• Even Better Than The Real Thing: Virtual Superstars And The Right Of Publicity In Cyberspace
• Open Source Software And Risks Associated With Its Use In A Commercial Environment
• Courts Confused By Pop-Up Ads
• First Sales In Electronic Commerce

Mr. Lee is a partner in Manatt's Intellectual Property and Internet Practice Group in Los Angeles, CA. He is the author of Intellectual Property In The Entertainment Industry, which will be published by Glasser Works later this year.

Celebrity heads on naked bodies. Digitally distributed sex tapes or wet T-shirt contest recordings. Photorealistic, digitally animated actors in videogames, motion pictures, or websites. These and other similar actions implicate the newest and least understood of intellectual property rights – the right of publicity. This article discusses the right of publicity and how courts have thus far applied it in cyberspace.

Background
The right of publicity is, in some ways, the most intuitive of intellectual property rights. While copyright and patent law protect what you create, and trademark law protects what you symbolize, right of publicity law protects who you are. Generally, the "right of publicity" describes an individual's right to control commercial exploitation of his or her identity.

The first case to articulate a "right of publicity" was Haelen Laboratories v. Topps Chewing Gum, Inc., which held that major league ballplayers had a commercial, proprietary interest in their names and likenesses, which allowed them to stop a chewing gum company from commercially exploiting their images on baseball cards without their consent. Recognition of the right subsequently spread throughout the United States. Depending on how you do the counting, a right of publicity for living people is recognized in from 28 to 42 states, a right of publicity for deceased people has been recognized in about 18 states, and the right has been accepted by the American Law Institute's Restatement of the Law of Unfair Competition.

As Zacchini v. Scripps-Howard Broadcasting Co. held: "the rationale for [protecting the right of publicity] is the straightforward one of preventing unjust enrichment by the theft of good will. No social purpose is served by having a defendant get free some aspect of the plaintiff that would have market value for which he would normally pay." Many courts and commentators recognize this policy.

The right of publicity generally prohibits unauthorized commercial exploitation of a person's "identity," and many rights of publicity prohibit commercial exploitation of a person's name, voice, likeness, photograph, or signature. However, one can violate a person's common law right of publicity by taking much more indirect indicia of identity than name, voice, likeness, or signature, such as by exploiting a pose, look alike, sound alike, a distinctive object, distinctive phrase, setting, animatronic simulation, signature musical stylings, or performance that evokes an individual to the public.

That does not mean that all such takings violate the right of publicity in all states. The right of publicity is a creature of state law, and what is legal in one state often is illegal in another. For example, New Jersey precludes unauthorized impersonation of a celebrity's performance, while Nevada expressly exempts celebrity impersonators from liability. Kentucky's and Massachusetts' rights of publicity statutes expressly prohibit only use of name or likeness, while New York and California case law also prohibit lookalikes, soundalikes, or more indirect evocations of identity in certain circumstances. There can even be significant differences within a state as to the scope of protection for living or deceased individuals. California, for example, recognizes expansive common law rights for living individuals, but apparently extends only specific, statutory rights to heirs of the deceased.

Choice of law rules also can create difficulties in determining which law applies, and unequal protection of similar laws. For example, the heirs of a deceased New York resident will have no publicity rights within the state of New York because that state follows a strict domicile rule to posthumous right of publicity claims, and recognizes no posthumous right of publicity. However, filing suit in New Jersey or Texas might resuscitate those rights, because those states recognize posthumous publicity rights and have adopted choice of law approaches that might permit their laws to apply to heirs of a deceased out-of-state domiciliary. Similarly, a foreign resident whose image was exploited in the United States might enjoy the benefit of U.S. right of publicity law in New England, which has adopted a "multi-factor" approach to conflicts of laws, but might not be able to do so in California because of some authority adopting a strict domicile approach. These conflict problems can be especially difficult when interstate commerce is involved. Although an injured party can generally file one suit to recover damages suffered throughout the United States under the "single publication" rule, determining which state's law will apply can be difficult, requiring a court to either choose one state out of fifty to apply to the nationwide conduct, or, alternatively, to determine that the law of all fifty states applies.

