Trade Dress

Lost and Found: Intellectual Property of the Fragrance Industry; From Trade Secret to Trade Dress

Lost and Found: Intellectual Property of the Fragrance Industry; From Trade Secret to Trade Dress
By Charles Cronin* Download a PDF version of this article here.  


The foundations of the modern fragrance industry can be traced to fragrance producers established in the south of France during the sixteenth century.[1] France’s Mediterranean coast offers excellent conditions for cultivating plants whose flowers, fruit, stems, and roots are used to produce fragrances.[2] For centuries fragrance manufacturers located themselves near growers in order to obtain and process the plant materials as soon as possible after their harvest.[3] By the nineteenth century many of the essences produced by these manufacturers were shipped to Paris to be purchased by hundreds of small perfume houses there that mixed them and sold the compounds in branded retail products like fine fragrances, soaps, and cosmetics.[4] Throughout the nineteenth century, the farms, essential oil producers, and manufacturers of branded fragrances, were family enterprises.[5] With small staffs, often comprised of related individuals who spent their entire careers with the same firm, it was relatively easy for these businesses to maintain proprietary information about distillation techniques, the composition of branded perfumes, and other elements of the production process. During the twentieth century, the fragrance industry underwent radical changes. Increasing real estate values in the area of Grasse, coupled with rising labor costs, prompted the sale of land once used to cultivate jasmine and other flowers for more profitable uses like condominium developments.[6] Today, most of the crops used in fragrance manufacture are grown and processed in countries like Algeria and India where land and labor are relatively inexpensive.[7] Most of the small perfumeries in Paris have disappeared or have been consolidated. By the end of the twentieth century, five fragrance and flavor companies – none of them French – had come to supply over half of the world fragrance market.[8] Whereas the perfumeries in Paris in the nineteenth century created their own branded proprietary blends, today most perfumes are developed and manufactured by a few large corporations with branches all over the world.[9] The inexorable consolidation in the fragrance manufacturing industry over the past century has made the remaining fragrance houses more vulnerable to misappropriation of their intellectual property, particularly of fragrance formulas that they develop at significant expense. Members of the close-knit cadres of the small fragrance houses of the nineteenth century worked in one location, and on behalf of one enterprise, their entire career. Today, perfumers, like professionals in other high technology industries, commonly change not only their locations, but also their employers. This itinerancy has engendered an element of unease among fragrance houses as to the security of their most valuable assets: formulas and other trade secrets that can now be readily obtained, copied, and shared by employees with access to the relevant information stored on the company’s servers.[10] Another late-twentieth-century development that has unnerved fragrance manufacturers is the improving accuracy of analytic technologies in revealing a fragrance’s chemical composition. Unlike digital technologies that have unsettled the media industry by enabling surreptitious copyright infringement, chemical analytic technologies do not enable the illegal acquisition or distribution of intellectual property. It is generally considered lawful to use these technologies, not only to obtain the fragrance formulas of competitors, but also to develop competing products.[11] The fragrance business is by no means the only industry that has had to cope with developing analytic and reproduction technologies, or increasingly itinerant employees. For centuries, many industries have struggled to maintain the confidentiality of proprietary business information, and the ongoing viability of these industries has depended in part on negotiating these challenges. Chartreuse liqueur, and Meissen porcelain manufacturers, for instance, effectively confronted such provocations, and may offer perspectives on how the fragrance industry might best come to terms with its weakened capacity to maintain proprietary knowledge. Since the early seventeenth century, monks of the Carthusian Order have produced Chartreuse.[12] In 1903 the French government appropriated the order’s monastery, exiling the monks to Tarragona, Spain, where they reestablished their eponymous liqueur manufactory.[13] The government sold the monks’ distillery in France, along with the Chartreuse trademark, to private investors who attempted to produce the monks’ cordial.[14] In a demonstration of divine justice, the “new” Chartreuse failed utterly in the market because its makers could not determine the complex formula; the Carthusians were careful to leave no trace of their secret recipe and methodology when they were expelled from France.