With the Internet playing an increasingly dominant role in the way people communicate and the World Wide Web becoming a major medium by which ideas are exchanged and business conducted, more disputes over intellectual property rights in cyberspace are making their way towards the courts. Moreover, in a medium in which geographical boundaries are almost meaningless, questions of where these disputes can be resolved are often difficult.

This Article intends to contribute to the present discussion concerning how to properly address the pertinent legal questions arising from the utilization of the Internet as a means of communication. To understand the Internet's development allows us to gain a better understanding of its particular significance today as well as in the near future. This leads to the insight of how important it is to find appropriate answers to the pertinent legal questions deterring copyright holders from posting their works on the Internet. [FN1]

*629 This Article provides an overview of copyright infringement on the Internet and the international jurisdictional dilemmas that arise in protecting intellectual property rights of Internet content providers. Part I briefly examines the Internet. Part II reviews international jurisdiction and choice of law implications in intellectual property disputes concerning the Internet. Part III analyzes the issue of whether framing is copyright infringement under United States law. This Article concludes that individuals from foreign nations who use American copyrights without authorization on foreign Internet service providers are subject to liability under the United States Copyright Act of 1976, through the exercise of personal jurisdiction over the international infringer.

A. The Significance of the Internet and the Scope of the Analysis

The Internet [FN2] is the term describing the world of electronic communications by the means of computer networks. The idea to connect single computers within a network to enable users to exchange ideas and data came into existence in the beginning of the 1960's when Internet pioneers were envisioning the so called 'Galactic Network'. [FN3] Shortly after that, in 1965, the first data processing connection between two remote computers via a telephone line had been established, although the circuit switched telephone line limited the 'network's' useful application. Nevertheless, *630 the first wide-area network had been built in 1965. [FN4] Departing from the point of the first two-computer-network in the 1960's and the subsequent creation of the ARPANET, a military communication network with three-hundred computers linked in 1981. [FN5] Thus, a much more sophisticated technical standard gave rise to the Internet as it is known today. [FN6] According to pertinent Internet growth-surveys as many as 2,215,195 Web sites available on the Internet. The development of this number is depicted in the first graph in the Appendix, beginning in June 1993: [FN7] The number of Internet hosts, starting with four in 1969, and domains, starting with 3.900 in 1989, have also tremendously grown. Following those numbers, the second and third graphs in the Appendix also provide a strong visual impression about the evolving significance of the Internet as well as they give an idea about the Internet's importance in the near future. [FN8]

*631 This booming quality of the Internet, expressed by the startling increase of the number of Web sites available and end users connected to the Internet, is usually explained with the Internet's potential commercial value as well as with people's general desire to disseminate information they consider worthwhile to disseminate. [FN9]

Since contemporary technical standard makes it so easy to digitize works of authorship in the first place making them accessible for everyone by posting and distributing them on the Internet, it goes without saying that the Internet challenges the existing patterns of established law. Contract law [FN10] as well as criminal liability [FN11] and privacy issues [FN12] are at stake. However, the emerging law of the Internet particularly invokes the challenge of copyright law. [FN13] Copyright protected works once digitized are much easier to copy than conventionally accessible works. [FN14] Moreover, copies from a digitized work appear generally in the same quality as the original. [FN15]

*632 A person who edits and provides digitized information and uploads it to the Internet might incorporate unauthorized copies of other individual's work or infringe other rights exclusively assigned to the author of a copyright protected work. Along the same line, Internet Service Providers ("ISPs") may be held liable for copyright infringement just by virtue of providing active and passive access to the Internet, thereby disseminating infringing work provided by third parties. [FN16] Moreover, copyright infringement may occur just by transmitting a work through the Internet. Particularly the use of the World Wide Web, which is the most popular segment of the Internet, [FN17] makes it easy to download and copy information desired by users who are interested in the information provided. Apart from the substantial copyright law, jurisdiction and choice of law issues have to be taken into consideration since the Internet works through a world wide network which does not stop at the border of one's home country.

This Article focuses on copyright infringement and liability from the point of view of an Internet content provider, whose Web page is being framed. In order to invoke jurisdictional and choice of law issues, it is assumed that a German citizen living in Germany using a German Internet service provider ("ISP") frames a Web site provided by a United States citizen living in the United States. The consumer, respectively the end user, of the framed Web site will be a United States citizen as well accessing the Web site from the United States.

*633 B. What exactly is Framing?

In order to understand the significance of the term 'framing' one must understand how the World Wide Web functions. The World Wide Web consists of countless Web pages providing information. Those Web sites have, to make each Web site individually accessible, a particular URL-address assigned to it. A Web site is usually written in 'Hypertext Mark-up Language' ("HTML") allowing it to incorporate Hypertext References ("HREF"), the 'links' which are designed to direct the attention of the user to other Web sites which are usually related to the content of the Web site being visited. An anchor that is electronically incorporated in the document being viewed represents such a link. The anchor may consist of particular words, sentences, or paragraphs, marked as anchor by usually underlining them. By clicking the computer's pointing device, a mouse, on the anchor the web browser immediately loads the URL-address of the linked Web sites and transfers to it enabling the user to view the linked Web site. [FN18]

Whereas linking just provides a reference point to the document being viewed without really incorporating it into one's own Web site, [FN19] framing goes farther. It is based on the modern Internet browser technology with its ability to open a number of frames on the screen for direct viewing of different Web sites containing *634 text, graphics or other HTML elements. [FN20] This technique gives the end user the advantage to browse different Web sites with related information at the same time. [FN21] Moreover, framing enables a Web site operator to incorporate remote Web sites completely into the Web site the end user actually called upon without terminating the connection to this Web site. [FN22] Thus, when a framed Web site is called upon by clicking on the link's anchor, the linked Web page may appear in part on screen surrounded by a frame that is provided by the 'framing' Web page, which is the Web page the end user actually called upon. [FN23] The frame often depicts logos, advertisement, or promotions placed by the author of the framing Web site. The logos, advertisement, or promotions of the framed Web site are sometimes covered by the frame and invisible to the end user. The frame then may generate the impression that the viewer is actually viewing information coming from the originating framing metasite instead of from the linked site. [FN24] But not only does the frame generate this impression, the URL does as well, because the URL displayed in the browser is the framing site's address and not the site's address actually being framed. [FN25]

In a case recently before the court of the Southern District of New York [FN26] the plaintiffs, among them the Washington Post, complained that the defendant, Total News, Inc., an Internet content provider, hosted a Web site consisting of framing links to the Washington Post's and other news provider's Web sites. According to the complaint, [FN27] viewers of the defendant's Web site were *635 able to choose the plaintiff's anchor transferring them to a corresponding Web site showing the news provided by the plaintiff and framed by the defendant depicting its own advertisement and logo. [FN28] The parties settled the case. [FN29] As for the settlement, the defendants agreed to permanently refrain from causing plaintiffs Web site to appear and to cease the practice of framing. However, the parties stipulated that defendants may link to plaintiff's Web sites via hyperlinks consisting of the linked sites in plain text. [FN30] This case demonstrates the significance of the question of whether framing infringes the author's exclusive rights granted by the Copyright Act of 1976 ["Copyright Act" ]. [FN31] Moreover, it shows that framing represents, since no precedent is available, [FN32] an unsettled area of law. [FN33]

Recalling the hypothetical, [FN34] it is reasonable to assume that the German content provider and his or her ISP might be responsible for the allegedly infringing use of the framed Web site and that they should be held liable for the damage they allegedly caused. Since a suit should be brought against them in the United States, two things are necessarily required. First, it has to be determined whether United States courts can exercise international personal jurisdiction over the German defendants and second, in case it can, what the applicable law is.

A. International Personal Jurisdiction

First, the general principles regarding international personal jurisdiction will be extracted from the pertinent law before those principles will be applied to the conduct of framing following the *637 hypothetical. [FN35]

Any given State's authority to affect its legal interests is accepted as a general principle. This principle is commonly described by the term "jurisdiction." Traditionally, three kinds of jurisdictions are distinguished: legislative, judicial, and enforcement jurisdiction. [FN36]

The settled principle that states have authority to affect their legal interests particularly to persons and circumstances does not mean that states are free to legislate whatever they want. It is generally accepted as a basic rule that a state's authority to legislate ends where legitimate interests of other states are affected. Within this framework of diverse interests states generally recognize that whatever happens on the territory of a State is of that State's primary concern. This principle is called the territoriality principle. It is also recognized that a state has a significant interest in exercising jurisdiction over persons or things that possess its nationality, the nationality principle, as well as the interest in protecting its nationals from conduct outside its territory, the passive nationality principle. In addition, states have the right to protect themselves against hostile acts (protective principle) and finally there are certain activities universally condemned (universal principle). [FN37]

In accordance with these general principles just pointed out, the section 402 of the Restatement (Third) of Foreign Relations further summarizes the legitimate interest states have to exercise jurisdiction. It provides that a state has jurisdiction to prescribe law with respect to:

*638 (1)(a) conduct that, wholly or in substantial part, takes place within its territory;

(b) the status of persons, or interests in things, present within its territory;

(c) conduct outside its territory that has or is intended to have substantial effect within its territory;

(2) the activities, interests, status, or relations of its nationals outside as well as within its territory; and

(3) certain conduct outside its territory by persons not its nationals that is directed against the security of the state or against a limited class of other state interests. [FN38]

The principle of territoriality is well established and free from controversy. As early as 1812 the courts had recognized that '[t]he jurisdiction [to legislate] of the nation within its own territory is necessarily exclusive and absolute' and that a restriction upon it would diminish a state's sovereignty. [FN39] Thus, no foreign country has the right to legislate another country's laws. Along the same line it is basically noncontroversial that a country has the right to regulate the conduct of its citizens or nationals anywhere in the world. [FN40] The passive nationality principle might be invoked when an act committed in one state causes injury in the territory of another state. Even though the act did not take place in the territory of one state, jurisdiction can be established there on the fact that the injurious effect took place in the territory. [FN41]

Jurisdiction to adjudicate is defined as a state's authority to subject persons or things to the process of its courts or administrative tribunals. [FN42] The exercise of jurisdiction to adjudicate does not *639 apply to any person. Countries are rather somewhat restricted by reasonableness. With regard to reasonableness, section 421 of the Restatement (Third) of Foreign Relations points out that '[a] state may exercise jurisdiction through its courts to adjudicate with respect to a person or thing if the relationship of the state to the person or thing is such as to make the exercise of jurisdiction reasonable.' [FN43] In other words, the relationship of a person to a state determines whether the exercise of jurisdiction over this person is reasonable. This idea of a relationship test in United States law differs quite substantially from the principle 'actor sequitur forum rei' [FN44] which is the starting point of any personal jurisdiction analysis in most of the country's international civil procedure laws. [FN45] Applying the 'actor sequitur forum rei' standard to the framer located in Germany would lead to the conclusion that an action against the alleged infringer would have to be brought before German courts. In the United States, however, plaintiffs, considering the home forum as the most favorable, seek to show that a defendant has minimum contacts with the plaintiff's forum. [FN46]

When dealing with the United States' jurisdiction to adjudicate, one has to distinguish between the United States' subject matter jurisdiction and the United States' territorial Jurisdiction.