States also have different terms of protection. New York recognizes the right of publicity only for living individuals, while Virginia recognizes the right for twenty years after death; Pennsylvania recognizes the right for thirty years after death; Florida recognizes the right for forty years after death; Illinois, Texas and Nevada recognize the right for fifty years, Ohio for sixty years, and California for seventy years after death. Oklahoma and Indiana recognize the right for one hundred years after death, and Tennessee recognizes the right indefinitely so long as the identity is being commercially exploited.

Various proposals for a federal right of publicity have been made to eliminate these problems. Both the International Trademark Association ("INTA") and American Bar Association ("ABA") have characterized current state right of publicity law as a "patchwork" of inconsistency. At the INTA's request, the ABA has drafted a proposed federal right of publicity bill for this reason.

The Right of Publicity in Cyberspace
Although the right of publicity is driven by state law, the Internet is an international medium, and computer programs, videogames, and DVDs are internationally distributed products. This gives rise to a host of issues involving personal jurisdiction and conflicts of laws that have not yet been addressed by any reported right of publicity decision. Instead, the limited case law to date has tended to apply straightforwardly the forum's right of publicity principles to the disputes before it, in five different settings.

The first setting involves that most traditional of Internet sports, cybersquatting. One court held that use of an individual's name as a domain name violated state right of publicity law. State and federal statutes also prohibit "bad faith" registration of names as domain names in certain circumstances. The federal cybersquatting statute prohibits bad faith registration of a "name of another living person, or a name substantially or confusingly similar thereto…" If the name qualifies as a trademark, other provisions also may apply. California has a cybersquatting statute that also prohibits bad faith registration of a deceased person's name as a domain name.

Parties can, of course, submit domain name disputes to arbitration. All persons who register a domain name contractually agree to Internet Corporation for Assigned Names and Numbers ("ICANN") arbitration under the auspices of approved arbitral bodies. ICANN regulations generally prohibit bad faith registration of domain names on grounds that are similar to those in state and federal cybersquatting statutes. These proceedings have the advantage of being quick and relatively inexpensive. However, an individual generally must establish trademark rights in his or her name to prevail, and there have been uneven results. See, e.g., Julia Roberts v. Russell Boyd, Case No. D2000-0210 (WIPO 2000) (Julia Roberts successfully retrieves "juliaroberts.com" domain name); Bruce Springsteen v. Jeff Burgar and Bruce Springsteen Club, Case No. D2000-1532 (WIPO 2001) (Bruce Springsteen does not retrieve domain name because of perceived lack of bad faith).

The second area involves all too common "cyberporn." Cyberporn often manifests itself as celebrity heads digitally superimposed on naked bodies, but also includes the digital distribution of sex tapes or images without the permission of the individuals appearing in the tapes or images. Courts addressing right of publicity claims brought against companies engaged in this conduct have, with little difficulty, found that distribution of tapes violates the individual's rights of publicity.

Third, a few courts have addressed right of publicity or related claims involving other Internet conduct. One decision applied the Communications Decency Act ("CDA") to a right of privacy claim and held that a website was not liable for violating an actor's privacy when it accidentally distributed her address and phone number, because of the CDA's statutory immunities. Perhaps because of its perceived expressive nature, the two other decisions that apply right of publicity claims to Internet conduct reflect a heightened sensitivity to First Amendment concerns. See Gionfriddo v. Major League Baseball, 94 Cal. App. 4th 400 (2001) (listing of names and historical information about baseball players on website did not violate the right of publicity because of statutory exception and free speech concerns); Stern v. Delphi Internet Services, 626 N.Y.S. 2d 694 (N.Y. Sup. Ct. 1995) (Internet service provider did not violate Howard Stern's right of publicity when it used his name and likeness to advertise its new service about Howard Stern because the use was incidental to a news reporting).

The right of publicity/free speech interface has been a difficult one for many courts to navigate. They have created up to seven different and often inconsistent approaches, tests, standards and guidelines to address this issue. See generally Mark S. Lee, "Agents of Chaos: Judicial Confusion in Defining The Right Of Publicity-Free Speech Interface," 23 Loyola Law School of Los Angeles Entertainment Law Review 471 (2003).

A fourth area involves the right of publicity and video games. The only reported decisions involve performers who agreed that their movements could be digitized and utilized in a video game, but objected to the scope of use. Ahn v. Videoware held that the right of publicity claim was preempted by the Copyright Act because plaintiffs were complaining about images they had agreed could be recorded in a copyrightable work. Pesina v. Midway held that there was no right of publicity violation because the digital alteration of the plaintiffs' likeness and persona made them unrecognizable. No reported decision has yet addressed a right of publicity claim involving the unauthorized but recognizable use of a person's name or likeness in a video game.