[15] Today the Carthusians thrive in France once again, thanks, in part, to their effective maintenance of this centuries-old trade secret.[16] Their ability to do so stems from the order’s remarkable avoidance of verbal communication, and extremely limited transfer of proprietary information; each monk vows to a life of silence, and only two monks know the formula for Chartreuse at any time.[17] Messien porcelain provides another perspective for how the fragrance industry might face threats to its trade secrets. Early in the eighteenth century Johann Böttiger, an alchemist working for the Saxon king in Meissen, discovered how to make porcelain.[18] Soon thereafter, Böttiger relocated his workshop to Albrechtsberg Castle, atop a high hill, which protected his trade secrets from the predacious eyes of competitors prowling the streets of Meissen.[19] Despite Böttiger’s precautions, his secrets of the materials and manufacturing techniques for porcelain were appropriated, and widely disseminated, within decades of his breakthrough.[20] Böttiger’s motley crew of laborers, artists, and chemists, were notoriously disloyal, tempted by potential financial windfalls from disclosing his secrets, or by establishing competing enterprises implementing them.[21] Despite the loss of its most precious asset, Böttiger’s enterprise still survives as “Meissen Couture” a luxury products manufacturer and retailer that sells an enormous range of products ranging from porcelain to clothing to furniture.[22] This diversification was essential for the survival of the enterprise. It was made possible, however, only by associating thousands of unrelated products to the porcelain on which the company was founded, and for which it is renowned.[23] All Meissen products are branded with the logo of crossed swords with which the company has marked its porcelain since the 1720s.
  The fragrance industry now faces a dilemma similar to that confronting the Meissen porcelain business 250 years ago, when Böttiger’s trade secrets were lost through breaches of physical security measures, and collegial disloyalty. To stanch a similar loss of their proprietary information, fragrance manufacturers could attempt to replicate the Carthusians’ success with Chartreuse. They could adopt the monks’ code of secrecy and silence, and relocate their manufactories to remote locations where they also cultivate the crops used to create new proprietary blends.[24] This solution, however, would be practically, agriculturally, and legally infeasible. Unlike the small cadre of monks whose conduct is governed by regulations that transcend the secular world, the fragrance industry, employs hundreds of thousands of direct and indirect workers, making secrecy practically impossible.[25] Legally, such restrictions would be indefensible, due to the extraordinary constraints on employees’ speech and movement.[26] Therefore, Chartreuse does not offer an apposite potential new business model for the fragrance industry. Meissen porcelain, however, may provide the fragrance industry an encouraging example of the value of reorienting its intellectual property focus. Despite the loss of its greatest asset, the Meissen porcelain business survived, not only by diversifying its merchandise, but also by invigilating over another significant intellectual property asset it has never lost: the crossed swords mark with which it has always branded its goods.[27] Like the secrets of porcelain manufacture, those used to create fragrances have been revealed, or are increasingly vulnerable to discovery, through reverse engineering and disclosure regulation.[28] Meanwhile, fragrances are increasingly being used as a component of trade dress in branding goods and services.[29] It is the corporate customers of the fragrance industry, rather than the fragrance manufacturers themselves, who benefit financially from fragrance trade dress. However, the increasingly widespread application of fragrances in this manner has added value to an expanding number of goods and services, and the fragrance manufacturers should reasonably expect to share in profits generated thereby.[30] This article proposes that while trade secret protections for fragrances have lost much of their efficacy, trademark and unfair competition law may offer currently unrealized legal protection of the use of fragrances as trade dress. Part I chronicles how reverse engineering has undermined the fragrance industry’s reliance, from time immemorial, on secrecy to protect its intellectual property. Part II considers the limited efficacy of copyright and patent protection for fragrances. Part III canvasses the growing practice of using of fragrance as a component of multisensory trade dress, and the potential legal protection of such uses through trademark and unfair competition law. The article concludes by drawing an analogy between the dissemination of fragrances and the performance of musical works, to enhance commercial spaces. It suggests that the fragrance industry might temper the economic injury incurred from the loss of intellectual property protection it once enjoyed by seeking a portion of the economic dividends generated by these enhancements of public and commercial venues.