The subject matter jurisdiction in the United States is highly determined by the existence of two independent court systems, commonly regarded as the systems of the federal courts and the states courts. As a basic rule, subject matter jurisdiction is generally *640 granted to the state courts. However, Article III, section 2 of the United States Constitution allows the legislator to designate specific subject matter to the exclusive or concurrent federal jurisdiction. [FN47] In diversity cases Congress has chosen a nonexclusive federal jurisdiction. The diversity statute prescribes in that respect that '[t]he [federal] district courts have original jurisdiction of all civil actions where the matter in controversy exceeds the sum of value of $ 75,000.00 exclusive of interest and cost and is between . . . citizens of a state and citizens or subjects of a foreign state.' [FN48] More specifically with regard to the hypothetical, the patent jurisdiction statute prescribes exclusive jurisdiction of federal courts in copyright cases. [FN49] Since the hypothetical is clearly a copyright case, which is, in addition, characterized by diverse nationality, federal courts have exclusive jurisdiction regardless of the value of the case and no state court in the United States is permitted to hear the case.

Traditionally, a prerequisite to exercise jurisdiction over a person was the physical presence of this person in the forum. [FN50] However, modern business life, particularly the increased mobility granted by modern means of transportation and communication and commercial transactions involving parties located in the whole country, challenged the standard of the physical forum presence. [FN51] As a consequence, courts began exercising jurisdiction over individuals who were not present in the court's forum. This practice lead to the development of the minimum contact standard.

On the assumption that individuals can be present in a forum without constantly being physically there, the Supreme Court, based on the Constitution's Due Process clause, [FN52] prevents state courts from exercising personal jurisdiction over defendants who lack sufficient contacts with the forum state. [FN53] Hence, state courts have personal jurisdiction over an individual not present in the forum if the individual has established minimum contacts with the forum.

Although the Supreme Court ruled with regard to a conflict between two parties located in the United States, the minimum contacts rule applies as well to a foreign defendant. [FN54] The underlying concept of the minimum contact test deriving from the due process clause is to preserve the sovereignty of the states in the federal system and to grant the right to a defendant to affiliate himself with one or another of those sovereigns. [FN55] Countries are obviously sovereign too so the same reasoning applies with respect to personal jurisdiction of a foreign defendant.

The Supreme court in International Shoe [FN56] distinguished between the two classes of jurisdictional situations with regard to the defendant's contacts. The Supreme Court specifically distinguished between 'general jurisdiction' and 'specific jurisdiction'. [FN57] The former invokes the permission of general jurisdiction over an individual if her contacts with the forum are systematic and continuous enough that the defendant might anticipate defending *642 any type of claim there. [FN58] The latter comes into play when the defendant has minimum contacts with the forum in such a way that she might anticipate defending that particular type of claim there. [FN59] It is needless to say that the German defendants in the hypothetical lack any systematic and continuos contacts with the United States and that they do not anticipate defending any type of claim there.

The lack of general jurisdiction leads to the more difficult question of whether the German defendants from the hypothetical are subject to specific jurisdiction. Here it needs to be analyzed whether the German defendants from the hypothetical might anticipate defending a copyright claim in the United States based on the conduct of framing. When analyzing the specific jurisdiction the Supreme Court considers the foreseeability that a defendant will become subject to suit in a particular forum and, in addition, certain other 'fairness factors'.

In World Wide Volkswagen the Supreme Court held that foreseeability does not just imply 'the mere likelihood that a product will find its way into the forum state." [FN60] Rather, "it is that the defendant's conduct and connection with the forum State are such that she should reasonably anticipate being hauled into court there.' [FN61] The Supreme Court pointed out that a defendant has clear notice that she is subject to suit when she 'purposefully avails' himself of the privileges and the advantage of conducting activities within the forum state. [FN62] It is clear that doing business with the aim of direct pecuniary gain in the forum state enhances the possibility of becoming involved in suit and renders it foreseeable for the defendant to become subject of a law suit. [FN63] The contact inquiry in a case involving a foreign defendant may consider minimum contacts *643 with the United States as a whole, rather than with any particular state. [FN64] Where there is an express contract forming the basis for the party's business in the forum, state foreseeability is easily to be inferred. [FN65]

Relying on its ruling in International Shoe, where the court held that once minimum contacts between the defendant and the forum state have been established, it has to be determined whether the assertion of personal jurisdiction comports with the 'traditional conception of fair play and substantial justice', [FN66] the Supreme Court enunciated five jurisdictional fairness factors to be considered in a separate assessment. In appropriate cases courts 'may evaluate [1] the burden on the defendant, [2] the forum State's interest in adjudicating the dispute, [3] the plaintiff's interest in obtaining convenient and effective relief, [4] the interstate judicial system's interest in obtaining the most efficient resolution of controversies, and [5] the shared interest of the several States in furthering fundamental substantive social policies'. [FN67] To render jurisdiction unreasonable the fairness factors need to be compelling to outweigh the fact that an individual has purposefully availed himself of the forum. [FN68] Apart from the five fairness factors, when the defendant is a foreign national, the procedural and substantive policies of the other nation whose interests are affected by the assertion of jurisdiction have to be considered as well on a case by case basis. [FN69]

With regard to the hypothetical, the specific jurisdiction analysis leads to the question of whether the German ISP or the German framer or both could reasonably foresee being sued in a United States court by having purposefully availed themselves of the privileges and the advantage of conducting activities within the United States. More specifically, the question arises of what kind of Internet related conduct justifies the assumption of a purposeful availment.

A number of cases in the United States have already dealt with jurisdiction issues in the context of the Internet. In Compuserve v. Patterson, [FN70] for example, the court found that the defendant had purposefully availed himself of the plaintiff's headquarters forum because the latter was posting the defendant's software and transmitting the fees for downloaded software the defendant had written. In Bensusan Restaurant Corp. v. King [FN71], to the contrary, jurisdiction could not be established due to the lack of purposeful availment. The defendant had created a Web site containing information about its jazz club in Columbia, Missouri, whose name ('The Blue Note') was identical with the name of a jazz club in New York. The plaintiff sued for trademark infringement in New York and the court ruled that a passive Web page just posting information cannot be the basis for purposeful availment. [FN72] The court pointed out that the argument that the defendant 'should have foreseen that users could access the site in New York and be confused as to the relationship of the two Blue Note clubs is insufficient to satisfy due process.' [FN73] In a later case, the Ninth Circuit developed a solution which seems to be mediating between the two opposite approaches the courts in Compuserve and in Bensusan advanced. This solution could be called the ' Interactivity approach'.

In Cybersell v. Cybersell, [FN74] where the plaintiff claimed that the defendant infringed the former's trademark by setting up a Web page under the plaintiff's name 'Cybersell,' [FN75] the Ninth Circuit distinguished between passive and interactive Web pages. A passive Web site, the court held, does nothing more than permitting anyone to access it and to view the information. This can be compared to the placement of a product into the stream of commerce that may be felt everywhere but is not an act purposefully directed toward the forum state. [FN76] An interactive Web site, on the other side, allows users to exchange information and encourages to communicate. [FN77] The opportunity to exchange and communicate information indicates that the content provider directed her activity in a substantial way to the forum. [FN78] Along the same line in other cases district courts also considered that the 'notion of transacting business over the Internet involves examining the level of Interactivity.' [FN79]

With regard to the hypothetical the user can do more than just accessing and viewing the Web site. He can also actively choose to click on the anchor that provides the link to the framed Web site. Instantly after activating the link the framed Web page will be downloaded [FN80] into the end users computer and can be viewed on the computer screen. The framing Web site, therefore, can be considered as interactive. [FN81]

*646 However, those courts distinguishing interactive Web pages from passive ones, and favoring interactive Web pages with regard to the determination of personal jurisdiction, also emphasis that there must be something in addition to plain Interactivity to establish purposeful availment. This additional element is the commercial nature of the exchange of information that occurs on the web site. [FN82] Usually, courts infer the commercial nature of the exchange from either the sales they have generated through the Web page or from e-mail communication directed at the making of a sale. The defendants in those cases allegedly either transacted business over the Internet or solicited business over the Internet. [FN83] Obviously, none of those activities occur with regard to the hypothetical because it is generally not the framing Web page operator's intent to enter into a contract of any kind with an end user who downloads the framing Web page. [FN84] But, nevertheless, the operator wants the *647 end user to access a particular site through her own framing Web site to ensure that her advertisement will be visible on the end users computer screen. For that reason, the framing Web page does not just simply inform the end user that there is the particular site out there, which can be accessed under its particular URL. It also does not offer a plain link to the particular site as a shortcut. The Web page enables the end user to download, or to acquire, the particular Web site directly within the framing Web site. [FN85] A process, which if not authorized by the framed Web page's copyright owner, would amount to copyright infringement. Moreover, seeing the end user's click on the link as a request to get access to the framed Web site and the fact that access is actually granted as the immediate positive response to the request allows to infer that the end user and the Web site operator actually communicate with each other, even though, admittedly, in a very limited technical fashion. Nevertheless, the opportunity to request a download and then to actually acquire the copyright protected information by initializing the communication seems to be sufficiently equivalent to the exchange of information recognized by the courts to establish purposeful availment. Following this conclusion the question remains, however, whether the exchange of information is also commercial in nature. One surely could argue that the Web site operator as the if framer derives her revenue from selling advertisement space to third parties. The more hits her Web page receives the more revenue she will derive. Communication with the end user leads to the receipt of a hit and thus increases the Web page operator's revenue that in turn, renders the communication at least in some sense indirectly of commercial nature. However, the relevant cases indicate the necessity of a direct commercial relationship between the Web site operator and the end user. Therefore, it would certainly be too overbroad to acknowledge calling on a framing Web page as communication of commercial nature. But it appears to be equivalent to communication of commercial nature when, as it is the case in the context of framing, the communication has some sort of an impact on the copyrights of an United *648 States citizen. [FN86]

However, when recognizing the communication in the context of framing as having an impact on the copyright owner's rights and, at the same time, as an equivalent to communication of commercial nature, then the question arises whether the Web site needs to have received a sufficiently large number of hits or whether the Web site's potential accessibility by virtue of being posted on the Internet would be sufficient. In order to counterbalance the lack of the commercial nature of the communication, a sufficiently large number of hits might be required to recognize purposeful availment in the context of framing in an international setting. However, unlike in the real world with respect to real goods, an Internet user does not need to specifically order from a particular provider the information she wants. Moreover, the ordered information does not need to be actively shipped to the customer. Thus, in the context of the Internet, there is no specific interaction necessary between the providing and the consuming parties since the Internet's content is in fact permanently and ubiquitously available to be downloaded. Furthermore, somebody calling on a Web page usually has no knowledge as to the Web page operator's or the ISP's geographic location. [FN87] In the context of pertinent minimum contacts standard, this permanent availability and the lack of knowledge about the party's location as unique features to the Internet, may invoke the argument that it is unnecessary for the user in the forum to interact with the Web site. As well, the potential for interaction with the Web site by virtue of it being posted on the Internet is sufficient to find purposeful availment.

The following, however, will show that the law requires that a framing Web page actually was accessed and that it actually must have been accessed by a number of visitors in order find purposeful availment.