A fifth area of concern, and one that is likely to raise difficult legal issues in the future, involves what could be called the "virtual superstar." Digital technology is on the verge of permitting creation of photorealistic "virtual" actors, either through reuse of existing footage or digital animation, and motion picture producers are actively exploiting this technology. For example, the "Matrix Trilogy" extensively uses digitized actors (with their permission). "The Lord of the Rings: The Two Towers" and "The Lord of the Rings: The Return of the King" both prominently feature "Gollum," a digitally animated character created by special effects artist from the performance of a real actor. "The Hulk" starred a digitally created humanoid creature in the title role. Dreamworks recently announced that it would "sample" and alter old movie footage to add new characters and create new motion pictures, and digitally animated characters are becoming ever more important in films.

This technology could create remarkable opportunities for actors. For example, an actor could render a performance, then have her appearance digitally altered to make her appear 18 to perform in a high school comedy, 25 to play a romantic lead, or 75 to play a character role, regardless of her real age. An actor's height, weight, physique, hair and eye color, features and other characteristics could be digitally manipulated to suit a role or make him or her appear more attractive. An actor's performing career could be considerably extended through such digital manipulation. Indeed, successful actors might wish to create and retain a library of their own digital images, poses, etc., now, for use in the future in the event such technology becomes a reality.

However, the same technology could make real actors obsolete. It could allow motion picture producers to create completely virtual "actors" such as was imagined in the motion picture "Simone," in which a producer used a digitally created "virtual" star to replace a "temperamental" real actor. Or, it could force actors to compete with themselves or deceased actors by allowing motion picture producers to superimpose images of famous living or deceased actors on an unknown who actually performs the role, or in photorealistic animated characters. New detective stories starring a virtual "Humphrey Bogart," or a new sex comedy starring a virtual "John F. Kennedy" and "Marilyn Monroe" could be in our future.

The legality of making films that appear to depict performances of individuals who do not actually render them is unclear under current right of publicity law. Both the U.S. Supreme Court and lower courts have held, for example, that the unauthorized recording and broadcast of an actual performance without the permission of the performer can violate the right of publicity, and a few decisions have held that the unauthorized imitation of a live musical performance can violate the right of publicity. However, some right of publicity statutes include express exemptions for "media" uses such as those discussed here and one federal appellate court recently held, arguably contrary to U.S. Supreme Court authority, that the unauthorized use of a digitally altered photograph of an actor in a magazine feature which made it appear that he was modeling clothing he was not actually wearing did not violate the right of publicity because it was protected "speech."

Given the apparently insatiable demand for content, the increasing expense of hiring well-known live talent, the competitive urge to attract viewers thorough "star power," the improving ability of technology to alter or imitate a real performance, and the legal uncertainties described above, this could become an area of significant controversy in the near future.

Mr. Larner and Ms. Reed are associates, and Mr. Porter is a partner, with Hogan in McLean, VA.

Open source software generally refers to software that is made publicly or freely available to users in source code and object code form. Although such software may be available free of charge, it is protected by copyright and generally is provided subject to license terms that permit use, copying, modification and distribution subject to certain conditions. The most common open source licenses include the GNU General Public License (GPL), the GNU Lesser General Public License (LGPL), the MIT License, the Mozilla Public License (MPL) and the Apache Software License.

While the concept of open source software is not new, the growth of the Internet over the past decade has allowed for ease of software distribution and collaboration and has led to the proliferation of open source software. As a result of the enhanced availability and low cost of open source software, more companies are contemplating its use; however, every company should understand the legal consequences of using open source software before adopting it for internal use or including it in a product for commercial distribution.

Advantages
Many companies and their programmers turn to open source software to save time and money. Much open source software is available to users at no cost and saves the time and expense of developing or purchasing an alternative. In addition, many companies are able to leverage the expertise and collaboration of a much broader spectrum of developers in the open source community than may be available in-house. Access to this broad community of developers allows for faster software development cycles and arguably results in more reliable and secure software products. Open source proponents argue that since the source code is freely available and subject to independent public scrutiny, flaws in the software and underlying code are more likely to be identified and can then be readily fixed.