I. The Fragrance Industry and the Challenge of Reverse Engineering

A. Regulation of Reverse Engineering in the United States and the European Union

In the twenty-first century, trade secrets have become increasingly vulnerable to disclosure, not only because of the ease with which information can be shared, but also because of advances in analytic technologies enabling reverse engineering.[31] Legislatures in both the United States and European Union are aware of this increased vulnerability affecting a broad range of industries, as evidenced in recent legislative proposals.[32] To obtain legal protection as a trade secret, information must be commercially valuable, not generally known, and subject to reasonable efforts to maintain its secrecy.[33] While trade secrets in the United States are not broadly protected under federal statute, they are regarded as intellectual property alongside information protected by patents, copyrights, and trademarks.[34] The recently proposed European Union Trade Secrets Directive, on the other hand, suggests that trade secrets, while intellectual “assets”, should not be protected as “formal” intellectual property rights like patents, etc., but rather as a “complement” or “alternative” to these “classical IP rights.”[35] Unlike owners of patents, copyrights, and trademarks, who are provided the right to prohibit most unauthorized uses of their protected intellectual property,[36] owners of trade secrets, do not enjoy this privilege. While trade secret law may protect a trade secret indefinitely, if another individual or organization acquires the information in question, it is no longer secret, and its original owner cannot prevent others from using it. Moreover, another person may independently develop secret know-how or, more commonly, will learn it through reverse engineering, i.e., by analyzing the composition of an object implementing the trade secret.[37] A trade secret’s vulnerability to reverse engineering depends on both the complexity of the secret and the nature of the product it implements. A material object, like a fragrance, is more tractable to reverse engineering than an intangible product or service, because it provides palpable and otherwise perceptible information. While it may be easier to “crack” trade secrets used in the production of material goods than of immaterial services, it may be more difficult to implement the acquired information, because the value of physical products depends – to varying degrees – upon the materials used in their manufacture. For example, the secret formula for a fine fragrance is more valuable to a company with established ties to suppliers of top-tier natural raw materials than to a start-up sourcing from an unknown grower selling adulterated plant essences. It is generally legally permissible in the United States to acquire trade secrets through reverse engineering.[38] This permissibility is desirable because it forestalls the possibility that trade secret law would provide monopolist protection for innovations, a right under the exclusive purview of federal patent law.[39] Federal legislation has, nevertheless, restricted unauthorized use of information acquired through reverse engineering of certain products. Congress has effected these limitations on reverse engineering by amending the copyright statute to provide sui generis protection for certain products like semiconductor chips, digital content anti-circumvention technologies, and original boat hull designs.[40] These protections constitute legislative “carve outs” from the broad right to reverse engineer a product and use the information learned thereby. In providing these protections, Congress’ purpose was to avert potentially gross unfairness that may occasion market failure, resulting from the easy replication of a technological advancement that may have cost another years of work and hundreds of thousands of R&D dollars.[41] On the other hand, the European Union’s proposed Directive on trade secrets would establish a liberal policy toward the acquisition of trade secrets through reverse engineering, akin to that found under United States law.[42] This approach is somewhat paradoxical: while the fundamental objective of the Directive is to harmonize and strengthen the legal protection of trade secrets across all member states, its permissive approach to reverse engineering would likely weaken extant trade secret protection available under the domestic law of some European Union members, such as Italy.[43] For example, while some states, such as Italy, consider trade secret law as a distinct form of intellectual property, the Directive transforms it into a component of unfair competition law.[44] Furthermore, as argued in a critique from the Max Planck Institute, the Directive’s liberal stance on reverse engineering is particularly troublesome to industries that depend upon innovative products embodying intellectual investment that is not protected as intellectual property.[45] The Institute’s Comments identify fragrance manufacturing as a prime example of such an industry, and suggest that the Directive’s lax approach to the acquisition of trade secrets eventually could result in the evisceration of innovation in this industry and lead to the failure of this market.[46]