In Maritz v. Cybergold the district court in eastern district of Missouri implicitly addressed the problem and recognized jurisdiction over the defendant. [FN88] The court found that '[b]y simply setting up, and posting information at, a Web site in the form of an advertisement or solicitation, one has done everything necessary to reach the global Internet audience. ' [FN89] The court then concluded that 'while modern technology has made nationwide commercial transactions simpler and more feasible, even for small businesses, it must broaden correspondingly the permissible scope of jurisdiction excerciseable by the courts.' [FN90] By contrast a district court in Arizona also seemed to rely on mere accessibility as such without paying attention to the question whether the Web site actually had visitors or not. [FN91] The court simply concluded that "[the defendant] should not be permitted to take advantage of modern technology through an Internet web page and forum and simultaneously escape traditional notions of jurisdiction.' [FN92]

Both foregoing court decisions indicate that an interactive Web site by virtue of being accessible or, from the content provider's perspective, by the mere invitation to visit a Web site by making it available on the Internet, is sufficient to find purposeful availment to the United States. Following this approach, it is reasonably foreseeable for the German content provider from the hypothetical to be sued in the United States. Since the German content provider's ISP made it possible for the framer to avail himself of the United States forum, the ISP, by virtue of contributing to the availment, also established minimum contacts in that respect. [FN93]

However, considering the applicable Federal Rule of Civil Procedure, it appears to be doubtful whether federal courts can exercise international personal jurisdiction just by virtue of having a framing Web site being accessible through the Internet.

Rule 4 (k) of the Federal Rules of Civil Procedure deals with the territorial limits of state court jurisdiction and with the federal court jurisdiction as well, the latter following the former consequently. Particularly, Rule 4 (k)(1)(A) of the Federal Rules of Civil Procedure states the very principle that '[s]ervice of summons or filing a waiver of service is effective to establish [federal] jurisdiction over the person of a defendant who could be subjected to the jurisdiction of a court of general jurisdiction in the state in which the district court is located.' [FN94] This provision indicates that, as a general rule, the district court situated in that state that would have jurisdiction over the particular defendant, is the appropriate one. [FN95] For that reason, it is appropriate to assess whether and if necessary under which circumstances state jurisdiction can be established against the German defendants. Considering that the framing Web site is ubiquitously available, any state may be affected by the defendant's conduct.

Generally, the state court's jurisdiction is limited to the citizens of the particular state where the state court is situated. Exercising personal jurisdiction in an international context therefore depends on the states so called long-arm statutes that provide rules for the state's right to exercise personal jurisdiction over a foreign defendant. The provisions of those long arm statutes must be satisfied to establish the consequential federal court jurisdiction as well, [FN96] unless the defendant is amenable to the federal court's exercise of the jurisdiction. [FN97]

For example the New York long-arm statute, section 302 of New York Civil Practice Law and Rules, provides two possible *651 bases for asserting personal jurisdiction over a foreign defendant. [FN98] The first possibility makes jurisdiction depend on 'commit[ing] a tortious act within the state.' [FN99] The second possibility is based on tortious acts committed outside the state but causes injury within the state and which demands the defendant's reasonable expectation that its acts have in-state consequences. Furthermore, the defendant must have 'derive[d] substantial revenue from interstate or international commerce.' [FN100]

Applying the second alternative with regard to the hypothetical the German framer puts up her link without deriving any revenue from international commerce at all and neither does the ISP. The framer might only, if at all, gain revenue from selling space for advertisement on her Web page and the ISP makes her living only by selling its service in Germany primarily to German customers. There is no revenue from international commerce to be found and both, neither the framer nor her ISP aim for revenue from international commerce. [FN101] It is, therefore, impossible to apply the New York's long-arm statute in this respect. Furthermore, long-arm jurisdiction cannot be established by applying the first alternative as well because no tortious act will be committed in New York just by virtue of the potential accessibility of the Web site. Moreover, downloading might actually not create a copy in the computer's RAM as legal commentators and European authorities are currently proposing, [FN102] thereby suggesting that framing does not lead to a tort at all. The lack of state court jurisdiction [FN103] necessarily *652 leads to the conclusion that it is impossible to establish federal jurisdiction as well since, applying Rule 4 (k)(1)(A) of the federal Rules of Civil Procedure, state court long-arm jurisdiction is a precondition to establish federal jurisdiction in that particular state. As a consequence, the German framer and her ISP seem to be off the hook.

However, Rule 4 (k)(2) of the Federal Rules of Civil Procedure does provide relief. [FN104] Rule 4 (k)(2) of the Federal Rules of Civil Procedure provides: '[i]f the exercise of [federal] jurisdiction is consistent with the Constitution and laws of the United States serving a summons or filing a waiver of service is also effective, with respect to claims arising under federal law, to establish personal jurisdiction over the person of any defendant who is not subject to the jurisdiction of the courts of general jurisdiction of any state.' [FN105] Rule 4 (k) (2) particularly exposes foreign defendants to federal jurisdiction under two conditions. First, the claim needs to arise under federal law and second, foreign defendants need to have sufficient contacts with the United States as a whole but lack sufficient contacts with any state to support personal jurisdiction under the states long- arm statute. [FN106] Both requirements are easily met since all state or common copyright law is preempted by the Federal Copyright Act, [FN107] and state's long-arm statutes do not provide general jurisdiction as provided in Rule 4 (k) (2). Applying the plain language of Rule 4 (k)(2) of the Federal Rule of Civil Procedure it could be sufficient just to analyze the nature of the *653 claim as a copyright claim and the lacking application of state's long-arm statutes. Consequently, it would be easy to verify minimum contacts and fairness factors based on the pure invitation to access a Web site in the United States. Its potential accessibility in the United States then would rationalize the finding of 'sufficient contacts with the United States as a whole'. [FN108]

However, case law seems to point in a different direction. In Eskofot A/S v. E.I. Du Pont De Nemours & Co. [FN109] a New York district court applied Rule 4 (k)(2) of the Federal Rule of Civil Procedure and held that 'personal jurisdiction may be asserted by courts where a foreign corporation, through an act performed elsewhere, causes an effect in the United States. ' [FN110] It is needless to say that the mere act of providing the link between the two Web sites and storing this link in Germany on a German ISP does not cause an effect in the United States at all as long as no end user in the United States calls on the link. Consequently it appears to be impossible to establish international personal jurisdiction in the context of framing activities just by virtue of the framing Web site being accessible over the Internet. For that reason it is clear that the framing Web site actually needs to be hit by American end users. Apparently, without a hit there would be no effect at all.

Hitting the Web site causes an effect in the sense that it creates copies in the end user's RAM, the Web site becomes visible on the end user's computer screen, and the Web site is also potentially subject for a long term storage in the computer's hard drive. At a first glance, however, it seems to be inappropriate to take just one hit in the United States to establish minimum contacts by virtue of having 'an effect' in the United States Even though it is very much likely that it is foreseeable for the German content provider and her ISP that the link will be, once it is posted on the Internet, accessed anywhere in the world, [FN111] fairness factors, derived from *654 the due process clause, appear as to render personal international jurisdiction for 'just one hit' unconstitutional. [FN112] On the one side, the public interest in the United States to prosecute a 'one hit' claim seems to be invisibly small because, at the first glance, only minimal impact can be found. Moreover, it seems to be too much of a burden for a defendant to come to the United States to defend a case just because the defendant's Web site has been hit there once.

However, there is different impact to be found depending on what the end user actually does with the information she downloaded. Obviously, the impact will be heavy in case the end user not only downloads the information into the computer's RAM but makes copies of it by saving the downloaded information on her hard drive. The copy from the hard drive might be easily disseminated to third parties thereby threatening the author's rights extensively. Under this circumstances, public interest to prosecute a one *655 hit claim is certainly substantial. Given that the defendants know about information on the Internet being ubiquitously available and the possibility that downloaded information can be saved on the hard disk, it does not seem to be too much of a burden for the German defendants to come the United States to defend a claim. In this respect it is relevant what the court in Maritz held namely that '[b]y simply setting up, and posting information at, a Web site in the form of an advertisement or solicitation, one has done everything necessary to reach the global Internet audience.' [FN113] But the outcome of the analysis changes when the end user does nothing more than just downloads the information into the computer's RAM without intent to permanently save it. The impact on the copyright owner's reproduction right is comparatively small and the threat to violate the author's reproduction right is consequently smaller too. To counterbalance the small impact the download into the computer's RAM has, it is reasonable to demand a sufficiently large number of hits. A sufficiently large number [FN114] of hits makes up the lack of a heavily weighing impact on the author's reproduction right due to permanent storage in the computer's hard drive. [FN115]

Consequently, after all, purposeful availment in the context of framing can be established as follows: First, when the information was downloaded into the computer's RAM for a sufficiently large number of times in the United States as a whole. Second, when the information was downloaded into the computer's RAM and stored on the computer's hard drive once.

With regard to the hypothetical it is assumed that the German framer's Web site was visited in the United States by a sufficient large number of end users. [FN116] As a consequence, the framer and her ISP purposefully availed themselves of the privileges of the United *656 States as a whole. Therefore, international personal jurisdiction over the German defendants has been established.

The German defendants still could invoke the forum of non convenience doctrine. The term "forum non conveniens" refers to court's discretionary power to decline jurisdiction when convenience of parties and ends of justice would be better served if the suit would be brought in a different forum. [FN117] In determining whether the doctrine of non convenience ought to be applied, the court should consider a number of relevant factors: (1) the relative ease of access to sources of proof, (2) availability of compulsory process for attendance of unwilling witnesses, (3) cost of obtaining attendance of willing witnesses, (4) possibility of view of premises, and 5) all other practical problems that make a trial easy, expeditious and inexpensive. [FN118]

It does not seem to be the case that framing a Web page opens court's discretionary power to decline jurisdiction. The sources of proof are accessible over the Internet and witnesses are most likely not needed, a fact that renders the availability of compulsory process for attendance of unwilling witnesses and cost of obtaining attendance of willing witnesses irrelevant. There are also no premises to be viewed, except, perhaps, the framing Web site at issue which can be accessed through the Internet anywhere. The fact that the defendants have to travel from Germany to the United States forum and the fact as such that they have to defend themselves before a foreign court might cause some remaining inconvenience. However, considering the world wide accessibility of the Internet, about which the defendants presumably were aware of and which they most likely appreciated, it is reasonable not to permit them the right to invoke the doctrine of the forum of non conveniens.

As an intermediate result it can be concluded that federal courts can exercise international personal jurisdiction in the context of framing when the framing Web site is sufficiently interactive and two alternative conditions are met: First, the Web site has been hit once in case the downloaded information was permanently saved on the computer's hard disk; or second, the Web site has been hit sufficiently often in case the downloaded information was only temporary stored in the computer's RAM. Furthermore, defendants are not able to invoke the doctrine of forum of non conveniens.

B. Choice of Law

After international personal jurisdiction over the German defendants has been established by virtue of finding sufficient Web site hits in the whole United States the question arises whether it is German or United States copyrightlaw that governs the case.

The Copyright Act of 1976 provides no guidelines as to the conflicts of law and neither does the national treatment principle of the Berne Convention. [FN119] The latter basically provides that a foreign author seeking copyright protection abroad has to be treated like a national. [FN120] The fact pattern here is different since it is a domestic author seeking protection in his home jurisdiction against a foreign infringer. For that reason determining the applicable law invokes the general principle of territoriality which traditionally governs the conflicts of law within internationally situated copyright cases. [FN121]

In the context of copyright cases within an international setting the principle of territoriality is traditionally of some importance. It describes the fact that national copyright law is applicable only within the country; that it does not reach across its borders. [FN122] The mirror like counterpart to the principle of territoriality is the principle of the country of protection (lex loci protectionis) stating that the origination and the scope of copyright protection is determined by the law of the country in which copyright protection is actually being sought. This principle is set forth in Article 5 (2) of the Berne Convention. [FN123] Following the principle of territoriality and its mirror-like counterpart the applicable law in copyright cases within an international context is generally the law of that country in which the infringing acts occurred. [FN124] Applying those principles, for example, in a case where somebody legally acquire an original work of authorship in Germany, brings it to the United States, copies it and sells it publicly in the United States, it is without any doubt that United States copyright law applies to solve the case.