Risks
While the advantages are appealing, individuals and companies may not fully appreciate the potential consequences and risks associated with using open source software. These software programs are typically distributed "as-is," without (i) any warranties of performance or ownership, (ii) support from the developers, or (iii) indemnities against intellectual property infringement claims. (The actual restrictions in the various open source licenses differ significantly, and the consequences of using a particular item of open source software vary depending on the terms of the applicable license.) Users must absorb the entire risk of the unknown development history of open source components and exposure to potential claims that these components infringe third party intellectual property rights. For example, in March 2004, SCO Group Inc. brought claims against AutoZone and DaimlerChrysler in connection with their use of open source software. Furthermore, companies that incorporate open source components into their own proprietary products are likely to distribute the products under commercially standard agreements that include performance warranties, intellectual property infringement indemnification and maintenance and support obligations – commitments that may be difficult or costly to keep with regard to open source software components.

Companies that incorporate open source components into their proprietary products for distribution should be sensitive to the fact that the most restrictive open source licenses require them to distribute the products into which they have incorporated the open source components on terms that are the same as those of the open source license. Complying with this obligation requires making the source code for the entire proprietary product freely available to customers for their use, modification and redistribution – clearly a requirement that is inconsistent with most companies' business plans. A company that uses open source software without complying with the terms of an applicable open source license is using copyrighted software without permission and is exposed to liability for damages for copyright infringement. In addition, courts have authority to issue an injunction preventing the company's future use or distribution of the infringing work.

Due to a general concern in the market over the uncertainties surrounding open source software, companies involved in mergers, acquisitions, equity offerings, financing and other corporate transactions are increasingly being required to make blanket representations that they are using no open source software and to make full disclosure of any exceptions to these representations. For many companies, making such representations is difficult because the use of open source products has not been consistently documented.

In addition, nearly all open source licenses have copyright and other notice requirements of varying complexity. For a company that uses an assortment of open source software, complying with the various notice requirements can be burdensome. Finally, there is a risk that using open source software may make software programs more vulnerable to attack by hackers and other security risks given that access to the underlying source code is readily available.

Managing Risk
Before deciding whether to use open source software, a company should have the terms of the applicable open source license carefully reviewed and should consider what alternatives exist, how the open source software will be used and what exit strategy is available if an issue later arises requiring the removal of the open source software.

Since many companies are unaware of the extent to which their development teams may have incorporated open source components into software that they use or distribute, companies should have an inventory conducted of all software they are using to identify the presence of any open source components. If any such components are identified, the applicable license agreements should be immediately reviewed in order to verify that the company is in compliance with those terms. If the company is not in compliance, it should act as promptly as possible to achieve compliance or, if compliance is not feasible, to remove the open source components from the company's software.

Following their initial review, companies should develop and implement policies to govern the use of open source software and educate employees on the consequences of bringing open source software into the company, whether for internal use or as part of the development process for software products it distributes. These policies should require review and approval of any use of open source software by an employee with substantial authority and appropriate training to evaluate open source licenses and the feasibility of the company's compliance with their terms. This individual should verify that compliance is feasible before implementation of an open source component is permitted. As with all third-party software they are using, companies should monitor compliance with applicable license terms on an on-going basis.

The Problem
Online advertising companies – such as Claria (formerly Gator Corp.) and WhenU.com, Inc. – distribute "adware" software programs, generally bundled for distribution with other software programs. Once downloaded by a user, these programs, also known as "spyware," monitor the user's Internet activity to determine whether any of the terms, World Wide Web addresses or content being viewed match certain information contained in a directory. If the program finds a match, it identifies an associated product or service category and then determines whether the user's computer should receive a pop-up or pop-under advertisement that is selected at random from a host of advertising clients that match the category of the user's activity. A pop-up display advertisement then appears on the user's computer screen, typically in front of all the windows the user may have open at that time.

