly a topic of conversation, they are now the subject of rapidly growing interest and speculation . Usage is exploding. The Internet is said to be doubling in size every year and studies have projected over 200 million users by the year 2000. New possibilities are emerging for the way we conduct all types of activities. Along with this growth, companies in some markets are moving to offer new services and struggling to create business models that work in the new environment.
These forces have been seen as a "ëdigital revolutioní...triggering structural changes comparable to last century's industrial revolution with the corresponding high economic stakes." As society moves into the new and uncharted region of cyberspace, it is becoming apparent that legal regimes ranging from tariffs to privacy to intellectual property rights will be affected.
Perhaps no area of law has sparked more comment on this matter recently than copyright. The impact of the information revolution on copyright has been termed "undoubtedly the most prominent, and arguably the most significant"issue in copyright law in the past decade. For many, the new environment has been seen as an insurmountable barrier to copyright doctrine. It has been suggested that "[t]he global information superhighway challenges virtually all the premises of copyright law.î Similarly, another author has argued "[i]ntellectual property law cannot be patched, retrofitted, or expanded to contain digitized expression...We will need to develop an entirely new set of methods as befits this entirely new set of circumstances.î At the same time, dissenters have suggested the new environment poses few insurmountable challenges.
The question of the Internetís impact on copyright is of great interest, and has sparked full governmental studies in the European Community and United States, as well as Australia, Japan, and Canada. Last July, the E.C. released a Green Paper posing questions about the copyright implications of the information society. In the U.S., a White Paper was released in September, presenting an assessment of the same issues and making legislative recommendations based on over two years of work. Both studies proceed from the proposition that content will drive the development of new information infrastructures and that fears of ineffective copyright protection would hinder the growth of networks by causing authors to withhold their material. Today, much of the content on the Internet is non-profit material. In order to generate the kind of large-scale investment that will build the backbone of these new networks, this will have to change. The goal of adapting copyright is, therefore, to overcome the perceived challenges to incentivizing the development of content on the network.
Does the Internet really raise significant challenges to copyright? If so, are they insurmountable? This paper will attempt to answer these questions in one area of E.C. law--the market for computer software. The central question posed is how the Directive on the Legal Protection of Computer Programs works in the networked environment, and how, if needed, it might be improved?
The problem of ensuring the protection of computer software in the information society presents the intersection of two problems copyright law faces as it tries to keep up with technological progress. First, copyright has had problems adapting to cover computer software. Second, global networks present their own problems in applying copyright doctrine. Providing for copyright protection in the multi-national environment of the E.C. further complicates this set of problems. How does the Software Directive respond to this triple challenge? The thesis of this paper is that the Software Directive provides an appropriate fundamental structure to protect software in the information society, and that, with some interpretive clarifications and modifications, the Directive can adequately redress the imbalance between various groups that the Internet creates.
Before developing these conclusions, it is helpful to review some necessary copyright background. Because the paper's inquiry merges a number of questions of copyright, this section covers a series of issues, including the basic philosophy of copyright, questions raised in copyrighting software, and problems of international copyright. Then the grounds to harmonize copyright protection in the E.C. are addressed, along with the hurdles such efforts entail. Next, the Software Directive is explained, using the elements of a basic copyright model to categorize its provisions and determine that it satisfactorily achieved both single market integration and appropriate balancing of various copyright interests. Subsequently, the paper turns to assess the challenges which the Internet does or will raise. Specific reference is made to the transnational, open, and distributed characteristics of global networks and how these characteristics can impact the system. The final stage in the analysis is integrating the various components to evaluate whether the Software Directive continues to fulfill its twin goals--promoting the internal market and balancing various copyright interests--in this new context. This analysis proceeds in three parts. First, it is determined that the internal market for software will continue to function well in the new environment and that, in fact, Internet delivery of software may provide a model for other product and service areas. Secondly, those software copyright questions not affected by the Internet are identified, noting that many of the most contentious issues in software copyright--questions arising from conflict among manufacturers--need not be reopened. Third, the problem of private copying is discussed in further detail from both a formalistic and a practical reference point. As a formalistic matter, with clarification of the reproduction right's application to RAM loading the Directive can continue to legally cover any copying taking place within the E.C. As a practical matter, however, enforcement problems upset the balance between rightholders and users and require changes in the system. Various options are considered. Finally, looking forward, the paper identifies benefits from breaking software out from much of the other discussion on information society copyright and suggests a media-specific approach to considering changes. This discussion reveals some of the issues that will be raised in addressing Internet copyright in other product areas and highlights the need to maintain awareness of the reallocation of privileges which this process is destined to cause.
In order to better understand the Software Directive and to assess its viability for continuing to function appropriately in the developing global information environment, it is necessary to have a grasp of fundamental concepts in copyright law. There are three aspects of copyright law which need to be understood in grappling with E.C. protection of intellectual property rights in software made available on the Internet. The first is understanding how the content matter of software can be brought within a copyright rubric and given protection. This is a question of basic copyright doctrine. The second key element is a foundation in the protections which are granted or lost when copyright matter travels transnationally. This has implications for software traveling among Member States, as well as software traveling to and from other nations on the Internet. The final is understanding what additional concerns the E.C. has for promulgating copyright regulation under E.C. law.
Copyright, in its various forms, seeks to balance the interests of authors in their intellectual work product with the interests of society in broad dissemination of ideas. Originating with Sixteenth and Seventeenth Century privileges granted to publishers by the French and British monarchies, copyright evolved into two main traditions--common law copyright and continental droit díauteur. The authorís rights model is applied in some form by most of the continental members of the E.C. While various national approaches of continental countries differ in specifics, such as perpetuality of rights, they share a basic focus which is centered around a natural rights idea of the individual retaining the right to control something he or she has created. Such control governs not only economic exploitation of the works (such as reproducing and distributing), but also "moral rights" elements, e.g. deciding whether or not your creation should be published, claiming authorship, and maintaining the work's integrity. Significantly, the droit díauteur philosophy also leads many countries to make certain rights inalienable by the author and to establish systems where rights to copyright protection can only evolve in an individual person, as opposed to, for example, a corporation. Another significant, and related, result of the droit díauteur approach is philosophical uncertainty about how far the doctrine of copyright can stretch when faced with new technologies. Consequently, materials such as broadcasts, which seem logical to protect but for which an individual author creating original work is difficult to identify, have been protected in many countries under "neighboring rights."
The second European copyright tradition--the common law copyright regime--is the Anglo-American tradition applied in the law of the U.K. and U.S. The focus of this regime is a more utilitarian approach based on economic reality. Harkening back to incentivizing publishers, the common law copyright is based on the practical need to maintain incentives for individuals and businesses to undertake new investment which will result in societal advancement. For example, the Congressional power to legislate in the copyright field in the United States is derived from the Constitutional grant "[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.î Under this model, advances in technology and collective business forms have not been difficult to reconcile with the underlying premises of the copyright regime.
Reconciling the differences in approach between these two systems is one of the challenges posed in developing a harmonized multi-party intellectual property approach for any new technology. As will be seen, this challenge applies to both developing copyright law for the E.C., as well as to coming up with international agreements on copyright. The differences between the two systems should not, however, be overemphasized. They share a number of similarities which are also relevant for understanding how copyright is applied in the software context. These similarities form the foundations of a general copyright model.
The basic copyright model identifies certain works as worthy of copyright, categorizes them as property, and then grants a certain bundle of rights to an appropriate rightsholder, subject to restrictions and exceptions. The U.S. Copyright statute is representative of the key elements involved in granting a copyright. It provides for protection of "original works of authorship fixed in any tangible medium of expression...." At the same time, the statute does not extend copyright "to any idea, procedure, process, system, method of operation, concept, principle, or discovery.î This basic formulation highlights the elements which raise issues when copyright is applied to new technologies and media.
In order for a work to receive copyright protection, it must meet some minimum hurdle of originality and creativity. Yet, different national systems frequently diverge on both the principle of how much originality is required and on the application of the originality principle to specific items such as computer software. Often, an authorís rights approach establishes a higher hurdle for creativity than a common law copyright approach. Similarly, while the various systems all require the work to be "embodied in perceptible form,"only some require fixation in a lasting, tangible form. Copyright doctrine also protects the central expression, but provides no protection for the underlying idea or function. This "idea / expression"distinction is critical with a functional product like software.