B. Challenges to the Fragrance Industry’s Traditional Reliance on Trade Secrets

The modern fragrance industry has a longstanding reputation for exceptional secrecy.[47] Its clandestine nature is an outgrowth of its origins in early medical and pharmaceutical endeavors in France, in which the creators of curative potions and elixirs carefully guarded their formulas.[48] By the nineteenth century, fragrance manufacturing had become largely independent of the pharmaceutical business.[49] Many enterprises had been established in the area of Grasse exclusively for the production of fragrances, particularly perfumes to be applied to the body.[50] Unlike most retail products incorporating materials produced by the fragrance industry today, these goods were considered luxuries to be enjoyed by a few affluent consumers, and produced in small quantities by family-owned enterprises.[51] It was relatively easy to keep manufacturing know-how and formulas secret among the small staffs of these enterprises.[52] Moreover, these teams were often comprised of related employees, most of whom who would spend their entire careers at the company.[53] While the industry still produces luxury products, like fine fragrances, today it is highly consolidated and creates scents used for innumerable consumer products like laundry detergent and hair gel.[54] The largest of these fragrance manufacturers have thousands of employees who commonly move among companies over the course of their careers.[55] The enormous expansion of both the fragrance industry’s now-itinerant workforce, as well as the number of consumer products it affects, has made it much more difficult for fragrance producers today to maintain their trade secrets.[56] This difficulty has been reflected in a flurry of trade secret misappropriation claims made by former employers of perfumers and flavorists who joined competing firms.[57] One of the most acrimonious of these disputes involves the ongoing prosecution by Givaudan, a large Swiss-based fragrance manufacturer, of a claim against its former perfumer James Krivda.[58] The circumstances surrounding this ongoing dispute illustrate both a company’s vulnerability to misappropriation of proprietary information given employees’ itinerancy, and the difficulty of establishing misappropriation without disclosing the secrets themselves. In 2008 Krivda left a vice-president position at Givaudan for a similar appointment at Mane USA, Inc., a direct competitor.[59] Givaudan asserts that in the days immediately prior to resigning, Krivda downloaded and printed from the company’s secure database over 600 proprietary formulas that he brought with him to Mane.[60] At trial Givaudan offered detailed evidence that Mane had capitalized on thirty-four of the formulas that it claimed Krivda misappropriated, by marketing fragrances identical to Givaudan’s under new names.[61] The trial court, however, granted in part the defendants’ motion for summary judgment based on a finding of insufficient disclosure by Givaudan of information about 600 of the additional formulas that Givaudan claimed the defendants had misappropriated.[62] Accordingly, testimony at trial was limited to a small fraction of the information Givaudan claimed Krivda misappropriated. In February 2014 a federal district court jury in New Jersey exonerated Krivda and Mane of all liability, and Givaudan is now seeking a new trial.[63] In prosecuting this claim Givaudan faced a commonplace dilemma of plaintiffs in trade secret litigation. To establish a meaningful claim of trade secret misappropriation, an owner must convey information about the secret both to the court and to the defendant. Once this information is voluntarily disclosed, however, it may no longer be protectable as a trade secret, because its secrecy has been unquestionably compromised.[64] What rattled Givaudan most about Krivda’s alleged theft was that Mane obtained the information without any cost. By purportedly absconding with hundreds of Givaudan’s formulas, Krivda provided its competitor a windfall of valuable information without the reverse engineering costs that would be incurred through legal acquisition. Krivda could have purchased hundreds of products incorporating Givaudan’s fragrances, and worked with Mane’s chemical analysts on the painstaking task of isolating and dissecting them. Given the brief life cycle of most fragrances, however, the market success of an innovative product depends significantly on the potentially dissuasive expense and time lag incurred in reverse engineering and developing a competing product. Therefore, if Krivda provided to Mane the trade secrets Givaudan claims he stole, Mane could not only avoid the temporal and financial cost of reverse engineering, but also produce, within the period of market viability, competing merchandise offered at a lower price. Moreover, because reverse engineering technologies cannot always provide exact and complete information about the chemical composition of a fragrance, a competitor can avoid any potential ambiguities by simply lifting the formula itself.