To many users, pop-up ads are annoying. However, a foreseeable result of pop-up programs is that a consumer looking to buy toys on-line, who has arrived at a website such as ToysRUs.com, may receive a pop-up ad for a competing toy store. Paradoxically, the user is left with two choices: either close the pop-up ad to access the initially desired website or click through the pop-up ad to the competitor's website. It is this diversion of commerce that strikes fear in the hearts of website operators. Understandably, website operators have tried to fight back, asserting traditional notions of trademark laws and copyright before the U.S. District Courts for the Eastern District of Virginia and the Eastern District Of Michigan. See U-Haul Int'l, Inc. v. WhenU.com, Inc., 279 F. Supp. 2d 723 (E.D. Va. 2003); Wells Fargo & Co. v. WhenU.com, Inc., 293 F. Supp. 2d 734 (E.D. Mich. 2003). On all theories, these district courts disagreed with the website operators. However, the U.S. District Court for the Southern District of New York has sided in part with the website operators. See 1-800-Contacts, Inc. v. WhenU.com, Inc., 69 U.S.P.Q. 2d 1337 (S.D.N.Y. 2003). Furthermore, in the wake of these cases, industry and the legislature have tried to remedy the situation.

Trademark Infringement
Trademark law protects consumers from confusion as to the source of goods or services. To prove trademark infringement, a trademark owner must show that it possesses a mark, that the infringer used the mark in commerce in connection with the sale, offer for sale, distribution or advertising of goods or services, and that such use is likely to confuse consumers. In the U-Haul and Wells Fargo cases, website operators unsuccessfully argued that pop-up software uses the websites' trademarks as its own to drive advertising and customer traffic to competing sites.

The Michigan and Virginia courts, however, disagreed, holding that a pop-up ad is not a use of the underlying website's trademark in commerce because, first, the pop-up ad is contained in its own window, separate and distinct from the window in which the underlying website appears. Secondly, the pop-up ad trademark and the underlying website trademark are displayed side-by-side, which amounts to "comparative advertising," which is not a violation of trademark law. Finally, and perhaps most interestingly, the court provided that use of the underlying website's uniform resource locator ("URL") to trigger the pop-up ad does not constitute placing the trademark in commerce because domain names do not act as trademarks when they are merely used to identify a business entity. In short, use of a URL and trademarks contained therein is nothing more than the business address for the website and not a source identifier of goods.

However, in a departure from the decisions reached in these courts, the New York court held that a pop-up ad triggered by the input of a word representing the trademark of another does infringe that trademark, and trades off the goodwill associated with the mark.

Cybersquatting
"Cybersquatting" occurs when an entity registers a domain name that embodies or incorporates a name or slogan that is a registered trademark, usually in an attempt to "ransom" the domain name to the trademark owner. Although in many instances interference with a web page, including cybersquatting, constitutes a use in commerce, the Virginia Court held that user-driven pop-up ads do not, because a user is not diverted from the underlying website to a competitor's website merely by seeing the pop-up ad. More specifically, the pop-up ad does not hinder or impede access to the desired website by using the underlying website's trademark and does not change the underlying appearance of the website. Moreover, because the pop-up advertising software is installed by the computer user, it is the user who, in effect, is controlling the computer display in its desired way.

Dilution
The doctrine of trademark dilution protects "famous" marks from overuse for goods or services that, as a rule, do not compete with those of the "famous" mark. Generally, to prevail on a claim of trademark dilution, a trademark owner must show that its mark is famous, that the infringer used the mark in commerce after the owner's mark became famous, and that such use dilutes the distinctive quality of the mark of the underlying website. According to the Virginia court, however, for many of the same reasons provided above, even if the underlying website uses a famous trademark as its URL, there would be no dilution because there is no use of the marks in commerce.

Again, the New York court disagreed, holding that the plaintiff, 1-800 Contacts, proved a likelihood of success on its anti-cybersquatting claim.

Copyright
Copyright protects original works of authorship that have been fixed in a tangible medium. Moreover, copyright gives the owner of the work the right to exclude others from making derivative works from that original work, i.e., from modifying the underlying work. Typically, to establish copyright infringement, the copyright owner must demonstrate ownership of a valid copyright and that the constituent elements of the work that are original were "copied," which is broadly interpreted to include the right of display and the right to make derivative works.

Website users have argued that the uninvited and unauthorized placement of pop-up ads on the website screen is an infringement of the copyright in the website screen. The New York court dismissed the copyright claims before it. The Virginia court found that there was no infringement of any copyright because the overlying pop-up ad does not alter the underlying web page in any manner. The hiding of the underlying website by the pop-up ad window is not an infringement of the right to display because the pop-up ad is in a separate window from the underlying website and it has no physical relationship to the window in which the underlying website might appear. By analogy, the Virginia court found this "cover up" to be no different than when a notice generated by the user's computer pops up in front of all the windows the user may have open at the time. Therefore, it is the computer user who controls how windows are displayed on the computer desktop, not the provider of the website. In effect, the court has switched control of the display right from the website provider to the computer user.