Assuming that a work is found to be an expression which is perceptible and sufficiently original to merit copyright, the rightholder receives a bundle of rights. While there is some variance from country to country, the basic economic rights include the right to control reproduction or copying of the work, its distribution, the sale of derivative works, and dissemination to the public. These rights may be subject to limitations, such as a "fair use" exception in the U.S. or permitted private copying exceptions in Europe.
One important element of this "limited monopoly" is the "first sale" principle, known as "exhaustion" in many European countries. Under the exhaustion principle, once an author has sold a copy of a work, he or she loses the exclusive distribution right with respect to that work. In other words, the purchaser may resell that copy of the work. Importantly, although the author has consented to giving his or her distribution right in that specific copy of the work, he or she has not waived other rights, such as the reproduction right. The concept of exhaustion should be kept in mind when considering creating E.C.-wide regulation, as well as in addressing some of the potential problems created by the Internet.
Copyright is a creature of national law. As such, there is no such thing as an international copyright. Rather, through a series of treaties and agreements, most notably the Treaty of Berne and the recent TRIPS agreement, nations undertake to set a minimum level of protection and accord authors from other signatory nations national treatment. In other words, a French authorís software program should be granted the same level of copyright protection in the U.S. as a U.S. authorís software program receives.
Rights under copyright therefore adhere to a principle of territoriality. Infringing conduct leads to a cause of action under the laws of the site of the infringement, and where that conduct crosses borders, it may lead to different causes of action under the national laws of each of the territories where it took place. The most difficult problems of territoriality arise where conduct may be legal in one state and illegal in another. In combination with this, the principle of exhaustion, which allows the resale of a work once the original author has already sold it, sometime applies internationally and sometimes applies only nationally. As will be seen in the next section, international application of exhaustion has the tendency to promote free trade, while national exhaustion more strongly preserves intellectual property rights.
Developing copyright protection in the E.C. introduces a further element into the already difficult to achieve copyright balance. In addition to balancing the protection of authors' rights in their works with the societal interests in the dissemination of ideas, the E.C.'s approach to copyright must also take into account the impact of intellectual property regulation on the establishment of a single market. Thus, the broader goals of European integration complicate the task of striking a balance and have led to the impetus to harmonize certain critical areas of copyright across the E.C. The Software Directive was the first milestone.
Promulgating intellectual property regulation traditionally had been left to the Member States. Up until the mid-to-late-1980's, the primary concern shown by the E.C. seemed to be ensuring that copyright did not function as a restriction on trade or hindrance to progress toward the internal market. However, in its 1985 White Paper, the Commission recognized the need to harmonize European protection of intellectual property; computer software, semi-conductors, and biotechnology products were cited as high technology areas where harmonization was urgent for completing the internal market. With the 1988 publication of the Commissionís Green Paper "Copyright and the Challenge of Technology--Copyright Issues Requiring Immediate Actionî a new era in E.C. involvement in copyright began. Work on the Software Directive was begun.
One of the explicit goals underlying an overhaul of the intellectual property system protecting software was to try to aid the development of the Community's software industry. The disparate state of intellectual protection for software may well have been a partial contributor to the industryís lagging position relative to its U.S. and Japanese counterparts. In any case, software copyright regulation within the Community raised a number of legal issues which impeded the functioning of a single market or compromised property rights.
Having twelve separate, and on some points widely differing, regimes addressing software, sparked the potential for significant distortions in trade. For one, it was not certain at that time whether copyright would be the approach chosen by all the member states for protecting software. In 1988, only France, Germany, Spain, and the United Kingdom actually had legislation which explicitly recognized computer programs as copyright works under the Berne convention. Some of the other member states were in the process of enacting copyright legislation or discussing copyright as a model for such protection. Secondly, the difference in regimes created significant difficulties under the Community's free trade provisions, specifically Article 30 and 36. The prospect of many of the Member States creating their own new legislative framework to copyright software had the potential to make these difficulties worse.
In the Treaty's chapter on eliminating quantitative restrictions between Member States, Articles 30 through 37 establish the Community's basis for reviewing intellectual property regimes which might affect trade. Articles 30 and 36 are the most relevant here and present a challenge in balancing the competing interests of national intellectual property law with Community trade law. Article 30 provides a potential barrier to Member Statesí copyright and other intellectual property legislation. A national copyright system which determines a software product to be infringing and restricts its import from one Member State to another could be considered a quantitative restriction or measure having equivalent effect. If this were the case, the supremacy of Community law would prohibit the restriction. Resultingly, the author's property right would be compromised. However, Article 36 provides a counterbalance to the free trade provision of Article 30. In permitting prohibitions or restrictions justified on the grounds of "the protection of industrial and commercial property" this article opens the door for national copyright protection provisions to be maintained, provided that they are not arbitrary.
The problem created by the interaction of these two Treaty provisions with respect to intellectual property is that it repeatedly places the Court of Justice in the position of having to strike an imperfect balance between promoting a market without internal borders on the one hand and protecting intellectual property on the other. The source of this problem is the intersection of the E.C.'s free trade ideal with two principles of copyright: territoriality and exhaustion. Recall that an author is entitled to a series of national copyrights in his or her work based on local laws and the Treaty of Berne, but has no standard international copyright. Rather, various national copyright regimes often differ in terms of what is copyrightable and what property rights are granted. Recall also that once an author has sold a copy of his work, subsequent purchasers have the right to resell that copy. If an author has sold a work into one Member State, should he have the right to control its sale into another Member State? If a product is not covered by copyright protection in a particular Member State and freely traded therein, should a different Member State where there is copyright protection be able to restrict the free movement of goods between states? How can these issues be decided in a manner which is consistent with Community trade principles, and yet maintains a high level of protection for authors?
The short answer is that, in a system where divergent copyright regulations are passed at the national level, these problems cannot be solved in a manner entirely consistent with both principles. The Community case law in this area illustrates a pragmatic but impossible balance. More importantly, it underscored the need for harmonization.
The philosophical approach taken by the Court of Justice to allow E.C. intervention into intellectual property rights highlights the problem. In an early intellectual property judgment in the Park, Davis case, the Court drew a distinction between the "existence" of an intellectual property right, which is not affected by the Treaty, and the "exercise" of those rights, which the Treaty could alter. Although the Parke, Davis decision took place in the context of Articles 85 and 86, the same distinction was drawn with respect to Articles 30 and 36 in the Deutsche Grammophon case. Plainly, a property right which "exists"but which an author cannot exercise, does not provide a high level of incentive or protection for an author. On the other hand, this solution may be necessary from the perspective of promoting the internal market.
Through a series of cases under Articles 30 and 36, the Court developed a jurisprudence implicating when national copyrights could be exercised and when they could not. Deutsche Grammophon was one of the major stepping stones along this development. The case involved a classic problem of parallel importing. The German plaintiff produced records and sold them in Germany, while granting exclusive licenses for distribution to other companies through its subsidiaries in other Member States. Some of the records sold by Deutsche Grammophon's French subsidiary in France were eventually resold by third parties into Germany for sale by Metro, a Hamburg retailer. Disagreeing with Deutsche Grammophon's argument that there had been no first sale in Germany, and that therefore the distribution of the records in Germany was in violation of their rights under the German copyright law, the Court of Justice held that it would violate the free movement provisions of the Treaty to permit Deutsche Grammophon to maintain its exclusive distribution right in Germany merely because its first sale of products took place in a different Member State. Essentially, the court made the entire E.C. the relevant market for determining exhaustion. This doctrine was further emphasized in Dansk Supermarket A/S v A/S Imerco, when the Court stated that:
Articles 30 and 36 of the EEC Treaty must be interpreted to mean that the judicial authorities of a Member State may not prohibit, on the basis of a copyright or of a trade-mark, the marketing on the territory of that product to which one of those rights applies if that product has been lawfully marketed on the territory of another Member State by the proprietor of such rights with his consent.
The issue which consequently arose is whether this principle operates to effectively gut rights other than the distribution right when products are marketed in two states with other differences in terms and conditions of copyright.
EMI Electrola GmbH v. Patricia, one of the cases which addressed other economic rights, demonstrated the other side of the coin from Deutsche Grammophon. In EMI, the plaintiff was again a German record distributor seeking to enforce its German distribution rights against products entering from another Member State. The court determined that the German distributor could exercise its distribution rights because the ability to copy and reexport was not due to EMIís consent, but rather due to the expiry of the duration of protection outside Germany. Although this decision is formally based on a "lack of consent,"the real issue operating here is protecting property in the face of differing terms of protection. The same conflict has the potential to arise from any host of differences in copyright regimes. Different terms of protection, varying compulsory license systems and divergent criteria for qualifying for copyright coverage could all set the economic preconditions for profitable parallel importing without the rightholder's consenting differential pricing. Permitting the maintenance of rights, as in EMI, clearly operates to create distortions in intra-Community trade. Yet, to hold the line against a race to the bottom in effective copyright standards, Article 36 must be given some bite.