However, the Internet's digital environment generally imposes a problem on the traditional territorial choice of law regime, because *659 a single use of a copyrighted work might lead to effects in a number of different countries. A content provider might upload a work on the Internet in one country which is then accessed by an end user who downloads it into her RAM of her computer in a different country. Applying a strictly territorial choice of law regime, one is unable to define which copyright laws exclusive rights may have been infringed because neither the content provider can control who in fact accesses the work she uploaded nor is the end user able to identify the location of the source of the work she browsed and then downloaded. [FN125]

A plausible approach to decide whether it is the broadcasting or the receiving country's law that ought to govern the conflict could be the determination as to whether it is the receiving or the broadcasting activity which in fact weighs heavier as a cause to the infringing act. With regard to framing, where the infringing act allegedly is the creation of a copy in the end users RAM of her computer, neither of the two conducts - sending and receiving--seem to outweigh the other in terms of importance. Both are logically necessary for the creation of the copy in the computer's RAM; without one of the two causes no infringement would occur. Moreover and generally in the context of the Internet, both alternatives come along with disadvantages which are surely not wanted. Applying the receiving country's law a content provider would expose himself to as many foreign copyright laws as there are countries with Internet access. [FN126] Assuming the content provider is eager to avoid becoming subject to suit in one of the Internet countries, she would be forced to consider all the country's copyright laws and to curtail her content to bring it in accordance with those laws. Proceedings like this would take too much trouble and would be insecure too. The other alternative, applying the law of the broadcasting country, has been initially proposed by the Commission *660 of the European Union. [FN127] Its proposal was based on the approach chosen by the EU's Satellite Broadcasting Council Directive [FN128] which applies the broadcasting country theory. Particularly Article 1 (2)(b) of the directive provides in that respect that the authoritative country for acquiring broadcasting rights is the country from where the broadcast is being conducted. Without questioning the plausibility of the broadcasting country theory in the context of satellite broadcasting, [FN129] in the context of the Internet serious problems would come along with it. Content providers could avoid liability by using ISP from a country which affords its nationals a comparatively lesser standard in copyright protection than the content provider's home country does. [FN130] The current system of international copyright protection would be rendered useless with regard to digital communication and exchange of information over the Internet. [FN131]

However, it is necessary to find an approach that reasonably enables the legal system to properly solve the choice of law question and to reach justifiable solutions in the context of the Internet.

Available alternative approaches can be divided into two categories. The first category relies heavily on a traditionally known method usually employed in tort cases within an international setting. The second approach reflects on a new and distinctive cyberlaw.

It is impossible to make a preference in favor of either the receiving country or the broadcasting country. [FN132] Therefore, it may be useful to redirect one's focus from the perpetration of the infringement to the infringement itself, particularly to the author of the infringed original work. Thus, the focus is directed to the place where the victim of the infringing conduct is situated. This approach seems to be well established in the context of torts committed in an international setting.

A tort is defined as 'a violation of a duty imposed by general law or otherwise upon all persons occupying the relation to each other that is involved in a transaction'. [FN133] Copyright law as such imposes a duty on third parties, namely not to violate the author's exclusive rights set forth in the Copyright Act. As a consequence, copyright infringement certainly constitutes a tort. [FN134] It should be, therefore, legitimate to analyze the law concerning international torts and to apply an analogous approach to copyright infringement on the Internet.

There are two underlying theories leading to a determination *662 which law should apply when a tort is committed in an international setting. The first theory with respect to international torts is the traditional one that is based on the 'lex loci delicti' approach. According to the lex loci delicti rule the applicable law is the law of the place where the tort has been committed. Thus, decisive is the place where the last event occurred which was necessary to make an actor liable. [FN135] However, while this rule seems to be an adequate one in the real word it is not in cyberspace. In cyberspace it is difficult to identify where on the Internet events actually occur. The location might be any country that is on-line. Therefore, the lex loci delicti approach turns out to be inappropriate to serve as a sound choice of law regime in cyberspace. [FN136] The second theory determines the applicable law by finding out the most significant relationship the participating parties have to a particular country. [FN137] It deserves further inquiry as to the question whether it provides a feasible basis for the choice of law decision.

To determine the most significant relationship the Restatement (Second) of Conflicts of Laws provides in section 6 (2) that in case there is no statutory directive concerning the choice of law the factors relevant to the choice of the applicable rule of law include:

(a) the needs of the interstate and international systems,

(b) the relevant policies of the forum,

(c) the relevant policies of other interested states and the relative interests of those states in the determination of the particular issue,

(d) the protection of justified expectations,

(e) the basic policies underlying the particular field of law,

(f) certainty, predictability and uniformity of result, and

*663 (g) ease in the determination and application of the law to be applied. [FN138]

Section 145 (2) of the Restatement (Second) of Conflicts of Laws furthermore provides what kind of contacts are to be taken into account when applying the principles of section 6. [FN139] Those contacts include:

(a) the place where the injury occurred,

(b) the place where the conduct causing the injury occurred,

(c) the domicile, residence, nationality, place of incorporation and place of business of the parties, and

(d) the place where the relationship, if any, between the parties is centered. [FN140]

Section 145 (2) (a) and (b) of the Restatement (Second) of Conflicts of Law basically resemble the traditional rule of the lex loci delicti: when the place where the injury occurred and the place where the conduct causing the injury occurred is identical it follows naturally that the law of this particular place ought to be applicable. [FN141] But it becomes more difficult to determine the applicable law when the place of injury and the place of cause are different from each other.

The difficulty in determining the applicable when the place of cause and place of injury are different can be illustrated by the following example: an American tourist buys a poisoned apple in Rome, carries it back home to Texas, [FN142] eats it and becomes sick. The injury clearly occurs in Texas whereas the cause was set in Rome invoking the question whether Texas law ought to apply or *664 Italian law. Logically, there are two approaches leading to a solution of this conflict. The judge might either choose between the place of injury and the place of cause or she might combine both places, choosing the law alternatively. In fact there seems to be no greater justification for choosing either the place of injury or the place of the cause to determine the applicable law. The former protects the valid confidence of the victim in her particular environment and the latter protects the actor who might have acted in accordance with the law applicable at her environment. Section 145 (2) (a) and (b) of the Restatement (Second) of Torts actually does not provide any help in that respect since both place of injury and place of cause seem to weigh equally without favoring one or the other as being the decisive contact.

However, considering section 6 (2) of the Restatement (Second) of Conflicts of Law, it appears to be the fact that all the factors relevant to the choice of the applicable rule of law point to one direction which is to always protect the victim of a tortious act, respectively to apply the law which favors the victim's claim in general. It is without any doubt that, as a matter of policy, justified expectations, and predictability, it is usually the victim who should be protected. The victim usually has no means to protect itself or to avoid the harmful event whereas the actor at least to a certain extent is able to control her actions and to consider future effects caused by her anticipated conduct. That makes it reasonable to apply the law of the victim's forum and toput the burden on the actor even though her conduct might be legal at her home country. [FN143]

Considering fairness standards toward the defendant it seems to be necessary to employ a foreseeability test with regard to the question whether a content provider could reasonably foresee that *665 her work will be accessed in a particular country. [FN144] In the context of the Internet, however, the use of the medium as such strongly implies foreseeability in this respect. A content provider is certainly aware of the fact that a work, once uploaded, is ubiquitously available in every country that has access to the Internet. Given this awareness, it must be concluded that it is the content provider's intent to actually make her work ubiquitously available. As a consequence, she can, respectively does, reasonably foresee that her work will be accessed anywhere. Particularly in case the content provider derives her revenue from operating a Web site either by granting access to it for a fee or by publishing third party advertising for a fee the content provider foresees the accessibility. In both cases the content provider depends on as much access as possible regardless from which country the access originates. [FN145]

However plausible an approach based on traditional rules with regard to the shortcoming of the principle of territoriality is, applying those rules for resolving conflicts of law problems in cyberspace is subject to fundamental criticism. This fundamental criticism finds its expression particularly in the proposal of a distinctive cyberlaw.

David R. Johnson and David Post have proposed a new dimension with regard to the applicable law in cyberspace. Their proposal is meant to overcome the traditional approaches for determining jurisdiction and choice of law. [FN146] They analyze that the traditional methods have their justifying basis in the real physical world where control over physical space, and people and things located there is a defining attribute of sovereignty and statehood. There is a correspondence between physical boundaries and jurisdictional boundaries which reflects a strong relationship between physical proximity and the effects of any particular behavior. [FN147] Cyberspace, say Johnson and Post, has no territorially based boundaries anymore and physical locations almost entirely lack the power to control activity in Cyberspace. [FN148] Moreover, the effects of on-line activity in cyberspace are not tied to geographically proximate locations either. [FN149] Thus, traditional principles are not able to function as proper guidelines in the virtual world because cyberspace without any doubt 'undermines the relationship between legally significant (on-line) phenomena and physical location.' [FN150]

The proposed alternative would be to recognize cyberspace as a distinct place with its own (electronic) boundaries. [FN151] When taking cyberspace as its own distinct place one could overcome the difficulties determined by a regime of inconsistent copyright laws *667 following the principle of territoriality. [FN152] Inconsistent copyright laws could be replaced by new doctrines particularly tailored to cyberspace. [FN153] Already in the real world much creative expression is entirely independent of an incentive structure and has more to do with building up recognition in the community and the accumulation of reputational capital. [FN154] Moreover, seeing cyberspace as a place with its own boundaries would facilitate the development of new doctrines of implied license and fair use with regard to the transmissions and creations of copies incidentally caused when browsing and caching the Internet. Based on its own cross-boundary 'spaceness' cyberspace could also provide means to resolve disputes which might occur within its boundaries based on the countries willingness to delegate authority and sovereignty to a self governing cyberspace body. [FN155]

It is obvious that distinctive cyberlaw does not yet exist. There might be some efforts around like the adoption of netiquette [FN156] providing core rules with respect to the style people on the Internet communicate with each other. [FN157] Moreover, an on-line mediation and arbitration center for domain name disputes under the roof of the WIPO as well. [FN158] But it is needless to say that those efforts, even though they might point into the right direction, *668 do not make up a distinctive cyberlaw solving actual problems in determining 'real' jurisdiction and choice of law issues in Internet related copyright cases with international participation. It seems to be very likely that the only promising way to create a truly effective cyberlaw will be the administration of multilateral treaties such as TRIPs or the WIPO administered Berne Convention and the Copyright Treaty from December 20, 1996. [FN159] The latter at least partially deals with Internet related aspects of technology, although not as to jurisdiction and choice of law issues, [FN160] and TRIPs which brings Berne's minimum standards of IP protection into the WTO regime of trade liberalization and, for the first time in the history of multilateral IP treaties, provides an enforcement procedure to its provisions. [FN161] The multilateral copyright treaties could serve as a basis for coordination of the multitude of interests of the parties connected to the net, thereby considering the different cultures, legal standards and values of the participating parties. [FN162] Even though it will be extremely difficult to internationalize copyright with regard to distinct cyberlaw, [FN163] maintaining the idea of a distinctive cyberplace with its own boundaries will, at the same *669 time, also facilitate the necessary process of identification and addressing pertinent problems to eventually create a smoothly working system.