The Virginia court also held that the appearance of the pop-up ad does not produce a derivative work. The court reasoned that the pop-up advertisement is an occurrence distinct from the underlying web page. The appearance of the pop-up ad on the computer screen at the same time as the underlying website is a transitory occurrence that cannot be exactly duplicated on another computer. Moreover, although a pop-up ad may modify the user's computer display, like other overlying windows that may appear on the screen, it is not a modification of the underlying screen and therefore would not constitute copyright infringement.

The Solution
While the courts move through the appeal process with respect to these cases, both industry and the government are continuing to take action to limit the use and occurrence of spyware, including adware that generates pop-up ads. For example, there is commercially available pop-up blocking software, and Microsoft has recently announced that it may be adding pop-up blocking tools to future versions of its Internet Explorer. Microsoft is also planning to incorporate pop-up blockers in Windows. Apple's OS X operating system, Unix, Linux and other Linux variants are designed to be more impervious to spyware.

While industry debates how best to address the solution, state and federal legislatures are taking up the issue. Earlier this year, Utah passed anti-spyware legislation with fines and other penalties for downloading spyware, while California is contemplating similar legislation, as is the U.S. Congress. It should be noted that WhenU is challenging the constitutionality of Utah's spyware law.

For now, it appears that pop-up advertisers must be very careful in their use of pop-up ads. According to the Virginia and Michigan courts, there is no refuge in the intellectual property laws of trademark and copyright infringement. However, in New York pop-up ads may be enjoined. To put an end to this confusion all of the cases are up for appeal. In the interim, users can avoid pop-up advertising by reading all agreements before downloading software and rejecting any that allow bundling. Alternatively, they can buy blocking programs. So, until a higher court decides, the only way to block these ads totally is to just say no, install blocking software, wait for legislation or, if you are a website provider, sue in New York or Utah.

(Prior version published by Edwards & Angell, LLP. Reprinted with permission.)

Mr. Hayes chairs Fenwick's Intellectual Property Practice Group in Mountain View and San Francisco, CA. He is the author of Advanced Copyright Issues On The Internet, published by Fenwick.

The "first sale doctrine" of copyright law is codified in Section 109 of the copyright statute, 17 U.S.C. § 101 et seq. Section 109 provides, "Notwithstanding the provisions of section 106 (3) [the exclusive distribution right], the owner of a particular copy or phonorecord lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord." The applicability of the first sale doctrine to "sales" through online commerce is uncertain.

Section 109 pertains to the sale or disposal of "the possession of [a] copy or phonorecord." The statute was, of course, originally drafted with tangible copies in mind. An immediate issue concerns whether an initial "sale" accomplished by an online transmission , rather than the physical distribution of a material object, constitutes a sale of a "copy" that would trigger the application of the doctrine at all. At least one commentator, Kent D. Stuckey, has argued that it does not, and the White Paper published by the Working Group on Intellectual Property Rights of U.S. President William J. Clinton's Information Infrastructure Task Force notes that the issue is uncertain. However, it seems plausible to analogize a transmission in which a complete authorized copy of a work ends up in permanent storage at the recipient's site (i.e., other than a transitory copy in RAM) as the distribution of a "copy" for purposes of the first sale doctrine, at least where it was intended that the recipient "own" the received copy. (In the case of computer programs, copyright owners often distribute copies of the program subject to a license agreement which states that the copy is being licensed, not sold, to the user as a vehicle to avoid the applicability of the first sale doctrine to the transaction.) Such a transaction seems highly analogous to a traditional sale of a copy, except for the distribution vehicle.

One could readily argue that in such instances the first sale doctrine should apply by analogy to permit a purchaser to further transmit his or her copy to a third party, so long as the purchaser deletes his or her original copy from storage, because in that instance, as in the case of traditional distributions of physical copies, no more total "copies" end up in circulation than were originally sold by or under authority of the copyright owner. As one commentator, Raymond T. Nimmer, has noted:

[The first sale doctrine's] balance was gauged over the years… Neither the copyright owner nor the copy owner receives all that it might desire. The balance could be recut today for cyberspace, but no clear reasons exist to do so. Absent that, this balance governs treatment of digital works, whether on the Internet or a diskette. Applying it is relatively simple. A purchaser who acquires a digital product that is not subject to a license has a right to retransfer the copy, make copies essential to use the work, and otherwise act as owner of the copy. If the "copy" is transferred, the transferor must relinquish all copies it possesses. Otherwise, it would in effect be making multiple copies inconsistent with the balance between copy and copyright owners.