The solution to what the Court recognized in EMI Electrola as "the present state of Community law, which is characterized by a lack of harmonization or approximation of legislation governing the protection of literary and artistic property" is harmonization. Based on the Community power under Articles 100 and 100a, the political organs undertook to approximate intellectual property protection of software in the Community.
The E.C. was faced with a double challenge when it undertook to harmonize the intellectual protection of computer programs. Foremost, because of softwareís functional nature, applying copyright protection to computer programs creates conceptual difficulties. Secondly, these difficulties are compounded when attempting to harmonize them among countries which do not share a single author's rights or common law copyright approach. Despite these challenges, the Software Directive is a workable and very practical solution to the copyright problem. It creates a framework to provide strong anti-piracy protection for software. At the same time, it strikes a balance between protecting rightholdersí investment in creating software and permitting activities which will foster advancement and market competition. It should be noted, however, that some issues which might have been harmonized were left open and could create uncertainty in the future.
The Software Directive represents a framework for protecting software which overcame many conceptual hurdles and created a workable copyright system for the E.C. It is significant for the prospects of improved global copyright coordination in dealing with the Internet that the Software Directive's premises and approach are substantially similar to those proposed by the U.S.'s National Commission on New Technical Uses of Copyrighted Works (CONTU)in 1978 and enacted by the U.S. Congress into American copyright law. The one major area of difference in approach is the E.C.ís granting of a specific exceptions to exclusive rights, particularly decompilation, rather than creating a general fair use exception.
The Directive might best be outlined in terms of the basic copyright model set out at page 12 supra. First, the Directive in Article 1(1) specifies those works which should receive copyright protection. Computer programs and preparatory design material shall receive copyright protection as literary works. Consistent with the classic "idea / expression"dichotomy, the Directive also states clearly that "expression in any form of a computer program"shall be protected while the underlying "ideas and principles"are not protected by copyright. This dichotomy is further reinforced in three of the Directive's recitals. Finally, and critically, Article 1(3) establishes that originality is only to be measured in terms of software being the "author's own intellectual creation" and that "[n]o other criteria shall be applied to determine eligibility for protection." Again, this is underscored in the recitals where it is detailed that "no tests as to the qualitative or aesthetic merits of the program should be applied.î
While most of the issues raised by the Directive will be discussed in terms of their viability for the Internet below, the significance of Article 1 merits mention here. The basic concepts established by Article 1 were central for overcoming the Community's most pressing harmonization problems with software. Most critically, by addressing the originality question, the Directive overcame what was probably the leading cause of disparate treatment of software--varying originality requirements among Member States. Prior to the Directive, new software products may have been accorded full copyright protection in some countries and no protection in others. This was largely a result of the differences between some Member States applying an author's rights approach and others using a common law copyright approach. At the extreme, the German system, following decisions in the Inkasso-programm case and the Betriebsystem case, applied the qualitative criteria of significantly surpassing the average ability of a programmer to determining originality. Such a requirement denied copyright protection to an estimated 85% of the software programs in Germany. At the other extreme, originality in the United Kingdom was based strictly on a program being the authorís own work, meaning that the author had not copied the expression. The result forced the classic issue of balancing between authorís rights and a single market, detailed in the discussion of the Article 30 / 36 conflict above. By placing the whole Community under a low originality, common-law copyright approach, the Directive unifies the Community under a high level of protection, achieving both strong support of author's property rights and a harmonized internal market in terms of having the same programs protected in each market.
Article 1 takes two further steps--by way of omissions--which may prove to be of importance in the future. First, the Directive expressly avoids providing a definition of "program." This was an intentional step to avoid the Directive becoming outdated as technology advanced. Second, the Directive also avoids providing a guide for identifying the borderline between idea and expression. Although expression is not limited merely to protection of the literal code, at some point of abstraction it becomes an idea. The Directive has left this borderline to be fixed by the courts. With little European caselaw on this subject matter, this omission has the potential to create protection disharmonies. Commentators, have differed on whether E.C. courts will borrow from American jurisprudence to solve this conundrum. Because it is unclear what source courts will use to find a principle for resolving this problem, it is even less clear what the actual doctrinal resolution should be in the particular case of any piece of software. While there is something to be said for the wisdom of having this question settled through concrete cases as technology evolves, uncertainty may become a problem.
If this area can be settled through an Article 77 reference procedure or receive consistent treatment by Member State courts, there will not be a problem. However, as the recent Lotus v. Borland decision illustrates, this is a conceptual issue rife with confusion and with the potential to meet widely varying treatment by Member State courts, at least until a case is finally referred to the Court of Justice. Although changes brought on by the Internet will not make these problems worse, the question is important in highlighting potential disharmonies which might have been avoided through action by the Commission.
After the criteria for which works will receive protection are set out Article 1, Articles 2 and 3 address who will be the holder of rights with respect to the property. Again, the Directive resolves one issue of conflict in Member State approaches and leaves another open for Member State determination. However, here one should expect that any actual differences among Member States will have minimal effect.
One of the differences between author's rights systems and common law copyright systems is in permitting legal persons to be designated as a rightholder for copyright. Additionally, various civil law regimes take differing approaches to determining authorship of a collective work. In an environment where most products are produced by teams working for an employer, these differences create varying mixes of rights among employers and employees, depending on what country a program was produced in. The directive provides for a legal person to be designated as the rightholder, or for collective work to be attributed to a person, where these arrangements are permitted or recognized by the Member State. The flip side to this proposition is that states which do not recognize these arrangements may continue not to recognize them. Thus, the directive does not harmonize the questions of legal personality and collective ownership. Further, where a program is created by a group of natural persons jointly, the Directive provides for joint ownership of the exclusive rights.A HREF="footnts.html#f71">
These differences should be of diminished importance in practice because the Directive grants an employer exclusive right to exercise all economic rights in a program unless the parties agree otherwise by contract.A HREF="footnts.html#f72"> Thus, the presumption is that full control over exploitation of the program will inure to the employer, although the employee may retain moral rights in certain jurisdictions.
Holders of software copyrights are granted a broad set of exclusive rights in Article 4, which are subject to specifically tailored exceptions drafted in Articles 5 and 6. As many of the issues raised by the Internet involve technological changes in reproducing and transmitting digital data, these provisions are the subject of greatest concern in the information society.
The rightholder's bundle of exclusive rights includes:
permanent or temporary reproduction of a computer program by any means and in any form, in part or in whole. Insofar as loading, displaying, running, transmission or storage of the computer program necessitates such reproduction, such acts shall be subject to authorisation by the rightholder;
translation, adaptation, arrangement and any other alteration of a computer program and the reproduction of the results thereof...;
any form of distribution to the public, including the rentalA HREF="footnts.html#f73">, of the original computer program, or of copies thereof.A HREF="footnts.html#f74">
In addition, applying the exhaustion principle as in Deutsche Grammophon, exhaustion of the copy of a program is applied Community-wide by a first sale of that copy in the Community with the consent of the rightholder. The application and enforcement of these provisions are challenged by the nature of activity on the Internet and will be discussed below.
The Directive creates two types of exceptions to the exclusive rights granted in Article 4. One set of provisions entitles the user to undertake activities which will allow him or her to appropriately utilize the program. In addition, there is a set of provisions which permit other software developers to undertake a set of narrowly defined acts in order to put other software products on the market.
Narrow exemptions are included to guarantee user's rights. "Lawful acquirers" are permitted to undertake the restricted acts of reproduction, translation, adaptation, arrangement, and alteration when these activities are "necessary for the use of the computer program...in accordance with its intended purpose, including for error correction."A HREF="footnts.html#f75"> Furthermore, a person "having a right to use the computer program"may make a back-up copy "insofar as it is necessary for that use.îA HREF="footnts.html#f76">
Specific exceptions to the rightholderís rights are also established for other companies in the industry that seek to market software products. As a matter of clarification, Article 5(3) specifies that one may "observe, study or test"the programís functioning to determine its ideas and principles while in the course of entitled use such as running, loading, and transmitting. A much more controversial right is granted in the seriously contested decompilation article. Article 6 expressly permits decompilation of the program for the singular purpose of determining interoperability of programs if such decompilation is necessary to accessing the interface information.A HREF="footnts.html#f77"> What is not entirely clear from Article 6 is whether a company may use decompilation to facilitate interoperability with a competing product.