C. The Impact of Gas Chromatography-Mass Spectrometry (GC-MS) Technologies on the Fragrance Industry

In the popular imagination, consumers have held to the romantic notion that fragrances, particularly perfumes applied to the body, are created from closely held and undetectable formulas. For instance, the conceit of Patrick Süskind’s novel Perfume centers on its murderous protagonist’s prodigious capacity to analyze scents.[65] Tom Robbin’s Jitterbug Perfume, on the other hand, tells the picaresque tale of the arduous quest to discover the formula of an ancient fragrance ultimately revealed to have contained the recherché ingredient of beet blossom essence.[66] In fact, using contemporary GC-MS technologies, one can learn with remarkable accuracy the formula of any fragrance.[67] GC-MS is a technique for separating the components of a vapor by observing the different speed by which each chemical component is expelled from a long tube through which a sample of the vapor is swept.[68] Once the components have been separated, a mass spectrometry apparatus identifies the various separated molecules and their relative volumes in the composition of the vapor.[69] GC-MS technology has disconcerted fragrance houses because it enables practically anyone to obtain a fragrance’s formula swiftly and inexpensively.[70] The most costly component of a fragrance, the formula is typically developed from months, or even years, of research costing hundreds of thousands of dollars.[71] Unsurprisingly, as GC-MS technologies have become increasingly affordable and effective, a parasitical industry has emerged, that manufactures and distributes lower cost “smell-alike” versions of well-known fragrances.[72] Five multinational corporations, four of which originated in Western Europe, dominate the world fragrance market.[73] For years this industrial concentration fostered a tacit agreement among the industry’s largest players. Under this informal understanding, the major fragrance houses would not cannibalize each other by manufacturing competing products based on formulas of a competitor acquired through reverse engineering.[74] Otherwise, a competitor could undercut an innovator’s market by selling products at prices that did not reflect the innovator’s R&D expenditures. The increasing accessibility and accuracy of GC-MS technologies in recent years has tested the stability of this “gentleman’s agreement” among fragrance manufacturers. Moreover, this understanding never extended to the client base of the major fragrance producers, ranging from consumer products giants like Unilever to couture houses like Dior (LVMH), that attach their brands to fine fragrances that are developed and manufactured by large external suppliers.[75] Moreover, GC-MS technologies have provided these clients a new means of negotiating lower prices for the development of new fragrances, as well as those for ongoing supplies of already commissioned products. If a fragrance house balks at the price negotiated by a client for ongoing supplies of a product that it developed for the client, the client could reverse engineer the fragrance, and then buy supplies of it at a lower price from a competitor of the initial supplier. The competitor would have legitimately obtained the formula without incurring the cost of creating it.[76] GC-MS technologies also make it more feasible for clients themselves to produce supplies of perfumes developed at their behest by fragrance houses. This discomfiting potential was realized in 2011, when the luxury conglomerate LVMH built a fragrance compounding facility outside Paris and began producing supplies of concentrates for their popular scents like “Miss Dior Chérie”, “Dior Homme”, and “Kenzo Flowers”.[77] Previously, they had obtained these supplies from Givaudan, Firmenich, and IFF, after these companies had developed the respective formulas.[78] LVMH claimed that the blends that they produced for these brands embodied subtle modifications of those that had been produced by Givaudan, et al., presumably a tactic to avert liability for breaching any contractual obligation to purchase concentrates from the companies that developed the original fragrances.[79] LVMH’s actions were particularly distressing to fragrance houses because they involved the production of successful and well-established perfumes.[80] Profits derived from sale of the liquid blends to produce these goods cover not only the costs of their development, but also the formulation of a constant stream of new proprietary blends that manufacturers use to compete for new business.[81] As the owner of fragrance brands like Givenchy and Dior, LVMH is one of the most significant players in the retail fragrance industry.[82] Moreover, the company is one of the most important clients for fragrance houses, continually commissioning the development of new products that capitalize on their deep R&D expertise. Therefore, fragrance houses are naturally reluctant to alienate themselves from such a powerful client. These companies could contractually preclude clients such as LVMH from certain uses of newly developed formulas, or from obtaining supplies of certain fragrance compounds from other sources. Such terms, however, would be difficult to negotiate in light of the bargaining advantage that analytic technologies now provide to these industry clients. Given the feasibility of legally reverse engineering and independently producing a fragrance, clients would agree not to do so only in exchange for price concessions, or guarantees regarding the ongoing manufacture and quality of a product, e.g., the sourcing of ingredients from a particular supplier. To summarize, over the past few decades, the availability and enhanced capacity of GC-MS technologies have significantly challenged the fragrance industry’s business model. The industry’s most valuable assets, proprietary formulas, can now be legally acquired by anyone with access to a well-equipped laboratory. The effects of the loss of trade secret protection, resulting from use of these technologies, have been compounded by unprecedented calls for greater government regulation of the industry’s products, which could require the public disclosure of the ingredients, or even the formulas, of proprietary fragrance compounds.[83] Now that trade secret protection has been compromised for the fragrance industry, are there other forms of intellectual property for which this industry should seek protection instead?