Although this argument makes sense in many instances, such as where a buyer has purchased a copy of a book that is delivered electronically, in other instances the policy choices with respect to whether the first sale doctrine should be applied by analogy seem less clear. One such example comprises works that are made available for on-demand usage, such as movies. The copyright owner clearly intends to make such works available only for one-time use by the recipient, and any further retransmission or distribution of the work to third parties would cut into the owner's on-demand market for the work. Yet depending upon the transmission technology used, a "copy" of the work may be made in whole or in part at the recipient's end. Indeed, under MAI Systems Corp. v. Peak Computer, Inc., even the data stored in RAM at the recipient's computer would constitute a "copy." It seems less clear that such "copy" should trigger the first sale doctrine and permit the recipient to further distribute that "copy," even if the recipient does not retain a copy.

As currently codified in Section 109, the first sale doctrine is drafted as an exception to the distribution right of the copyright holder. However, the new rights of transmission and access under the World Intellectual Property Organization ("WIPO") treaties are seemingly broader than the current distribution right under U.S. law. An issue therefore arises as to whether the first sale doctrine should prevail over these new rights of transmission and access, in addition to the right of distribution. Both WIPO treaties contain provisions stating that nothing in them shall affect the freedom of contracting parties to determine the conditions, if any, under which the exhaustion of rights afforded by the treaties will apply after the first sale or other transfer of ownership of the original or a copy of a work with the authorization of the owner. The WIPO treaties thus seem to contemplate that the interplay between the doctrine of first sale and the new rights of transmission and access will ultimately be resolved through implementing legislation.

Although the implementing legislation for the WIPO treaties in the United States afforded the U.S. Congress the opportunity to resolve the ambiguities in the scope of the first sale doctrine as applied to the Internet, the federal Digital Millennium Copyright Act ("DMCA") does not address the issue. One of the proposed bills to implement the WIPO treaties, H.R. 3048, would have added the following new subsection (f) to Section 109 of the copyright statute with respect to applicability of the first sale doctrine to works in digital format:

(f) The authorization for use set forth in subsection (a) applies where the owner of a particular copy or phonorecord in a digital format lawfully made under this title, or any person authorized by such owner, performs, displays or distributes the work by means of transmission to a single recipient, if that person erases or destroys his or her copy or phonorecord at substantially the same time. The reproduction of the work, to the extent necessary for such performance, display, distribution, is not an infringement.

This provision seems to have been drafted to apply to the paradigm situation, discussed above, in which the original sale of a work via transmission in digital format results in a complete copy of the work residing in permanent storage at the purchaser's site. So long as the original purchaser erases his or her copy at substantially the same time, new subsection (f) permits the purchaser to transmit that copy to a third party without liability (including any reproductions, displays or performances that are attendant thereto).

The applicability of this provision to the case of on-demand transmissions for simultaneous viewing or other usage by the original purchaser (such as movies or online games) is not clear. In those instances, as discussed above, it is unclear whether the purchaser should be treated as the "owner of a particular copy or phonorecord in a digital format" by virtue of the initial on-demand download of the work in order to trigger application of the new subsection (f). In any event, this provision was not adopted in the DMCA.

The European Copyright Directive appears to take the position that obtaining a copy of a copyrighted work through an online service does not exhaust the copyright owner's rights in a way that would allow resale or retransmission of such copy. Specifically, paragraph 29 of the recitals to the Directive states the following:

The question of exhaustion does not arise in the case of services and on-line services in particular. This also applies with regard to a material copy of a work or other subject-matter made by a user of such a service with the consent of the rightholder. Therefore, the same applies to rental and lending of the original and copies of works or other subject-matter which are services by nature. Unlike CD-ROM or CD-I, where the intellectual property is incorporated in a material medium, namely an item of goods, every on-line service is in fact an act which should be subject to authorization where the copyright or related right so provides.

(Originally published in slightly different form by Fenwick & West LLP. Reprinted with permission.)