The lack of distinctive cyberlaw on the one side and the necessity to provide "justice' in a truly existing conflict between parties with conflicting interests consequently leads to the conclusion that traditional choice of law concepts ought to be applied. Following the analogy of tort principles in an international copyright dispute, discussed earlier, it appears to be quite easy to determine the applicability of United States copyright law. The victim, the author of the framed Web site, resides in the United States and the victim in an international tort case deserves protection. Because distinctive cyberlaw is lacking and the principle of territoriality looking for the place where the conduct causing the infringement took place does not provide a decisive help either, the traditional victim approach should be analogously applied in international copyright cases in the context of the Internet. Since, with regard to the hypothetical, the victim is a United States resident the United States Copyright Act of 1976 ought to be ruling the case. [FN164]

C. Intermediate Result

As an intermediate result it can be concluded, with regard to the hypothetical, that United States federal courts can exercise personal international jurisdiction over the German defendants because minimum contacts to the United States as a whole can be established. Furthermore, United States copyright law governs the case because the victim of the allegedly infringing act is a United States resident. The traditionally in international copyright cases used choice of law rules, which are based on the principle of territoriality, do not apply.

The determination whether framing is copyright infringement particularly points to the question whether the work of authorship being framed is protected by copyright law. In case it does not, no copyright infringement occurs and a framer cannot be held liable in the first place. Analyzing whether a work being framed is copyright protected invokes the question as to the requirements of copyright protection in general. These questions entail what the subject matter of copyright protection is and in case the framed work falls within subject matter of copyright law, whether the act of framing infringes one of the author's exclusive rights.

A. The Subject Matter of Copyright

The subject matter of copyright is defined by section 102 of the Copyright Act of 1976. [FN165] According to section 102 (a) works of authorship include in particular literary, musical, pictorial, and audiovisual works as well as sound recordings. [FN166] Section 102 (a) furthermore prescribes that copyright protection subsist only "in original works of authorship fixed in any tangible medium of expression . . . from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine." [FN167]

The language of the statute invokes two questions. The first question refers to the standard of 'originality' a work of authorship needs to have to qualify for copyright protection. The second question emerges from the significant characteristics of the Internet and refers to the problem whether original works of authorship transferred over the Internet are fixed in a tangible medium pursuant to the language of the statute.

The term originality refers basically to the fact that an original work of authorship does not have to be novel but rather independently created and not copied from other works. [FN168] But it is not just the independent creation of a work in itself that confers copyrightability. Copyrightability can be claimed only in the expression the work of authorship and not in its underlying ideas. [FN169] This aspect of the standard of originality is usually referred to as the idea/expression dichotomy. Furthermore, the work has to reflect a minimum degree of creativity.

The idea/expression dichotomy points out that underlying ideas of a work can not be copyright protected at all. Following the copyright clause of the constitution, [FN170] which provides that copyrights might be granted to promote the progress of science and the useful arts, it is impossible to grant a copyright for an idea since it would confer a monopoly to the author and thereby inhibiting the promotion of science and the useful arts.

*672 The idea/expression dichotomy found early application in Baker v. Selden. [FN171] There, Selden was the author of a book that described a new way of accounting. He developed an account book ledger design that incorporated his new idea. Baker copied the account book ledger and Selden sued him for copyright infringement. [FN172] The court held that Selden's book was subject of copyright protection but not the account book ledger because the latter itself just represents the idea which then found its expression in the detailed descriptions of the book. [FN173] In other words, Seldan teaches that the written expression is copyrightable, the idea - the account book ledger as such - has to remain unprotected to ensure that it cannot be monopolized by the person who 'invented' the idea, thereby inhibiting the promotion of science and the useful arts. However, in cases where the idea and its expression is inseparable, 'copying the expression will not be barred, since protecting the expression in such circumstances would confer a monopoly of the idea upon the copyright owner'. [FN174] The non-copyrightability of ideas and expressions being inseparable is commonly referred to as the merger doctrine.

The idea/expression dichotomy was explicitly recognized in section 102 (b) which provides that "[i]n no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.' [FN175] Denying copyright protection to any idea or process does not mean that a written description of the idea or process cannot enjoy copyright protection, [FN176] unless, invoking the merger doctrine set forth in Baker v. Selden, the number of possible expression of the idea is *673 limited.

Traditionally, American courts granted copyright protection according to the 'sweat of the brow' doctrine. [FN177] Courts particularly relied on section 5 (a) of the Copyright Act of 1909, [FN178] provided categories under which a work might be copyrighted. One of these categories was '[b]ooks, including composite and cyclopaedic works, directories, gazetteers, and other compilations.' Section 5 (a) of the Copyright Act of 1909 was seen as a specification as to the works 'in which copyright can be claimed." [FN179] Applying the "sweat of the brow' doctrine courts conferred copyright protection already when plaintiff had expended some deal of labor regardless whether the work does contain some sense of creativity at all:

The right to copyright a book upon which one has expended labor in its preparation does not depend upon whether the materials which he has collected consist or not of matters which are in publici juris, or whether such material shows literary skill or originality . . . . The man who goes through the streets of a town and puts down the names of each of the inhabitants, which their occupations and their street number, acquires material of which he is the author. [FN180]

The Supreme Court overruled the "sweat of the brow' doctrine in 1991 in the Feist case. [FN181] Rural, the plaintiff in Feist, was a telephone company which distributed the yellow and the white pages listing of Rural's subscribers in an alphabetical order together with their towns and telephone numbers. Feist, the defendant, was a publishing company that specialized in areawide telephone directories *674 and its yellow pages covered also Rural's service area. Both parties competed vigorously with each other for advertisement. To compile the white pages Feist asked Rural to license its white page listing what Rural refused to do. [FN182] As a result, Feist used the listings without Rural's consent and incorporated them into its white pages. [FN183] Rural sued for copyright infringement and the District Court, relying on caselaw, held that telephone directories are copyrightable. [FN184]

The court acknowledges the copyrightability of compilations in general and refers to the idea/expression dichotomy and states that in absence of original written expression only the compiler's selection and the arrangement may be protected and that raw facts may be freely copied. [FN185] Holding facts as non-copyrightable, the court states, is in accordance with the Copyright Act of 1976, which provides in section 102 (b) that "copyright protection for an original work of authorship [does not] extend to any idea." [FN186] Even though a compiler may use the selections and arrangements others have used as a starting point, thereby indicating, that novelty is not required, the court points out that the selection and arrangement of the facts has to be done independently and that the selection and arrangement needs to display some level of creativity. As a result the court explicitly dismissed the sweat of the brow doctrine. [FN187] The opinion does not clearly define the threshold of creativity. However, the court ruled that names, addresses, and telephone numbers are merely non protectible facts and that Rural's typical and obvious telephone listing--name, town, and telephone number--as well as the arrangement of the listings in alphabetical order does not meet the minimum creativity requirement necessary to confer copyright protection. [FN188]

*675 Consequently, framing a Web page does not constitute copyright infringement when the Web page lacks a minimum of creativity as required by the Feist case. The determination of the originality of a work of authorship, however, has to be made on a case by case basis. Since it is just a spark of creativity which is needed to make a work eligible for copyrightability, it would be good advice to treat a doubtful case as copyrightable matter rather than non-copyrightable.

The way that the Internet works and information is communicated through the Internet heavily depends on storing information in the Computer's memory devices, particularly the ISP's and cache providers hard disk. [FN189] Following the definition of the term 'fixation' provided in section 101, the storage of the information in those memory devices does qualify as fixation pursuant section 102 (a). [FN190] Section 101 states: 'a work is fixed in a tangible medium of expression when its embodiment in a copy . . . is sufficiently permanent or stable to permit it to be perceived, reproduced, or otherwise communicated for a period of more than a transitory duration '. [FN191] It appears to be rather settled law that original works of authorship, stored in the hard disk of a computer, are fixed pursuant section 106 (a), because it goes without saying that a original work of authorship, stored in the hard disk of a computer, is a stable copy of the original work which can, with the aid of a machine, be perceived for a period of more than a transitory duration. [FN192] Courts have long recognized that a ROM [FN193] is a tangible means of expression and that a computer program imprinted *676 in a ROM is fixed in a tangible medium. [FN194]

B. Infringement of Author's Exclusive Rights

The owner of the copyright of the work stored on the Internet server's hard disk may invoke his exclusive rights granted by the Copyright Act when his work is being framed. Author's exclusive rights might be infringed either directly by the content provider, the framer, and/or the end user, or indirectly by the Internet service provider who has contributed to an alleged copyright infringement. Author's exclusive rights are expressed in section 106. [FN195] An author has the exclusive right to do and to authorize: the reproduction of the copyrighted work in copies; the preparation of derivative works based upon the copyrighted work; the distribution of copies of the copyrighted works to the public by sale or other transfer of ownership, or by rental, lease, or lending; and the display of the copyrighted work publicly. [FN196]

Direct Infringement is based on the notion that somebody's conduct directly infringes authors exclusive rights. With respect to the hypothetical, the framer being the person setting up the framing link in the first place is naturally the focus of the analysis.

The right to reproduce an original work of authorship is considered to be the most fundamental rights of all the author's exclusive rights provided in section 106. [FN197] Since section 106 (1) refers to 'reproduction of copies,' the section has to be read in conjunction with section 101. The latter provides that '[c]opies are material objects . . . in which a work is fixed by any method now known or later developed, and from which the work can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device.' [FN198] The significance of this provision in the context of framing is obvious. The conduct of framing must constitute a material object in which the framed work is fixed and from which it can be perceived, either directly or with the aid of a machine. Since the framed Web site will finally appear on a computer screen it seems to be, at a first glance, an easy task to qualify framing as a violation of author's reproduction right. But this turns out to be different when taking a closer look.

It has already been stated that digitized works stored in a webserver's hard disk fulfill the fixation requirement and generally constitutes a copy in a tangible medium. [FN199] However, the conduct of framing in itself may not create a copy at all. It is rather arguable that the framer merely creates a link between his Web site to the Web site that is going to be framed in purely technical fashion. However, linking does not create the fixation of the framed Web site on the server's hard disks and, therefore it does not create a copy. One might argue that a copy of the framed Web site occurs when an end user calls on the framing link. But this copy at most comes into existence in a place related to the user, most likely in his computer, and not in a place which related to the framer. [FN200]

However, the technical prerequisite of 'framing' is 'linking', namely to establish a link from the framer's Web site to the framed *678 one. When 'linking' would constitute copyright infringement, the framer could be directly held responsible. [FN201] But even when analyzing the technical aspects of linking no creation of a copy can be found. The hypertext link employed by the framer contains only the URL address of the linked Web site technically implemented in the Web site's HTML code. The URL address merely serves as a locator indicating where on the Internet the desired information is available. As such the URL address does not contain a copyrightable expression. [FN202] Considering the functional aspect of the URL address, it appears likely to be a system or method of operation which, when activated, causes the browser software to jump from one Web site to another. Nevertheless, a fact or a method of operation is not copyrightable. [FN203] Moreover, it is widely assumed that by uploading a Web site the content provider grants an implied license to link. [FN204] Consequently, the hypertext link itself never constitutes a copy or otherwise processes any of the data from the linked Web site and the author's exclusive right to reproduce is not implicated. [FN205] Consequently, the framer does not infringe the author's reproduction right by the mere act of' linking' the latter's Web page to his framing Web site.