Of the Directive's remaining provisions, Article 7, which provides for special measures of protection, is of particular interest in the emerging information environment. This article directs member states to provide for "appropriate remedies" against three types of infringements: putting an infringing copy of a program into circulation, possessing an infringing copy for commercial purposes, and putting into circulation, or possessing for commercial purposes, any means whose sole intended purposes is to remove or circumvent copy protection. Article 7 also provides for seizure of both infringing copies and of the means to circumvent copy protection. Interestingly, the NII White Paper in the U.S. has suggested that changes in the National Information Infrastructure mandate adding a prohibition on devices which have the sole purpose of removing technological protections against copying digital copyrighted materials.
The Software Directive is a good piece of legislation. Judged by a number of criteria which might be used to evaluate the E.C.'s system of software copyright, the Directive functions well. Software is provided a high level of protection while at the same time specific and limited user and market needs are accommodated. With few exceptions, areas of conflict between Member States (and copyright systems) are harmonized in a manner which is practical and attuned to the realities of the software market. Yet will this legislation survive the advent of the Internet? The rest of this paper devotes itself to analyzing the Software Directive's longevity considering the emergence of the information society.
The Internet and other "information superhighways" like it are global networks of loosely interconnected computers and computer networks which are connected by telecommunications lines.80 In reality, they are not a single organized network of computers, per se, but rather series of different networks which share the same protocols for data transfer. These systems facilitate the sending and receiving of text and other digital data files from and to computers around the world. Over the network, a user may use a personal computer to connect to remote computers via phone or telecommunications lines. Typically, the user first goes through a host computer which establishes connections to remote sites on the Internet. Most host computers are academic servers, company servers, or commercial on-line service providers (e.g. CompuServe). Once connected through a host to a remote computer, the remote computer acts as a server, permitting the user to access and download information from the server and send information to the server.81 A user may connect with a series of remote computers in different jurisdictions. Often, when "surfing the web"a user will jump from cite to cite, gaining access to various content, but not knowing where the user actually is in "physical space.î82
Recent changes in technology products and markets have spurred a substantial advance in the capacity, speed, and reach of these global information networks. Increased personal computer penetration into homes in many industrialized countries is one trend that has contributed to the reach of networks. At the same time, other technologies have advanced rapidly. Particularly important for the information society have been advances in information digitization, compression techniques which reduce the amount of bandwidth83 required to transmit information, microprocessing technology which has enabled the rapid receipt and decoding of these compressed signals, and satellite, transponder, and fibre cable scope resulting in increased capacities.  Other market factors, such as telecommunications deregulation prompting providers to seek additional sources of revenue to cover their heavy capital investment needs have also contributed to the expansion of network technologies.
The powerful networks which have resulted from these changes make it possible to transmit immense amounts of data around the world at incredible speeds.  One commentator has called the emerging possibilities the "most far reaching challenge in the history of copyright law.î This challenge is based on the contrast between the systemic characteristics which copyright needs to be effective and the possibilities engendered by the new technologies. While copyright requires a limited ability to copy a work and the ability to account and monitor for use, the information society enables the user to distribute almost any work over a telecommunications network, download and store it, alter it, and retransmit it easily. This new environment erodes the practical constraints which created a type of de facto copyright. Because making copies of software and other digital works does not degrade their quality, users can create an endless supply of quality copies from one original, or one copy.
These sweeping changes create a system with a number of elements which have ramifications for copyright doctrine and enforcement. The material traveling across the Internet is digital and intangible. In addition to making it easy to copy, alter, store, and retransmit the intangible nature of the materials, along with other system characteristics, makes it difficult to track, follow, and control activity. The ability to detect the origin of messages or shut off connections to certain countries or servers has serious flaws and can be circumvented. Moreover, network and communication technology is rapidly evolving and continually novel. Thus, a process that is technically infeasible today may be possible next month. It seems significant that Czarnota and Hart's detailed guide to the Directive, written only five years ago, does not raise any questions of the new environment. Finally, many of the interactions are cross-jurisdictional; users can easily control activities on computers in other countries. How can the E.C. structure a regime that protects intellectual property within the Community if many of the actions which compromise those interests take place outside its jurisdiction?
Collectively, these elements could indeed pose a serious challenge to any system of copyright. The question, then, is how the Software Directive fares in the face of this challenge.
The Software Directive achieves a level of harmonization which is comprehensive, well structured, and protective enough that the Internet trade in software does not pose a threat to the functioning of the internal market. In fact, the combination of the harmonized regime of software protection with a greater interconnectivity between authors, users, and software firms in various Member States may make the emerging on-line market for software a model for other products in the single market.
As described above, the Software Directive harmonizes Member State software regimes across almost all major issues which arise in the area of software copyright. Application of the Directive to Member State laws should provide the same level of protection to the same programs across the entire E.C. With originality requirements specified at a low hurdle and exclusive rights as well as well-specified exceptions to those rights drafted in great detail, software copies which are infringing in one Member State should be infringing in all other Member States. Similarly, products which have legitimately borrowed ideas or interface specifications should be accorded the same privilege across countries. Furthermore, the Directive clarifies what type of contractual and licensing terms are permissible on a Community-wide basis. Therefore, country laws should not provide variations in the favorableness of their contract regimes to avoid rights given in the Directive. Such sweeping harmonization leaves little room for laws which could act as a quantitative restriction as between Member States.
Moreover, the exhaustion provision in Article 4(c) ensures the workings of the internal market by providing for exhaustion of products with a first sale within the Community, but precluding exhaustion within the Community for a first sale in a non-Member State. This keeps differing national views on international exhaustion for software from distorting trade.
There are two exceptions to complete harmonization which harbor the potential to create some trade distortion, but their effect should be minimal. First, the Directive articulates neither a borderline between "idea"and "expression"nor a choice of a test for making this determination. This is a somewhat narrow band of uncertainty, but at the margin, it leaves the potential that "cloned" elements in a product may cause that program to receive protection in one Member State but not another. For instance, Borland's QuattroPro, the subject of the Lotus v. Borland case for copying menu hierarchies discussed above, might potentially be considered infringing in the Netherlands, but not in France. One can hope that a national court will soon seek reference on this issue so that the question can be more clearly settled than it is in the United States. In any case, this problem will apply to programs whether they are sent via the Internet or sold in more traditional distribution channels, such as in mass-market shrink-wrapped packages or through service providers such as value-added resellers. If divisions on this issue actually evolved, the Internet would probably shift the balance toward preserving free trade at the expense of an author's right. Community-wide access to the program on the Internet would make it difficult to restrict users from gaining access to the program through a server in another Member State.
By permitting differences between systems which allow corporate entities to be authors and other systems which require that authorship be vested in a natural person, Article 2(1) creates a second area where disharmony could effect a distortion. The exact problem which this might raise is a conflict in determining who has which exclusive rights in a program, and in which countries they have those rights. Network technology has created the possibility of telecommuting, where employees in multiple jurisdictions can communicate and even work on a product simultaneously. If you have a product which has been created simultaneously by two employees collaborating from different Member States and those states take a different approach to authorship, will employees retain moral rights to the product in that country? Even though Article 2(3) gives an employer full economic rights in a product, "the exercise of moral rights by employees may quickly conflict with the employer's interest of undisturbed exercise of the economic rights." Could the employee-authors require, for instance, require that the work bear their name prior to it being marketed in their Member State? It remains to be seen how Member States with strong authorís rights traditions will resolve this consequence. As a result of the strong message in Article 2(3) that employers are to be given economic control over their products, it seems unlikely that much impact should arise from the retained moral rights. Moreover, as a practical matter, its unclear to what extent any employees would seek to assert such control over computer software which is arguably a less intimate form of personal expression than other types of literature. In terms of the major concern for trade, getting a product on the market, it appears that moral rights are rarely invoked to prevent exploitation.
Beyond purely legal issues regarding potential restrictions to, or promotion of, a more unified single market, the practical effects of the Internet should also be recognized. While economic factors such as transportation costs and distribution channel relationships do not rise to the level of quantitative restrictions or discriminatory duties, as a matter of logic they make it easier for a local producer to supply manufactured products to consumers within their own country.