II. Patent and Copyright Protection for Fragrances

A. Patent

U.S. and E.U. law provides patent holders with a twenty-year monopoly on the manufacture, use, and sale of their inventions.[84] Even those who independently develop or reverse engineer an invention covered by patent are prohibited from unauthorized manufacture, use, and sale of products or services that implement it. However, this sweeping prohibition is tempered by the patent holder’s obligation to disclose, at the time of registration, the composition and functioning of his invention, which information enters the public domain when the patent term expires.[85] Despite the strength of patent protection, the fragrance industry does not rely heavily upon this form of intellectual property for the protection of fragrance formulas, and those of fine fragrances in particular.[86] An invention must be useful to be patentable.[87] Like the jewelry business, the fragrance industry promotes high-end perfumes as pure luxuries. Ascribing utility to these goods tarnishes their cachet of exclusivity, and thereby the economic value associated with entirely discretionary products.[88] Moreover, the fragrance industry eschews the trade-off between patent’s twenty-year term of monopolistic control and full disclosure of the patented invention. This is not only because the market for many of the industry’s high-end products lasts more than twenty years, but also because longevity in the marketplace of some of these products actually makes them more valuable over time.[89] While the fragrance industry does not primarily rely on patents to protect the formulas used to produce fragrances, there have been thousands of applications in U.S. Patent Class 512 covering “perfume compositions”.[90] However, most of these applications are for innovative means for extracting, manufacturing, or delivering fragrances.[91] The relatively few registered patents that protect the formulas of fragrant compounds are grounded in claims of the product’s useful capacity to supplant noxious odors or – more dubitably – to promote physical and mental health.[92] Fragrance companies also rely upon patents to protect some of their most valuable assets: new fragrance molecules that they have developed, known as “captives”.[93] These proprietary molecules typically are not valuable because of the beauty of their scent, but rather for their capacity to create original, safer, or less costly fragrances.[94] The handful of fragrance companies that dominate the world market create and own most captives because only these companies can afford the significant R&D investment required for their creation.[95] Companies that develop patentable molecules may initially manufacture fragrances employing these captives, but much of their profit is derived from selling or licensing them to other fragrance manufacturers that more exhaustively explore and capitalize on their potential.[96]