Section 106 (2) grants to the copyright owner the exclusive right 'to prepare derivative works based upon the copyrighted work.' [FN206] The definition in section 101 provides that a derivative work is a work "based upon one or more preexisting works . . . in which a work may be recast, transformed, or adapted." [FN207] Courts are split with regard to the question whether an adaptation needs to contain new copyrightable expression to qualify as a derivative work. [FN208] In the case of framing, however, it is not necessary to choose sides since the outcome of the analysis is not determined by the question whether the framed Web site does add originality or not.

The purpose of the adaptation right is to put the copyright owner in the position to control more than simply verbatim forms of copying. [FN209] Framing does produce more than simply a verbatim copy of the framed work. It incorporates the framed work, embodied in the Web site, into the framer's Web site and makes both Web sites appear as one entity, using the underlying work as it has been framed. Furthermore, a derivative work does not need to be fixed to cause a copyright infringement. [FN210] That a derivative work does not need to be fixed might lead to the conclusion that the link *680 itself, the URL address, represents the derivative work.

The allegedly infringing frame is apparently not fixed as long as it is not activated and has no immediate impact on author's exclusive right. However, the framing link represents a dormant, potential copyright infringement that becomes real once the link is activated. This situation of a dormant copyright infringement is analogous to the situation the plaintiff was in Midway. [FN211] In Midway the plaintiff manufactured video game machines with data for the game stored on printed circuit boards. The defendant sold printed circuit boards for video games that sped up the plaintiff's video game when properly inserted in replacement of one of the plaintiff's circuit boards. The court ruled that the circuit board sold by the defendant was a derivative work. [FN212] Analyzing the courts ruling concerning the defendant's speed up version it has to be pointed out that it did not work by itself but rather in conjunction with the plaintiff's video machine. Obviously, the derivative work was not fixed within the device the defendant sold. It was merely dormant and became active only when it was inserted in the plaintiff's video machine. The situation with regard to the framing link is similar. Once it is uploaded it merely needs to be activated to cause a copyright infringement assuming that activating the link creates a copy of the framed Web site surrounded and therefore adapted by the framing Web site somewhere in cyberspace. [FN213] As a consequence, the framing link constitutes a dormant adaptation analogous to the adaptation the defendant's device caused in Midway.

However, considering the legislative history, a work being just dormantly of infringing quality does not qualify as an adaptation. Legislators were intending that 'the infringing work must incorporate a portion of the copyrighted work in some form.' [FN214] Courts followed this incorporation requirement. In the Lewis Galoob *681 Toys, Inc. v. Nintendo of America, Inc., [FN215] the defendant, Galoob, manufactured an add-on device to alter features of one of the plaintiff's, Nintendo copyrighted video game. [FN216] The add-on device was to be inserted between a game cartridge and the Nintendo Entertainment System and did not alter the data that were stored in the game cartridge. The defendant argued that the alterations only occurred in the computer processor and therefore necessarily had to remain unfixed. The court concluded that a derivative work must be fixed to be copyright protected, but a derivative work does not have to be fixed to infringe the copyright of the underlying work. [FN217] On this basis, the court held that Galoob's device could only work in conjunction with Nintendo's video game and therefore did not supplant, duplicate, or recast Nintendo's work. [FN218] Another court's holding in the Vault Corp. v. Quaid Software Ltd. [FN219] is similar. There the plaintiff Vault developed software to protect other software from becoming copied. Quaid the defendant developed a program disabling Vault's protection software by loading a computer program into the computer's memory. Quaid's program also shared a thirty-character sequence with Vault's sequence. The court held that the infringing work must incorporate a portion of the copyrighted work in some form. [FN220] It particularly saw the thirty-character sequence as an incorporation of the copyrighted work. [FN221]

Applying the applicable case law it has to be concluded that the mere conduct of framing--creating a link by connecting two URL addresses to each other--does not incorporate the framed Web site into the framing one. Furthermore, the dormant copyright infringement, which is represented by the framing link, constitutes no impact on the author's exclusive rights. The framer's conduct *682 can be better described as a conduct which grants the ability to download the framed Web page, [FN222] but as such framing in sense of technically providing the link does not yet incorporate the framed Web site into the framing one. For that reason, the conduct of framing does not infringe the author's exclusive right to prepare derivative works.

Pursuant to section 106 (3), the copyright owner has the exclusive right 'to distribute copies . . . of the copyrighted work to the public by sale or other transfer of ownership, or by rental, lease, or lending.' [FN223] Thus, the statute grants the copyright owner the exclusive right publicly to sell, give away, rent, or lend any material embodiment of his work. [FN224]However, the mere link between two URL addresses connecting the framed and the framing Web sites does not create the necessary material embodiment of the framed Web site containing the protected work. [FN225] The lack of material embodiment of the work in the connecting URL addresses leaves no room for the presumption that framing infringes author's exclusive distribution right because distribution actually does not take place at all.

Section 106 (4) grants an author the exclusive right to perform a copyrighted work publicly. [FN226] Section 106 (5) confers the exclusive right to publicly display a copyrighted work. [FN227] In both cases the copyrighted work must be either a literary, musical, dramatic, or choreographic work, or a pantomime, a motion picture, and other audiovisual works. [FN228] To 'display' a work is 'to show a copy of it, either directly or by means of a film, slide, television image, *683 or any other device or process or, in a case of a motion picture or other audiovisual work, to show individual images nonsequentially.' [FN229] To 'perform' a work is 'to recite, render, play, dance, or act it, either directly or by means of any device or process or, in the case of a motion picture or other audiovisual work, to show its images in any sequence or to make the sounds accompanying it audible.' [FN230]

In terms of a motion picture the language of both definitions clearly indicates that the 'performance' of a motion picture or other audiovisual work requires 'show[ing] its images in any sequence' and that the display of a motion picture or other audiovisual work requires only the showing of 'individual images nonsequentially'. [FN231] Regarding motion pictures and other audiovisual works, 'display' is limited to individual pictures of an audiovisual work (stills) whereas 'performance' covers the audiovisual work as a whole or any sequence of it. [FN232] In terms of nonaudiovisual works like literary, musical, and dramatic works the exclusive display right will be infringed when a manuscript or printed version of the work is being transmitted. [FN233] The impact on framing on the exclusive display and performance rights is the same, so that the following analysis applies to both the exclusive right to display and the exclusive right to perform.

As it has already been pointed out, the display of a copyrighted work as well as its performance needs to be done 'in public' to qualify as infringing conduct. Section 101 provides the definition of 'to perform or display a work publicly.' [FN234] One might argue that *684 the Internet is not a place according to the meaning of the statutory language since the Internet is not a place open to the public where people are gathered. However, case law indicates that audio transmission to hotel rooms for the benefit of the guests qualify as transmission to the public [FN235] as well as video transmissions to four people inside a closed room. [FN236] Therefore, the transmission of a work via the Internet clearly encompasses the transmission to the public pursuant to section 101. [FN237]

*685 However, following the language of the definition of 'digital transmission' in section 101 linking a Web site by creating a connection between two URL addresses does not constitute a transmission. Section 101 defines a 'digital transmission' as a transmission . . . in a digital or other non-analog format. Assuming that the definition of a digital transmission applies not only with regard to section 106 (6), it nevertheless indicates that a (digital) transmission is characterized by a transfer of some information. [FN238] Establishing the framing link as a conduct lacks such a transfer. It merely provides the technical link between two Web sites without actively transferring or communicating information as long as the link is not activated.

As an intermediate result it has to be stated that the conduct of framing does not infringe any of the author's exclusive rights set forth in section 106. Neither does the conduct of framing lead to the creation of a copy or to a derivative work, nor to a display, performance, or distribution of a copyright protected work. Consequently, a person, who is engaged in framing a third party's Web site, can not be held directly liable for his conduct under copyright law.

The end user of a framed Web site is the person who actually calls on the site and views it on his computer screen. In the context of the Internet it is worth discussing whether the end user by virtue of using the Internet interferes with the author's exclusive rights.

The reason that the end user's conduct--calling upon a particular Web site-- triggers the question whether this conduct interferes with the author's exclusive reproduction right is based on computer technology. The use of any modern computer application necessarily involves the creation of a copy in the computers RAM. [FN239] In the context of the Internet the browser software 'loads' the Web page from the servers host computer into its RAM. Once the information is uploaded it becomes available for the computer's processor and is finally visible on the user's screen. The significance of a RAM is its lack of capability to store information permanently. Once the computer is turned off the information disappears from the RAM and remains unrecoverable. The information is likewise lost when new information is loaded into the RAM going along with the user's particular application requiring the RAM space formerly occupied. The RAM, therefore and in contrast to a hard disk or a ROM, is a temporary storage device.

It has already been pointed out that section 106 (1) provides the right to reproduce an original work of authorship, speaks of 'reproduction of copies,' and is to be read in conjunction with section 101, which defines the term 'copy'. [FN240] The definition provides that '[c]opies are material objects . . . in which a work is fixed by any method now known or later developed, and from which the work can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device. ' [FN241] The definition of 'copy' indicates the requirement of 'fixation' as well defined also in section 101. Section 101 provides in that respect that "a work is 'fixed' in a tangible medium of expression when its embodiment in a copy . . . by or under the authority of the author, is sufficiently permanent or stable to permit it to be perceived, reproduced, or otherwise communicated for a period of more than a *687 transitory duration.' [FN242] Unlike a hard disk, which allows permanent and stable storage of information, the RAM is arguably not as permanent and stable to allow the information to be perceived for a period of more than transitory duration. Therefore, the Computer's RAM is a temporary storing device.

The Ninth Circuit in MAI Systems v. Peak Computer, Inc. [FN243] was the first court to address the question whether the storage of information in the RAM is sufficiently permanent and stable to permit it to be perceived for a period of more than transitory duration. [FN244] The plaintiff MAI, manufactured computers and designed software to run those computers, which it licensed to its customers. The software license allowed MAI customers to use the software for their own purposes including the loading of the software into the computer's RAM. The license prohibits the use or copying of MAI software by third parties and the software includes operating system software. The defendant Peak, maintained and repaired computer systems for its clients. The process of maintaining and repairing necessarily involves turning on the computer, consequently running MAI's operating software by storing it into the computer's RAM, thereby checking whether computer and software are functional. Running the operating software also allowed Peak's technician to view the systems error log that enables him to diagnose a problem. MAI claimed that Peak, by virtue of loading the software into the computers RAM, created a copy that allegedly infringed MAI's exclusive reproduction right. [FN245] Peak argued that 'this loading of copyrighted software does not constitute a copyright violation because the copy created in RAM was not fixed.' [FN246] The court, without really discussing the issue and without weighing different positions, basically "find[s] that the copy created in the RAM can be ' perceived, reproduced, or otherwise *688 communicated,' [and] hold[s] that the loading of software into RAM creates a copy under the Copyright Act.' [FN247]