The Internet removes this local advantage. De facto impediments to competitive parity in getting products to other intra-Community markets are substantially reduced in the networked context. Software products may be transported directly over telecommunications lines at almost no marginal cost. They can also be marketed and delivered directly from the supplier to the consumer. In this environment, a software producer based in Munich has almost no cost or logistics advantage over a competing producer based in Dublin in distributing a product to a Stuttgart. The two products can compete based more strictly on their development cost and their quality. The combination of the Directiveís removal of legal impediments to a Community-wide software market and the Internetís removal of cost impediments set software up as an example to other product areas.[103 ]
As important for understanding what copyright subjects the Internet has the potential to affect is clearly recognizing what issues the Internet will not impact. In this regard, an entire range of issues which arise the context of software copyright will not be influenced by the Internet.
Copyright, in simple terms, controls two different types of threats to rightholders.[104 ] On the one hand, copyright regulates competitive product development between manufacturers to ensure that competing products do not appropriate expression from software that is already on the market. This function sets the bounds of what level of idea borrowing is permissible. On the other hand, copyright also regulates the use and wholesale copying of works by pirates, private individuals and firms. As will be demonstrated below, the problems raised by the information society are mostly questions involving the second threat. In contrast, the Internet should have little impact on copyright's role in regulating conflict among legitimate manufacturers.
It is significant that the challenges posed by the information society will have minimal effect on the major issues of intra-manufacturer rights. Jean-Francois Verstrynge and others have commented that conflict among the competing interests within the software industry made this Directive "one of the intensely debated proposals ever to come before the Council of Ministers and the European Parliament."[105 ] Such conflicts were an impediment to developing any Directive at all.[106 ] These controversies will not have to be rekindled in the framework of addressing the Internet. To the extent that the new debate will pit groups against one another, it is more likely to be an aligned software industry against users and other groups such as on-line service providers. Because of the Commission's perspective that strongly protected content such as software will drive the much needed development of the information infrastructure, balancing between groups should prove easier than trying to help a software industry that is divided about how it should be aided.[107 ]
Two widely debated software copyright questions have been the focus of much discussion in recent years, but should not be further complicated by the advent of the information society. The heated decompilation issue is one. This was the most intensively debated provision of the Directive.[108 ] However, it will not be raised in terms of the Internet, because decompilation is primarily a policy issue about when a competitor can use your specific expression. Decompiling only requires one legitimately purchased copy of a program. No changes in cross-jurisdictional information transfer or the ease of anonymously transferring or copying a program should alter the basis for the policy debate.
Whether copyright is a conceptually appropriate doctrine to apply to software is another issue which should not change because of the Internet.[109 ] This too is a policy question of managing competitive product development between competing companies.[110 ] Does it make sense to permit a company to study a productís functioning and then independently write a program--using wholly different expression--to perform exactly the same function? The idea / expression distinction of copyright permits this, given the Directive's approach of protecting software as a literary work within the meaning of the Berne convention.[111 ] Whether this result of copyright makes sense has been, and continues to be, a topic of serious debate within the intellectual property community.[112 ] However, as with decompilation, the changes introduced by the Internet should not alter this debate.[113 ] However, although these controversial areas can be avoided, the Internet does pose important and difficult issues for the other side of copyright protection--regulating private copying and piracy.
One can evaluate the Software Directive's effectiveness from both a formalistic perspective and a practical point of view. Formalistically, one may ask whether the doctrine's provisions, as a technical matter, will continue to apply to copying and use in the Internet context in accordance with the principles on which the Directive was grounded. Practically, one can ask, simply, whether it will really work. Will real behavior conform to the goals of the Directive?
One pair of commentators has looked at some of these questions and concluded that the Software Directive "will continue to work well for the information infrastructure."[114 ] In general, this assessment is correct. However, from both a formal and a practical viewpoint, the Internet requires changes or clarifications of current doctrine. Recognizing and implementing these changes will facilitate improved protection for rightholders and a more certain environment for needed investment in the information society.
The Directive provides a good formal structure for regulating infringing intra-Community network traffic in software. As a matter of law, the Directive provisions will continue to appropriately regulate infringing actions which could prejudice the interests of rightholders. At the same time, the legitimate needs of users will not be prejudiced.
The Articles 4, 5, and 7 protection structure, when transposed into the Internet environment, continue to control unauthorized copying. In order to distribute or receive an unauthorized copy of software on the Internet, that copy would have to be reproduced in one or more of several places.[115 ] The original holder of the software must upload the software onto a system. This involves either creating a copy to be available for access from a server or packetizing a copy for transmission attached to an e-mail message. The recipient, to keep the software, would also have to reproduce the software onto his or her computer. Each of these is covered by the Directive's exclusive reproduction right.
The extensive and adaptive exclusive reproduction right granted in software unproblematically applies to this transaction. It covers "permanent or temporary reproduction...by any means and in any form."[116 ] Furthermore, specific acts which are necessary to transfer software over the Internet, such as loading, transmission, and storage, are governed by the reproduction right insofar as they necessitate reproduction.[117 ] Although the Directive creates exceptions for making back-up copies[118 ]8 and for other actions which are necessary to use the program in accordance with its intended purpose[119 ], such exceptions would not apply to sending an infringing copy to someone else. The reproduction involved in such loading, transmitting, or storing of an infringing copy would not be "necessary for use...by the lawful acquirer in accordance with its intended purpose."[120 ] Thus, the example of sending a copy of infringing software or making such a copy available generally on a bulletin board should be no different from copying software onto a floppy disk and mailing it to someone or leaving it in a computer lab for general copying. Because this entire transaction has taken place within the E.C., and each step is regulated by the provisions of the Directive, choice of law as between Member States is irrelevant for the purposes of ensuring that reproduction is governed by the protection offered by the Software Directive.[121 ]
The case of basic infringement within the E.C. is unproblematic, but one issue which arises in the context of sending information over the Internet requires attention. Mirroring and caching is a potential problem for copyright doctrine which neither the NII White Paper or the EC Green Paper addresses.[122 ] Mirroring and caching is the copying of materials from a site in order to improve the performance of the system.[123 ] For example, a service provider may copy the materials from the site onto its system in the course of transmission so that the provider will not have to go back out over the Internet to redownload the data should the user request the materials again. In so far as this makes a download more efficient or ensures that the user can reaccess the copy should there have been difficulty in the transmission from the service provider to the user, such caching should be permitted. A user has legitimately accessed a site and, through payment or otherwise, gained permission to make the copy. The copy could be considered to be a reproduction necessary for transmission, permitted under the Directive. However, if the same cached copy may later be accessed by other users without paying the appropriate fee to the rightholder, the copy should be considered to be infringing. Policy-makers should consider creating a clearly defined, limited privilege for caching software which disallows permitting subsequent users to access cached software.[124 ]
Whereas trade in software on networks within the E.C. presents few problems for applying the Software Directive, the extra-Community connections of the Internet create significant formal legal issues. The global information infrastructure makes it easy access remote computers in other jurisdictions and impact their functions without actually crossing into those jurisdictions.[125 ]
In terms of questions of formal legality, the most significant challenge to the Software Directive will be in regulating the transfer of software between the Community and a jurisdiction which has lower copyright standards, in law or in practice, than the E.C.[126 ] There are two points at which a rightholder can seek to control unauthorized copying on the Internet. Where the content is uploaded, and where it is downloaded and used.
In the transnational context, seeking to control uploading must be done in the country where the content is being placed on a server. There appear to be two permutations of the third country problem. One is where a third country bulletin board contains pirated software considered to be infringing in both the third country and the Community. The second is where the software is legally on the bulletin board in the third country, but considered infringing by the E.C.
With the first problem, the rightholder can take legal measures in the third country to stop the infringing material from being made available. In practical terms, this is prefereable because it is logistically easier to seek out one infringing bulletin board than hundreds or thousands of users who access it. Such an action would proceed under the laws of the jurisdiction where the material is being placed on the bulletin board.[127 ] This situation should be no different than pursuing the trade in tangible pirated goods. Complications arise where copies of software which are infringing within the E.C. are not infringing in the third country. Assume, for example, that the third country applies a similar originality hurdle to software as Germany used prior to the Software Directive. Under such a regime, a large percentage of software products currently protected within the E.C. could be uploaded onto a bulletin board in that country without violating the local reproduction right. Within that country, the products could be freely accessed, downloaded and copied. The rightholder would have no recourse against those making copies available in the third country. Yet, over the Internet, this bulletin board can also be accessed by users from the European Community, who are able in turn to download the software. Therefore, the rightholder must seek redress at the other point in the chain--downloading in the E.C.