B. Copyright

Copyright registration is easier to acquire than patent,[97] because the registering author must simply claim that his work is more than de minimus original expression.[98] In other words, an author may register expression already copyrighted or in the public domain, as long as it is his expression produced independently of the preexisting work.[99] Copyright owners enjoy a “bundle of rights” in a protected work, including those to reproduce and perform it, and to create derivative works.[100] The term of copyright is significantly longer than that of a patent, although the protection copyright offers is moderated – particularly in the United States – by “fair use” and other exemptions permitting certain unauthorized uses of copyrighted works for educational and archival purposes.[101] Globally, the scope of copyrightable expression has grown exponentially since enactment of the Statute of Anne in 1710, which established authors’ rights in their books.[102] The margins of protection have expanded, accommodated by evolving copyright statutes that provide illustrative, but not exhaustive, examples of copyrightable works. The French copyright statute, for instance, delineates fourteen exemplary categories of works typically considered works of protectable authorship, but prefaces this list with a broad statement extending copyright protection to all intellectual creations, “regardless of their embodiment, merit, or purpose.”[103] The United States Copyright Statute offers a similarly broad definition of copyrightable authorship, followed by an illustrative list of eight categories of copyrightable works.[104] Because copyrights are easy to obtain and provide lengthy terms of protection, they would appear to be an attractive means of protection for the fragrance industry. Copyrights, like patents, provide owners near monopolies on the use of their protected information, so one could not reverse engineer and reproduce a copyrighted fragrance without authorization from the copyright owner.[105] Moreover, the term of protection offered by copyright is now several times that of a patent, typically providing owners control over their works for the better part of a century.[106] Fragrances are ultimately embodied and perceived as particular combinations of airborne molecules. Nevertheless, fragrances, like pharmaceuticals, may ultimately be reduced to works of information fixed in visual symbols comprising a formula. In this respect – and in others – they are akin to music scores whose visual information is used to produce a performance by which a work of music is typically broadly disseminated, and ultimately perceived as sound. A skilled and patient musician can “reverse engineer” and reproduce a music score from repeatedly listening to a performance.[107] Digital audio technologies can dissect the sounds of performances of even relatively complex musical works and render increasingly accurate scores in traditional music notation.[108] Just as an audio recording of the reading of a book is a copy of a literary work, a music score derived from the sounds of a performance is a copy of the musical work underlying both the performance and the score.[109] Likewise, if we consider man-made fragrances to be copyrightable works of authorship, the dissection and reconstruction of a fragrance, whether by a human with preternatural olfaction or by a mechanical apparatus for molecular analysis, results in a copy of that fragrance, which only the copyright owner is authorized to make. There is no indication, until the latter half of the twentieth century, that perfumers regarded copyright as a means to protect their original blends of fragrances.[110] Apart from the fact that the original focus of copyright protection was literary texts, there was no need for such protection given the difficulty of copying a fragrance by separating the components and determining their role in a particular blend.[111] Given this impediment, fragrance formulas could enjoy perpetual protection as trade secrets rather than merely a term of perhaps fourteen or twenty-eight years as copyrighted works of authorship.[112] By the end of the twentieth century, the breadth of copyrightable subject matter had grown to include works as disparate as fictional characters, and computer programs, far beyond the contemplation of those who promoted authors’ rights in the eighteenth century.[113] At the same time, technologies for molecular analysis had become so advanced and accessible that perfumers could no longer depend upon secrecy to prevent competitors from learning the formulas of their fragrances. Copyright surfaced, therefore, as a potential new means of protecting fragrance formulas, attractive to an increasing number of perfumers who consider themselves authors and artists creating original aesthetic works.[114] Although France is no longer a leading fragrance producer, it remains an influential force in the fragrance industry.[115] Though France has lost much of the agriculture and extraction work associated with the industry, it has retained the expertise for manufacturing fragrances and creating new blends. French ventures have capitalized upon this element of national patrimony, offering education and degrees for the study of fragrance creation.[116] Even today many perfumers at major fragrance companies have trained, at least in part, in France.[117] Not surprisingly then, the most significant débat on whether fragrances are copyrightable expression occurred in France.[118]