This decision has sparked fundamental criticism among legal commentators. They pointed out the legislative history to the Copyright Act which indicates that '[t]he definition of 'fixation' would exclude from the concept purely evanescent or transient reproductions such as those projected briefly on a screen, shown electronically on a television or other cathode ray tube, or captured momentarily in the 'memory' of a computer.' [FN248] The language easily supports the finding that Congress did not consider RAM copies to be fixed. [FN249] In connection with the language in the legislative history, a further argument was made dealing with the fact that Congress enacted section 117 (1) and (2), which exempts from liability the copying of a computer program 'created as an essential step in the utilization of the computer program . . . for archival purposes.' [FN250] Apparently, Congress did not intend every digital copy to be an infringement of author's reproduction right. [FN251] Apart from that, the court in MAI became criticized for not discussing the 'transitory duration' prong of the fixation test [FN252] and not analyzing preexisting cases that have suggested that RAM copies are not fixed. For example, in Apple Computer v. Formula International [FN253] the court stated that "RAM can be simply defined as a computer component in which data and computer programs can be temporarily recorded." [FN254] This utilizing programs in a computer "would be only a temporary fixation. It is a property of RAM that when the computer is turned off, the copy of the programs recorded *689 in RAM is lost.' [FN255] However, there are a number of rulings which are consistent with the court's approach in MAI [FN256] as well as there is scholarly authority and the United States government suggesting that RAM copies are fixed. [FN257]

Transferring the ruling in MAI into the context of the Internet one might argue that browsing the Internet necessarily involves the making of copies in the RAM thereby creating a copy in the RAM that in turn violates the copyright law. The argument was made that the MAI ruling precludes people's right to read or grants the copyright owner an unknown exclusive reading right. [FN258] This argument seems to assume that the simple act of reading information in the real world is actually free. But this is not the case. In the 'real world' consumers have to pay for any kind of information retrieval, regardless of whether it is for books, newspaper, magazines and so forth. Access to those sources always was and still is connected with a detriment. Even borrowing books from libraries is impossible when the user does not pay either directly by submitting a fee for every single use or indirectly by paying a membership fee, tuition, or by paying taxes, so that libraries actually have *690 to acquire the books on the market in exchange for money. One might say that I could get a book from my friend and read it for free without any copyright law interference. But even in this case the same reasoning applies: it is at least the friend who had to pay for the book in the first place thereby securing the author's deserved remuneration. [FN259] Obtaining a copy from the Internet does not involve any monetary transfer at all and potentially deprives the copyright owner of the revenue. Comparing the act of browsing the Internet with the conduct of free browsing a magazine or a book to facilitate the decision whether the book or the magazine is to be bought also misses the point. The hard copy might be subject to browsing but it cannot be taken away without paying. This is different with a virtual copy from the Internet. Once a copy is downloaded it can be easily stored on the hard disk and then distributed in an unlimited number either as a virtual copy or printed out as a hard copy in exactly the same quality as the original copy. Therefore, the free reading argument has to be rejected in favor of the court's ruling in MAI and its subsequent court decisions. [FN260]

*691 The outcome of the analysis is that the end user, by loading the framing Web site into his RAM, creates a copy of this Web site and at the same time, incidentally creates a copy of the framed Web site as well. This leads to the conclusion that the end user infringed the reproduction right attached to the framed Web site.

Following the Digital Millennium Copyright Act, [FN261] Congress made clear that a hardware maintenance service does not violate the reproduction right when it runs a computer program for the sole purpose of repair and maintenance. [FN262] Therefore, MAI v. Peak became irrelevant leaving room for a new evaluation of the question whether the creation of a copy in a computer's RAM leads to a copyright infringement or not. Considering the narrow exemption created by Congress and the generally prevailing fear about the violation of copyrights in the digital context, however, it is not unlikely that courts will consider RAM copies as a violation of the copyright holder's reproduction right.

Since in the context of the Internet one usually has to deal with overlapping copyrights, [FN263] it might be the case that the end user of a framed Web site does not only violate the reproduction right. However, concerning the distribution right it is obvious that clicking on a framing Web site does not constitute the distribution of the framed Web site to the public. It is just the end user who views the site; the public does not take part. The same reasoning applies in connection with the public performance and display right; clicking on and viewing the framing Web site is not done publicly.

However, it might be the case that the end user creates a derivative work since she actually transforms or recasts the framed Web site by activating the framing site and letting the framed Web site appear on the screen. This question naturally invokes the discussion of two issues. First, whether a derivative work needs to add new copyrightable expression to the underlying work and second, whether it is sufficient as to the creation of a derivative work that the end user does nothing else than clicking on the provided link without actively participating in the conduct of recasting and/or transforming the underlying work. In this respect it is probably appropriate to say that courts demand that the creator of a derivative work participates in the creation of that work at least to some extent. After all, in Mirage [FN264] the creator of the derivative work had to mount the photographs on the tiles, thereby showing that the creator is somehow substantially involved in creating the derivative work. It is, however, questionable whether this sort of active participation in the creation of a derivative work can be shown when an end user does nothing more than clicking on a link incorporated in the framing Web site. Perhaps, clicking on the link is not sufficient to determine the creation of a derivative work. [FN265] *693 However, this problem does not need to be fully discussed and resolved since the standard for infringement is the same for both the infringement of the reproduction right and the infringement of the adaptation right. It, therefore, does not matter which of the rights is actually infringed. [FN266] Since it has already been analyzed that the end user infringes the reproduction right, no further investigation with regard to the adaptation right is necessary.

The end user is not helpless after having established that downloading a framing Web site infringes the author's reproduction right. As a defense she might appoint the first sale doctrine or that the provider of the framed Web site in fact granted an implied license to view the content of the Web site by virtue of posting it on the Internet. A third argument could be that the end user's calling on the framing Web site is a fair use.

The first sale doctrine is codified at section 109 (a) of the Copyright Act and allows the owner of a particular copy of a work to dispose of possession of that copy in any way, for example by selling, leasing, loaning or giving it away without violating the copyright owner's exclusive right of distribution. [FN267]

It could be argued that the end user, by virtue of downloading the framed Web site, creates a copy of the framed Web site that was lawfully made by its author. Once the author posts his content on the Internet it seems to be reasonable to say she disposed of the possession of this copy that was received by the end user. However, *694 it is undisputed that the first sale doctrine is applicable only to those situations where the owner disposes of physical possession of a particular copy. [FN268] A transmission through the Internet does not create a physical embodiment of the work and no physical embodiment of the copy stored on the host server. The lack of disposal of a physical embodiment certainly leaves no room for the assumption that downloading information from the Internet is protected by the first sale doctrine. But legal commentators have, concerning Internet transmissions, argued that only technically the first sale doctrine does not apply. [FN269] According to this view it is a fact that copyrights overlap and that transmissions over the Internet not only touch on an author's exclusive distribution right but also on his reproduction right. [FN270] Following this point of view, the argument could be made that the first sale doctrine should apply when the transmission is followed by the deletion of the particular file which contained the transferred information on the host server so that the number of copies do not increase after a digital transmission. [FN271] As a consequence, the usual effect of a reproduction would be taken away and author's right to control the number of copies made from his work would not be harmed. [FN272] This argument has been rejected by the United States government stating that the application of the first sale doctrine would nevertheless weaken the author's exclusive reproduction right and that the reproduction *695 right is simply not within the scope of the first sale doctrine. [FN273]

It is certainly not a common practice that ISP's delete files automatically after they were downloaded for the first time, so, with regard to the hypothetical, it can be reasonably assumed that the file containing the framed Web page was not deleted after the end user's download. It goes without saying that the framer has no intent and particularly no opportunity to delete the file(s) containing the framed Web site. As a consequence, the first sale doctrine is not a valid defense for the end user who downloads a framed Web site even under the assumption that the first sale doctrine does not apply only in a technical sense.

The question, whether an implied license can be found, refers to the law of the transference of exclusive rights. The Copyright Act provides in section 204 (a) in this respect that '[a] transfer of ownership . . . is not valid unless an instrument of conveyance, or a note or memorandum of the transfer, is in writing and signed by the owner of the rights conveyed or such owner's duly authorized agent.' Section 101 of the Copyright Act furthermore provides the definition of the term 'transfer of copyright ownership'. Following the language of the definition given by section 101 "[a] 'transfer of copyright ownership' is an . . . exclusive license, or any other conveyance . . . whether or not it is limited in time or place of effect, but not including a nonexclusive license.' The language clearly indicates that nonexclusive licenses can be granted but are excepted from the 'in writing requirement' of section 204 (a) of the Copyright Act. [FN274] Consequently, nonexclusive licenses can be conveyed by oral agreement or can be implied from the conduct of the parties. [FN275]

The case law indicates the grant of an implied license in two situations. In Effects Associates, Inc v. Cohen [FN276] the latter hired Effects,*696 the plaintiff, to improve a film Cohen made. The parties agreed orally that Effects would be compensated for that work. [FN277] After finishing the work and handing it over, Cohen only paid half of the amount she promised to pay. [FN278] However, Cohen used the work by incorporating it into another work to be distributed by a third-party. [FN279] The court found that Effects was the copyright holder but it did not find copyright infringement holding that. The court noted that "Effects created the work at defendants request and handed it over, intending that defendant copy and distributes it.' [FN280] The court concluded that Effects "impliedly granted nonexclusive licenses to Cohen . . . to incorporate the special effects [and] to distribute the film.' [FN281] The general rule behind the court's ruling is clear: 'when an individual creates a work at a request of another, hands it over, and intends for that recipient to copy and distribute it, an implied license for the recipient is created.' [FN282] In another case the United States Court of Federal Claims held that an implied license is created when a person hands over his work to another with no pecuniary expectations and with the anticipation that the work will be copied and published. [FN283]

With regard to the situation in the Internet and the applicable law in mind it cannot be doubted that a person posting his work on the Internet grants an implied license to download and to view it or, in other words, to browse it. [FN284] Authors know in advance that a work, once posted on the Internet, will be available for the millions of users connected to the Internet. Posting a work on the Internet even indicates that the author wants his work to be viewed by as many people as possible unless she disclaims the implied license in terms clearly visible on the homepage or she cancels an implied license in force by putting up an advanced notice on his Web site. [FN285] *697 It seems to be a precondition for the assumption of an implied license in the Internet, however, that the Web site is accessed or visited in the same manner as it has been originally posted on the Internet. The author most likely wants to have his work viewed and consumed the way she created it, not in an altered version whatsoever or in connection with a different work. This precondition is arguably not fulfilled when a Web site containing an original work of authorship is framed. The framed work appears on the user's screen somehow altered and surrounded by a frame showing the content of the framing Web site. Moreover, it seems to be implausible to assume that a copyright holder, without further written notice visible on his Web page, intents to permit framing. Framing implies alteration of the copyrighted worked as opposed to pure linking where the only purpose is to transfer the end user to the linked site which then will be seen without any alterations at all.

However, if there was a common practice among Web site operators to impliedly license framing one could make the argument that any Web site posted on the Internet can be framed. Insofar it is not the Web page operator's intent which is decisive but rather what appears to be common practice with regard to Web page operator's. However, cases dealing with framing currently litigated in the United States point to the fact that there is no common practice among Web site operators to grant implied licenses for framing. [FN286] As a consequence, the idea of an implied license is not applicable in the context of a framed Web site; there is no basis to assume that a copyright owner has impliedly given his consent to view his framed Web site. [FN287]

'The doctrine of fair use allows a holder of the privilege to use copyrighted material in a reasonable manner without the consent of the copyright owner. ' [FN288] The underlying principle of fair use was first addressed in 1841 in Folsom v. Marsh [FN289] where the court sought to distinguish between an excusable unauthorized use of a work and an infringing use. The court stated 'if so much is taken, that the value of the original is sensibly deminished, or the labors of the original author are substantially to an injurious extend appropriated by another, that is sufficient, in point of law, to constitute a piracy pro tanto.' [FN290] In modern days, the fair use argument became an affirmative defense and the last resort against a claim for copyright infringement. As an affirmative defense, the burden of proof shifts from the plaintiff to the defendant who wants to avoid liability. It is his obligation to prove the fairness of the use rather that the copyright owner needs to prove an accused use not being fair.