In some cases, as a formal legal matter this action is covered squarely by the Directive. Although a Member State cannot impound the software at the border, as they could with a hard copy, the user's action of saving the copy from his or her download would involve reproduction of an infringing work and would be covered by the exclusive grant of right in Article 4(a). To the extent that the users in the E.C. download and save copies of the software, the lack of international exhaustion for software would mean that these copies would be considered infringing.[128 ] The Software Directive would appropriately cover these copies.
However, under a different, easily imaginable scenario, it is currently unclear that the Directive would provide protection to software accessed from a foreign bulletin board. This would involve a European user downloading and using software in the computer's RAM without ultimately saving it.[129 ] One could imagine a bulletin board in a third country offering software--games for instance--which a user could access, download into RAM, play or use, and then discard without saving onto a hard drive, floppy disk, or other storage device. The user could do this as often as he or she wanted, so that the practical effect would be no different that having oneís own copy of the program. Would copying the program in the computer's RAM be covered by the reproduction right?
Regulating the loading of a copy into RAM as a reproduction is the key element of being able to assert legal regulation over cross-jurisdictional use if the software is not subsequently downloaded. Applying the reproduction right to RAM is, therefore, critical to protecting rightholder interests. Yet the status of programs loaded into RAM is currently unclear under the copyright laws of both Europe[130 ] and the United States.[131 ] To ensure the protection of computer programs in the transnational environment created by the Internet, the reproduction right in RAM copies needs to be explicitly clarified.
There is a strong legal argument for the proposition that loading a program into RAM is governed by the reproduction right. Although it is not technically clear from the use of "reproduction" in Article 4(a), the structure of the Directive makes such an interpretation logical. The Directive's granting of very broad exclusive rights with narrow and detailed exceptions suggests an intention to apply the exclusive rights broadly. Interpretation in concert with the purpose of these broad rights would involve appying the reproduction right to RAM.
The legislative history also suggests that coverage of loading into RAM was intended. The Article 4(a) restrictions on loading, viewing, running, transmission or storage of a program were originally drafted as restricted acts "[i]nsofar as they necessitate a reproduction of the program in part or in whole...."[132 ] After the Parliamentís input, this section was changed to make these acts subject to authorization by the rightholder "[i]nsofar as they necessitate a permanent or temporary reproduction of the program.î[133 ] Furthermore, the explanatory memorandum stated that the purpose of this change was to make it
"more explicit that because the acts of loading, displaying, running, transmission or storage do at the present time involve an act of reproduction of the program, and since all kinds of reproduction, both permanent and temporary are technically a violation of the author's rights in the program, any such reproduction requires the authorization of the rightholder.[134 ]
The use of "at the present time"in the explanatory memorandum suggests a recognition that some of these acts may not require reproduction in the future.[135 ] The method of loading from the Internet does, however, involve the reproduction of code from somewhere into RAM, just as loading from a floppy disk, and should not, therefore, be considered any different.
Such an interpretation is also in accord with the policy direction of the statute. In addition to creating the sole remedy for the transnational use described above, this interpretation serves important policy goals beyond regulating Internet transmissions. It would prevent users from loading multiple computers with one copy of a disk, thus stopping "parallel use of the software, which violates the creator's interest in commercial participation in exploitation of the program."[136 ] Based on the legislative and policy grounds, as well as the importance of providing some type of control for transmissions of infringing material originating outside the E.C., storage in RAM should come under the exclusive reproduction right in the Directive. The Commission should clarify this position in the course of updating the Directive to accommodate the information society.
No matter how effectively the Directive may technically address software piracy within the E.C., the practical realities of non-compliance and enforcement difficulties in the Internet environment may nonetheless mandate change. The effect of the Internet will alter the current rights balance between users and rightholders. Therefore, the law will either have to recognize that it is permitting a reallocation of privilege or take steps to counteract that shift.
The problem with enforcement is "obvious." Given the size, fluidity, and scope of the information superhighway, how can a copyright owner discover and stop infringement?[138 ] Although some commentators have suggested that the Internet will not cause a large increase in unauthorized distribution of protected works[139 ] , I would argue that, absent some countervailing influence, the Internet structure establishes incentives and an environment which will make the copying problem substantially worse.
The global information infrastructure changes the calculus by which users currently assess their compliance with copyright laws. The Internetís transnational character lowers the costs and the risks of physically transporting pirated goods.[140 ] The marginal cost of providing each pirated copy, even internationally, is practically zero.[141 ] Whereas a European traveler might have serious qualms about trying to return through customs with a large number of infringing software programs acquired outside the E.C., downloading them over the Internet seemingly entails little threat of discovery. For the purveyor of pirated works, it will be easier to access major consumer markets from copyright havens which do not protect, de facto or de jure, software.[142 ] Even within the E.C., the ability to act anonymously over the infrastructure, or at least without the feeling that you are being regulated, decreases oneís perceptions of risk. For both the user and the provider of infringing software, the Internet reduces the incentives to comply with copyright regulation.[143 ]
At the same time that it minimizes the perceived risk from infringement, the new technology of the information society increases the harm caused by any one individual's decision not to comply with copyright. The ease of duplication, forwarding, and retransmission radically increases the breadth of dissemination that a single infringer's conduct may have. One user can forward a copy of a software program to fifty or many more friends on a mailing list in a few key-strokes or clicks of the mouse. A single uploading of a software program onto a bulletin board can make it accessible to thousands of potential customers--seriously diminishing the market for that work.[144 ] Compounding these activities across multiple actors can quickly outstrip any impact copying in the analog world could have. With a floppy disk, the logistical hurdles which have traditionally deterred copying--time and investment--were barely evident. On-line, they are non-existent.
Finally, the culture surrounding both the copying of software and the sharing of intellectual property on the Internet suggests relatively widespread willingness to ignore traditional copyright norms. Software copying is an extensive and well-known phenomenon.[145 ] Recent studies of software piracy estimate that over 50% of the packaged software used in Western Europe and 35% of that used in the United States is copied.[146 ] The "anything goes"attitude on the Internet compounds this problem.[147 ] As was illustrated by the mass loading of protected software onto a bulletin board in U.S. v. LaMacchia, many users seem to perceive a right, or invincibility to deterrence, to conduct mass copying over the Internet.[148 ]
Collectively, these trends work together to fundamentally alter the balance between users and rightholders. At the same time that the Internet lowers the risk of infringers getting caught and offers users easier access to infringing software, it substantially raises the level of harm each act of infringement can have. Without changes, the software market is likely to become subject to higher levels of more damaging infringement.
Despite the suggestion elsewhere that the enforcement conundrum spells the end for copyright in cyberspace[149 ], the E.C. should move forward with efforts to support legal and technological measures improving the enforcement of copyright on the Internet. Emerging technological possibilities hold promise for counteracting some of the changes brought on by the networked age. Furthermore, on the legal front, experience has shown that more vigorous monitoring and enforcement can reduce the piracy of computer software.[150 ] Rather than scrapping copyright, policy-makers should adjust the current system to rebalance incentive changes created by the Internet.
A non-legal, technological solution may be the most promising route to containing software piracy. The Commission has already recognized that digital technology holds future possibilities for identifying, tattooing, better copy protecting, and managing software.[151 ] These technologies include encrypting transmissions so that they can only be downloaded by users who have a decoder, improving copy protection measures to constrain creating or running a pirated copy, and identifying documents with digital tattoo markings that can track when they are being transmitted across a network.[152 ] While each has its specific problems, these technologies could restore the balance, or even heighten current levels of protection, in policing software copyright.[153 ] The ability to monitor transmissions of copyrighted works through tattooing, particularly, would have a significant in increasing the chances of being identified for sending and receiving infringing software. This effect could have a major impact in restoring the balance.
Unfortunately, technological solutions to copying have rarely worked for long. Because of user desires to make back-up copies or download software onto hard drives for more effective use, strong resistance to technological anti-copying measures led software firms to abandon many of their efforts at copy protection.[154 ] Moreover, the large and talented community of sophisticated hackers often undermines any technological solutions which the industry creates.[155 ]
Such threats to technological solutions and the speculative nature of how they may eventually pan out require that the market not rely entirely on improved technology to solve the enforcement dilemma. Technological solutions will work to a point, but supplementation via other means is necessary.[156 ] This supplementation can take two directions: new laws can engernder and protect new technical devices or they can directly impact user's perceptions of the risks of getting caught infringing and the costs of doing so. The Software Directive's enforcement measures are a good first step in this direction. Yet, they are only a first step.
The requirement that Member States create appropriate remedies against persons who have or who circulate devices for removing technical copy protection in computer does support the evolution of technical protection.[157 ] Such a measure was one of the few legislative changes seen as necessary by the NII White Paper.[158 ] Yet, while this provision is certain to be effective against widescale marketing of devices to circumvent copy protection, it is relatively weak in its limitation to "any means the sole intended purpose of which is to facilitate the unauthorized removal or circumvention"of copy protection.[159 ] As such devices are developed with possibly more than "one sole intended purposeî, it will remain to be seen whether the multi-use possibilities of copy protection removal devices leaves Article 7 (c) with any bite.[160 ] However, even if Article 7 (c) proves to be ineffective, a broader restriction hardly seems possible on this matter. Restricting all such devices could damage the competing interests which led to the Article 5 and 6 exceptions.[161 ]
The further two protection measures in Article 7 are also an important step is enhancing protection.[162 ] While Unfortunately, the interaction of these two protection measures no longer functions as well in the new environment. As the LaMacchia case illustrated, there can be tremendous scope to non-commercial software piracy on the Internet. The Software Directive does wisely mandate measures to address placing infringing copies into circulation, whether done for commercial purposes or not.[164 ] Such a provision unequivocally increases the possible costs perceived by someone for placing infringing software up on a bulletin board. What the Directive does not do is create enforcement mechanisms, beyond seizure, for possession of infringing copies for non-commercial purposes.[165 ] This may reflect a hesitancy to go after individuals users, which would most likely continue. But it must be recognized that the enforcement environment has changed.
The Article 7(a) and (b) structure was adequate to provide remedies against a majority of infringing copies prior to the Internet. Given that the prevailing form of copying is corporate "over-use"of software and that the suppliers of infringing software to individuals, prior to the advent of cross-jurisdictional networks, needed actors within the E.C. to distribute their works,[166 ] there was little need to address remedies to non-commercial individual infringers. Over the Internet, however, individuals may have greater and greater access to infringing software supplied by persons beyond the jurisdictional reach of Article 7 (a). This has the potential expandthe importance of infringement taking place outside the corporate context. One possible solution is to bring non-commercial users within the scope of the remedy system, although this may be an unpopular measure.[167 ] A further possible solution is international harmonization.
Like technological developments, international harmonization is out of the E.C.ís exclusive control. Therefore, it is not a solution the Community should await without taking other action. Nevertheless, because it is much easier to put a dent in piracy by identifying one infringing bulletin board than by pursuing each infringing user, international harmonization of software copyright standards and enforcement should be prioritized by the Community.[168 ] Although it is beyond the scope of this paper to evaluate these efforts, a few observations about the Directive bode favorably for harmonization in this area.[169 ] First, the Directive brings the software copyright regimes of the Member States very close to that of the U.S.[170 ] Furthermore, the approach taken by the Community in the Directiveís final form represents a shift from an authorís rights perspective to a more utilitarian view of copyright.[171 ] Given the substantial approximation of the E.C. regime with that of the U.S., and the U.S. commitment to driving international agreement in protecting works on the global information infrastructure, there seems to be a good possibility that such an agreement could emerge under joint E.C.-U.S. pressure.
In addition to looking to technology, consideration of increasing penalty provisions, and promoting international harmonization, two further measures may be suggested which improve the chances of identifying infringement. Each results from either trying to address the overwhelming scope of trying to find infringement in the expansive mass of cyberspace.[172 ]
The first is relatively simple--consider a collective licensing society approach.[173 ] Users are free to do this today and, in a way, they already are. For instance, in the context of their trade association, various software manufacturers have taken collective steps to seek out and challenge copyright infringement in a variety of jurisdictions.[174 ] However, given the scope of the Internet, it is unclear that any one industry group could effectively police cyberspace in a cost-effective manner. While a possible solution to the scale problem is to pool resources across product industries, it is unlikely that traditionally independent businesses such as software firms would desire to come within the ambit of, and cede rights to, established licensing groups.[175 ] It seems even more unlikely that they would want to submit themselves to a compulsory system like that which exists in some jurisdictions. Assuming that collective enforcement will continue to proceed on an ad hoc basis in various industries, it is still likely that such efforts by the industry will alter some users' willingness to load infringing software on the network. Nonetheless, it is practically impossible that such groups could identify the users receiving infringing software. If a bulletin board is operating in a low protection jurisdiction, as postulated above, a collective enforcement approach would have little effect.
Imposing liability on the operators of computer bulletin boards is second route to enforcement which might increase chances of finding copying and the correlated user perceptions of infringement being identified. This policy has received attention in the U.S. over the past two years. A series of recent federal district court cases have each providing for some liability for Internet service providers or bulletin board operators. In Playboy Enterprises Inc. v. Frena, the district court held that a bulletin board operator may be directly liable for infringing the distribution and display rights by making available to the public infringing copies which a user has uploaded onto the system.[176 ] Similarly, Sega Enterprises Ltd. v. MAPHIA held a bulletin board operator directly and contributorily liable for copyright infringement where users uploaded computer software games onto the bulletin board for other users to download.[177 ] In this case, the defendant operator apparently had encouraged and clearly knew of the infringement.[178 ] The most recent case to consider the problem conflicts to some extent with these cases, but nonetheless finds a possible basis for operator liability despite the large size of the service providerís client base and lack of evidence that the provider encouraged the use of the system for infringement. In Religious Technology Center v. Netcom, the court reasoned that any finding of direct infringement for a bulletin board operator would be inappropriate because they have merely set up and operated a system for the functioning of the Internet--similar to owning a copy machine--and because there are other parties directly responsible and liable for the infringement.[179 ] However, the court refused to dismiss the case against Netcom, suggesting that Netocomís leaving the materials on the system after possibly knowing of their infringing nature stated a legal claim for contributory infringement.[180 ]
These cases demonstrate the difficult challenge in developing a doctrine of third-party liability for on-line infringement. Such service providers can play varied roles in setting up and encouraging use of their systems.[181 ] Their role can be analogized to current legal categories in a variety of ways, from a common carrier, to a copy machine owner, to a publisher, to a retail store. Critically, service providers also are positioned to have the potential to play a strong role in copyright enforcement.
Despite calls to provide a liability exemption for on-line service providers, the NII White Paper recommended retaining the possibility of liability at this time.[182 ] Interestingly, this is a policy matter which was not even raised in the E.C. Green Paper. As discussions of regulating copyright in the information society continue in the E.C., service provider liability is a difficult policy issue which needs to be addressed.
There are strong arguments on both sides of this issue. Service providers have pointed to the unfairness and impracticality of trying to police everything that takes place on their systems.[183 ] Such policing raises privacy concerns and specters of systems operators reading private e-mail. In addition, shifting such costs over to service providers has the potential to chill necessary investment to expand the system.
Despite these drawbacks, there are compelling reasons why the E.C. might consider calling for some form of liability for service providers. It may be the only enforcement mechanism within the Communityís control which has the potential to redress the altered balance. Of the non-technical options considered, it is the sole mechanism which would seem to give a private user pause before downloading an infringing software product. Just as some services currently regulate their content, albeit imperfectly, to screen obscenity and pornography, services might be asked to screen for software traveling on line. The specter of being discovered would provide deterrence to users who would otherwise perceive little risk in uploading or downloading infringing content. In addition, not exempting providders retains their to develop technical and marketplace tools to reduce infringement and thereby lower their own exposure. This is certainly a proposal worth considering.
The Community may ultimately decide this is a step it does not want to take. Privacy or investment concerns may outweigh perceived improvements in compliance. Because bulletin board operators play different roles and assert different levels of control, one might fear adverse effects from creating different regimes of regulation. Creating liability for those who do play a role regulating content could cause all operators to take a laissez-faire approach, ultimately reducing compliance further. Regardless of the final decision made on this issue, it is a subject the E.C. needs to put on its agenda and ensure is solved with an E.C.-wide policy.
As alluded to several times, regulating the Internet poses difficult problems of balancing interests between various groups. The Software Directive can accommodate the Internet, but the changes presented in this paper will need to be addressed. In assessing policy changes to adapt copyright doctrine to the new environment, there are at least four different, but overlapping, perspectives that enter the debate. One can take the point of view of rightholders, of users, of intermediate service providers, or of promoter of network expansion. Keeping the impact on each of these groups in mind is critical in structuring a wise system. Minimizing the conflicts may also be important for creating a consensus agreement.