The common law doctrine of fair use found its modern expression in section 107 providing that 'the fair use of a copyrighted work, including such use by reproduction in copies . . . for purposes such as criticism, comment, news reporting, teaching, scholarship, or research, is not an infringement of copyright." [FN291] In determining whether the use made of a work in any particular case is a fair use, the language of the statute employs four factors to be considered. [FN292] First, "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." [FN293] Second, the nature ofthe copyrighted work. [FN294] Third, "the amount and substantiality of the portion used in relation to the copyrighted work as a whole." [FN295] And last, "the *699 effect of the use upon the potential market for or value of the copyrighted work." [FN296] The uses listed in the statute do not exclude other uses not listed from being recognized as fair use and do not create a presumption that a listed use is fair. [FN297] It is clear that courts have to evaluate all four factors set forth in section 107 in any case to determine whether a particular use is fair or not.

The purpose and the character of the use contrasts commercial uses with nonprofit educational uses. The Supreme Court in Sony Corp. v. Universal City Studios, Inc., [FN298] stated with regard to this contrast that all commercial uses were to be presumed unfair. [FN299] The Supreme Court's subsequent Harper & Row, Publishers, Inc. v. Nation Enterp., decision does not appoint its former presumption but ruled more lenient that a use for commercial purposes usually weighs against a finding of fair use. [FN300] The Campbell v. Acuff-Rose Music, Inc. decision finally overturned the Sony presumption and differentiated the actual character of the use by distinguishing transformative use of the work from a mere duplication. [FN301] In case a user duplicates the work and uses the reproductions commercially the use weighs heavier against a finding of fair use, whereas in case a user makes a transformative use and creates a derivative work the use weighs more in favor of fair use. [FN302] Concerning the mere reproduction feature courts held in a commercial context, for example, that off-campus copy shops that manufacture and distribute photocopies on teachers specification and order are not covered by the fair use doctrine. [FN303] Concerning the hypothetical the end user certainly does not make a transformative use of the work she *700 downloads. He rather creates a reproduction. [FN304] Following Campbell, the first factor in the fair use analysis would weigh against a finding of fair use if the end user's use of the reproduction would be commercial.

In American Geophysical Union v. Texaco Inc., [FN305] the Second Circuit suggested, in order to discern the 'purpose and character of the use,' that the nature and the objectives of the user need to be considered. [FN306] Concerning a user's objectives, a fair use defense will not prevail when the use of the reproduction appears as a form of commercial exploitation. [FN307] The court then concludes that '[t]he greater the private economic rewards reaped by the secondary user (to the exclusion of broader public benefits), the more likely the first factor will favor the copyright holder and the less likely the use will be considered fair.' [FN308] With regard to the hypothetical there is hardly an economic reward for the end user to be found. He just gains free information through the framed Web site she could have gained for free by accessing the framed Web site directly anyway. On the other hand, the end user's creation of a copy is also incidental because technically induced in the computer's RAM. The copy will be most likely deleted as soon as the end user calls on another Web page or as soon as she turns the computer off.

However, browsing the Internet could have some commercial impact, for example, when a company checks out its competitor's Web site to compare and improve its own professional performance. But even in this case, the use seems to be more like an 'intermediate use' which does not weigh against a finding of fair use. With regard to intermediate use in Sega the defendant copied computer code to study the idea contained within that program for the subsequent development of a competing computer program. [FN309] Using a copy for such a purpose, the court held, was intermediate use *701 'only and thus any commercial exploitation was indirect or derivative.' [FN310] As a result, the defendant copied the computer program for a 'legitimate [and] essentially non-exploitative purpose.' [FN311] Whereas the copy is not literally an intermediate one, the same reasoning may apply to the use with commercial impact in the context of framing. Just analyzing a Web site downloaded into the computer's RAM to compare professional performance is arguably an indirect or derivative commercial exploitation and, therefore, essentially non-exploitative.

The end users use with regard to the hypothetical is non- commercial. Therefore, the first factor in the fair use analysis weighs in favor of a finding of fair use.

Courts have held that the nature of the copyrighted work weighs in the copyright owner's favor when works of fiction [FN312] and unpublished works [FN313] were copied. In turn they have held in defendant's favor when factual works [FN314] and published works [FN315] were copied. In the context of framing it might be necessary to distinguish between different kinds of works being framed. Those works might be either factual or fiction, weighing either for or against fair use.

Applying the factual/fiction distinction in Total News, [FN316] the framed works were news sites that by nature present factual works weighing more in favor of a fair use. Applying the published/unpublished distinction in the context of the Internet is more difficult. Perhaps, a work put on the Internet is not published following *702 the definition of 'publication' in section 101. [FN317] The conduct of uploading and downloading a work does not constitutes a transfer of ownership as to the work being downloaded into the computer's RAM the posting does not constitute an offering for further distribution of copies neither. The posting seems more likely to be a public display pursuant to the definition of 'public display' provided in section 101 [FN318] which does not in itself constitute "publication'. The lack of publication clearly weighs against the end user calling on a framed Web site.

However, in the NII paper the United States government suggests that, in the context of the Internet, 'it is quite possible that a court might evaluate whether a work in digital form should be treated differently from a work in a conventional print or other analog form for the purpose of evaluating" the nature of the copyrighted work.' [FN319] This statement indicates that a work posted on the Internet perhaps ought to be regarded as published. The actual availability of the work leaves no room to consider a work put on the Internet as not being published. Consequently, a work posted on the Internet has to be regarded as published pursuant to the definition provided in section 101. Therefore, the use of a work posted on the Internet weighs more in favor of a finding of fair use.

The third factor in the fair use analysis takes the amount of the protected work taken by the infringer into account, measuring the quantity as well as the quality. Concerning the quantity of the portion which was used it seems to be the prevailing view that "[it] generally may not constitute a fair use if the entire work is reproduced.' [*703 FN320] On the other hand, with respect to the quality of the portion being used, even a relatively small portion can cause copyright infringement. [FN321] Consequently, the judge is obliged to decide on a case to case basis how much of the framed Web site has been incorporated into to the framing Web site. For that reason, it is, in the context of the Internet, difficult to state whether the factor 'amount and substantiality of the portion used' generally works more in favor of the defendant or against him. It is, nevertheless, most likely that the framing Web page entirely displays the copyrightable expression embodied in the framed Web page. [FN322] And, furthermore, it seems to be impossible to find a similar justification as to the time shifting purpose in Sony [FN323] acknowledged as a valid justification for the entire reproduction of a copyright protected work. As a result, the third factor in the fair use analysis weighs more against the finding of fair use.

The last factor is the 'economic effect of the use' which is the most important and central factor in the fair use scrutiny. [FN324] This part aims at the commercial impact of the use on the copyright holder. [FN325] The case law dealing with regard to the fourth fair use factor seems to indicate that, once the plaintiff is able to establish that she would have had significantly higher revenues from the work when it had not been copied, this factor weighs heavily in favor of the plaintiff. Thus, a use of a work must not "impair materially *704 the marketability of the copied work' [FN326] and therefore requires the analysis whether, 'if [the challenged use] should become widespread, it would adversely affect the potential market for the copyrighted work.' [FN327]. With regard to the hypothetical, the conduct of framing has two negative effects on the potential market. One is more indirect and the other is more direct.

Internet content providers usually gain revenue from placing advertisements from third parties on their Web pages. [FN328] The amount of money a content provider can demand for his service is determined by the number of visitors his Web site gets. The number of visitors can be figured out by counting the number of 'hits' a Web site receives. [FN329] The higher the number of visitors of a particular Web site is the more valuable the Web site becomes in terms of gaining revenues from advertisement. Since the framing Web site usually also provides third party's advertisement, the conduct of framing adds to the number of content providers competing with each other to do business with third parties willing to place advertisements on the Internet. [FN330] It is a basic economic principle that increased competition in a given market for a given product *705 is likely to bring prices down because the competing parties tend to undercut their competitor's prices. Thus, once a content provider has lowered his prices to meet the increased competition, the competing content provider is forced to match the lower price to retain its customers and to stay in business. Unless the competing content provider is unable to attract more visitors to his Web page, thereby increasing the commercial value of the Web site, the consequence of competition for advertisement, induced by framing a Web site, is a lower price. Consequently, in the future a framed content provider will gain less revenue for the same number of web banner advertising the was able to place before. [FN331]. Less revenue for web banner placement clearly indicates a negative impact on this particular market pointing to the conclusion that the fourth factor in the fair use analysis 'economic effect of the use' weighs against a finding of fair use.

However, one could argue that current case law applying a fair use analysis indicates that the use of a copy and the dissemination of a copy to third parties must be direct and that it is the direct use which is generally regarded as having a negative impact on the copyright holder's commercial interests. In Sega Enterprises Ltd., v. Accolade, Inc. [FN332] the plaintiff Sega was a copyright owner of a computer video system operating with cartridges. [FN333] The defendant Accolade reverse engineered Sega's video game programs by producing a 'copy' to be able to decompile the information stored in Sega's game cartridges. Accolade then loaded the decompiled code in its computer and experimented with it to discover Sega's interface specifications. Accolade recorded the information it found and used it in the second stage to create its own games for Sega's computer video system without copying the Sega's program. The court saw the intermediate copying as such and considered the fact that the decompilation of Sega's software affected the market for its video game system. However the court stated that the market was merely indirectly affected and that consumers are *706 very likely to buy more than one video game. [FN334] To foreclose a competitor from this market would be an attempt to monopolize the market that can not be a justification for resisting the invocation of the fair use doctrine. [FN335] In Sony, Corp. v. Universal City Studios the Plaintiff Universal Studios brought suit against the defendant Sony for providing consumers with VCRs granting the opportunity to tape record copyright protected television shows. [FN336] The court held that the shows were recorded just for time shifting purposes without further dissemination of the copies to third parties. Considering these facts copyright holders would suffer no economic harm. [FN337]

Sega and Sony seem to apply, when discussing a negative commercial impact weighing against fair use, as an underlying principle that only direct use of the copy and the making it available to third parties will be regarded as a negative economic effect of the use. This analysis finds confirmation in Sega Enterprises Ltd. v. MAPHIA, [FN338] which indicated that the direct use of a copy and its dissemination to third parties is generally regarded as providing a negative impact on the copyright holders commercial interests. [FN339]

Comparing the facts in Sega, Sony and MAPHIA, it is clear that the commercial impact of framing is quite indirect concerning the use of the copyrighted work. But courts also seem to start from the general assumption that, as already pointed out, a use is unfair if 'it adversely affects the potential market for the copyrighted work.' [FN340] In the context of the Internet it is a given fact that providing space for advertisement on a Web page basically is the only way to *707 commercially exploit a Web page. [FN341] Thus, the only potential market for the Web page respectively for the copyrighted work is not, as it happens to be the case in the real world, to sell copies of it but to sell space for third parties' advertisement. It does not seem to be reasonable to distinguish between direct exploitation through selling copies of a work at the one hand and indirect exploitation through selling advertisement space 'on' virtually existing copies at the other when this is lit