As the NII White Paper has been debated in the U.S., one of the most heated issues has been reconciling the conflicts between users and rightholders. This debate has been carried on at a very contextual level. Those advocating from a userís point of view have focused on how copyright regulation on the information superhighway could make "browsing through a borrowed book, lending a magazine to a friend, [or] copying a news article for your files" illegal. It has been postulated that the information society is evolving into an evironment where current free access and browsing rights in libraries will become pay as you go copying via computers.
These possibilities could be real. At some point, if the best information is available on-line for a cost, timely retrieval is available only to those who will pay for it; consequently, public access will be limited. The types of questions such possibilities raise are important for a society. One's answer may well depend on deeply held political values about information, rights to knowledge, and free speech. The conflicts may force difficult trade-offs.
Values and world view are implicated in the software debate as well. Are people comfortable with an environment where 50% of software is copied? Would it be acceptable for the only users purchasing new software to be those who want the support relationship and timely product updates that come with a valid registration number? Is it necessary to further strengthen protection when, even with currently high levels of copying, the industry has continued to profit and supply the market with new products? While people differ about these questions, such issues have hardly the political or emotional pull of questions about access to news, information, and cultural products. It is hard to imagine people becoming incensed about the fact that they cannot "loan" a friend a duplicate copy of Microsoft Word. When you contextualize the copyright issues in software, the answers seem much more apparent.
This suggests that, at some basic emotional level, the issues in regulating software are fundamentally different from many other products which will be on the information superhighway. Software can avoid many of the difficult questions facing other items. Thus, these issues should be treated separately. Both the NII White Paper and the E.C. Green Paper have addressed the issue of copyright on the Internet monolithically. Such an approach blurs the line between product areas in a way that overlooks real differences in culturally accepted use.
Going forward, the E.C. should address protecting software on the Internet as a discreet matter. The evolution of the Software Directive has already shown that, in the context of supporting the development of an industry for functional products, Member States have been willing to think of software differently than many traditionally copyrighted products. Similarly, instead of looking to one overarching regulatory structure for the information society, necessary changes should be discussed specifically with respect to the software market and other individual product groups.
Such an approach would facilitate stronger protection for software. Legitimate userís rights concerns about lending, borrowing, and public access that arise in other areas should not impact the consideration of what balance is appropriate for computer programs. Given that harmonization of copyright for different products in the E.C. has proceeded at different speeds, such a disaggregation should prove easier than it might be under the unitary copyright statute that governs U.S. law. Finally, while permitting stronger protection measures to be established for software products, a product-specific approach will allow the much needed, explicit discussion about how to balance the legitimate interests of groups in other affected markets as well. It is important for the Community promote product protection to build this new digital age, but it is important not to leave individuals behind while doing so.
Angel, John. Legal Risks of Providing Services on the Internet. 11 Computer Law & Practice 150 (1995).
Auerbach, John. Premium Fees the Next Wave for 'Net Surfers, The Boston Globe 1 (April 18, 1996).
Burk, Dan L. Transborder Intellectual Property Issues on the Electronic Frontier, 6 Stan. L. & Pol'y Rev. 9 (1994).
Christie, Andrew. Designing Appropriate Protection for Computer Programs, 1994 Eur. Intell. Prop. Rev. 486.
Christie, Andrew. Reconceptualising Copyright in the Digital Era. 1995 Eur. Intell. Prop. Rev. 522.
Cinq, Robert A. Note, Making Cyberspace Safe for Copyright: The Protection of Electronic Works in a Protocol to the Berne Convention. 18 Fordham Int'l L.J. 1258 (1995).
Czarnota, Bridget and Robert J. Hart. Legal Protection of Computer Programs in Europe--A Guide to the EC Directive. London: Butterworths (1991).
Dixon, Allen N. and Laurie C. Self. Copyright Protection for the Information Superhighway, 1994 Eur. Intell. Prop. Rev. 465.
Geller, Paul E. "International Copyright: An Introduction," in 1 International Copyright Law and Practice, INT-15. (Meville B. Nimmer & Paul Edward Geller, eds.) New York: Matthew Bender (1995).
Ginsburg, Jane C. Putting Cars on the "Information Superhighway": Authors, Exploiters, and Copyright in Cyberspace. 95 Colum L. Rev. 1466 (1995).
Goldstein, Paul. The EC Software Directive: A View from the United States of America, in A Handbook of European Software Law, 203 (M. Lehmann & C.F. Tapper, eds., 1993). Oxford: Clarendon Press (1993).
Hardy, Trotter. The Proper Legal Regime for "Cyberspace". 55 U. Pitt. L. Rev. 993 (1994).
Hoeren, Thomas. The Green Paper on Copyright and Related Rights in the Information Society. 1995 Eur. Intell. Prop. Rev. 511.
Lehmann, Michael. The European Directive on the Protection of Computer Programs, in A Handbook of European Software Law 163 (M. Lehmann & C.F. Tapper, eds.). Oxford: Clarendon Press (1993).
Lehmann, Michael. TRIPS, the Berne Convention, and Legal Hybrids. 94 Colum. L. Rev. 2621 (1994).
Mallam, Paul. Copyright and the Information Superhighway: Some Future Challenges. 6 Entertainment L. Rev. 234 (1995).
Meijboom, Alfred P. The EC Directive on Software Copyright Protection, in Copyright Software Protection in the EC 1. (Herald D.J. Jongen and Alfred P. Meijboom, eds., 1993).
Metalitz, Steven. The National Information Infrastructure, 13 Cardozo Arts & Entertainment L. J. 465 (1995).
Miller, Arthur R. Copyright Protection for Computer Programs, Databases, and Computer-Generated Works: Is Anything New Since CONTU? 106 Harv. L. Rev. 978 (1993).
Perritt, Henry H. Jr. Law and the Information Superhighway. New York: John Wiley & Sons, Inc. (1996).
Rosenbloum, Robert A. The Rental Rights Directive: A Step in the Right and Wrong Directions, 15 Loyola Entertainment Law Review 547 (1995).
Samuels, Edward. Copyright Concerns on the Information Superhighway. 1994 Annual Survey of American Law 383 (N.Y. University School of Law, 1995).
Samuelson, Pamela. Consequences of Differences in the Scope of Copyright Protection on an International Scale, Unpublished Paper delivered at Symposium on "Information, National Policies, and Internal Infrastructure", Harvard Law School (January 29-30, 1996).
Samuelson, Pamela, Randall Davis, Mitchell Kapor & Jerome H. Reichman. A Brief Manifesto on the Legal Protection of Computer Programs. 94 Colum. L. Rev. 2307 (1994).
Samuelson, Pamela. The Copyright Grab. Wired 4 (Jan. 1996).
Samuelson, Pamela. Digital Media and the Law, 39 Communications of the Association for Computing Machinery 23 (October, 1991).
Samuelson, Pamela. Intellectual Property Rights and the Global Information Economy. 39 Communications of the Association for Computing Machinery 23 (January 1996).
Shaw, Josephine. European Community Law. London: The MacMillan Press (1993).
Stewart, Stephen M. International Copyright and Neighbouring Rights. London: Butterworths (2d. ed. 1989).
Stone, Peter. Copyright Law in the United Kingdom and the Euopean Community. London: Athlone Press (1990).
Sucker, Michael. The Software Directive--Between the Combat against Piracy and the Preservation of Undistorted Competition, in A Handbook of European Software Law 11 (M. Lehmann & C.F. Tapper, eds.). Oxford: Clarendon Press (1993).
Verstrynge, Jean-Francois. Protecting Intellectual Property Rights within the new Pan-European Framework: Computer Software, in A Handbook of European Software Law 1 (M. Lehmann & C.F. Tapper, eds.). Oxford: Clarendon Press (1993).
Vorhees. The Wired Triology: The White Paper Ain't Perfect, but it's not the Disaster Naysayers Make it out to Be. 3 Information Law Alert (December 20, 1995).
Wilkinson, Anne. Software Protection, Trade, and Industrial Policies in the European Community, in A Handbook of European Software Law 26 (M. Lehmann & C.F. Tapper, eds.). Oxford: Clarendon Press (1993).
Zimmerman, Diane Leenheer. Copyright in Cyberspace: Don't Throw out the Public Interest with the Bath Water. 1994 Annual Survey of American Law 403 (N.Y. University School of Law, 1995).
Top of the page