C. France and the Netherlands: Odor in the Courts[119]

In the early 1970’s, the French fragrance manufacturer De Laire contracted with the couture house Rochas to create several new fragrances.[120] De Laire agreed to divulge the formulas of these fragrances to Rochas in exchange for Rochas’s promise to purchase from De Laire all concentrates of any new fragrances that the fashion house chose to add to its perfume line. After providing Rochas the formula for one or two perfumes, but subsequently receiving no orders to produce them, De Laire sued, claiming that Rochas had infringed its copyright by producing a fragrance using De Laire’s formula. De Laire’s claim failed when an appeals court upheld a lower court’s finding that perfumes are practical works and therefore eligible only for patent protection.[121] Moreover, because perfumes are not tractable to meaningful and consistent description by those who perceive them, they cannot be considered copyrightable “works of intellect”.[122] The holding reflects a view that perfumes cannot be considered original expression because human olfaction is too crude to perceive and describe fragrances except in broad terms. Fifteen years after the Rochas dispute the French perfume house Molinard created a fragrance marketed as a “smell-alike” of “Angel,” the popular perfume created by Olivier Cresp for Quest International, commissioned by fashion designer Thierry Mugler.[123] When Mugler sued Molinard for copyright infringement, the Paris Tribunal de Commerce discounted the defendant’s argument based on Rochas that as products of industrial technique, perfumes couldn’t qualify as original works of personal intellection.[124] Comparing the formula of a perfume to a music score, the Mugler court suggested that variations among perceptions and reactions to a scent are akin to idiosyncratic receptions among those listening to the same musical work, and do not indicate ineligibility for copyright protection.[125] Several years after Mugler the cosmetic giant L’Oréal sued Bellure, an importer of “smell-alike” fragrances, claiming it was infringing L’Oréal’s copyright in Trésor.[126] Deciding the dispute in favor of L’Oréal, the Paris Cour d’appel noted that the fact that the French Intellectual Property Code does not include fragrances among its list of copyrightable works was not dispositive on the issue of copyrightability.[127] All works of intellect are eligible for copyright protection, even those that might also be patentable, or are not fixed, if they are perceptible and reveal the imprint of the creator’s personality.[128] At the same time L’Oréal was prosecuting its claim against Bellure in France, its subsidiary Lancôme initiated a copyright infringement action against the Dutch perfume seller Kecofa in the Netherlands.[129] Lancôme claimed Kecofa’s Female Treasure was not only a counterfeit of Trésor, but also infringed its copyright. The dispute advanced to the Netherlands’ Supreme Court, which ultimately confirmed the lower courts’ findings that fragrances are copyrightable.[130] The Court also noted that the Dutch copyright statute has a catholic scope of protection, and cannot be read as excluding fragrances that are perceptible, original (i.e., bearing the personal imprint of their creator) and not purely technical (i.e., useful).[131] Back in France, shortly after the conclusion of the Kecofa litigation in the Netherlands, the same Paris appeals court that determined L’Oréal’s Trésor could be protected by copyright reached a consistent conclusion in a claim involving designer Jean-Paul Gaultier’s perfume Le Mâle.[132] Gaultier’s perfume manufacturer, Beauté Prestige International, sued a competitor, Senteur Mazal for infringing the copyrights and trademarks of Gaultier’s perfumes by which it marketed “smell-alikes” sold at prices much lower than those of the Gaultier products.[133] Like the court in the earlier Mugler decision, the Gaultier court discounted the defendant’s argument that variances in human perception of fragrances make it impossible to establish that a perfume possesses the originality required for copyright protection.[134] Responding to this argument the court noted that literary, graphical, and musical works are also perceived variously, but these variations in perception do not undermine the originality of these works.[135] The opinions in these Dutch and French cases, which emphasize the role of the perfume creator, give short shrift to that of the user in determining whether fragrance may be copyrightable. The decision in L’Oréal’s case against Bellure, for instance, expressly found that a perfume could reveal the personality of its creator, and thereby be an original work.[136] Neither L’Oréal, nor any of the other cases, however, addressed the issue of what constitutes the “revelation” of a work, a question that necessarily implicates those perceiving the “revealed” work.[137] The shortcoming of these decisions lies not in their determination that the work of perfumers can be a complex intellectual and aesthetic endeavor akin to writing and painting, but rather in the fact that they do not consider the lack of human olfactory capacity to perceive the complexity of the work rendered from this intellectual investment. The ultimate issue in all of the cases involving the copyrightability of fragrance, therefore, is how perceptible must expression be to be protectable by copyright. Despite the pro-copyright outcomes of the Kecofa, Mugler, and Gaultier cases, the Cour de cassation recently delivered a severe check to those advocating copyright eligibility for fragrances in France, by addressing this issue of perception.[138] In 2006 Patrice Farque was prosecuted for selling counterfeit fragrances at a flea market outside Paris.[139] When the case foundered for lack of evidence Lancôme et al. claimed that by selling imitations of its fragrances Patrice Farque had infringed Lancôme’s copyright in these products.[140] When Lancôme’s case ultimately advanced to the Cour de cassation (commercial section) in 2013, the court determined unequivocally that fragrance is not copyrightable expression.[141] The court separated the act of creating a fragrance from that of perceiving it, finding that while the development of a perfume may involve creative intellection beyond technical know-how, this original thought cannot be broadly communicated because it cannot be sufficiently perceived.[142] With its 2013 decision in Lancôme v. Farque the Cour de cassation dashed expectations that French copyright law might offer the fragrance industry a new means by which to protect its perfume formulas. Courts in the United States have not yet fielded the question of whether fragrances may enjoy copyright protection. In fact, the existing classifications under which works may be registered in the United States would not accommodate an application for a work of fragrance.[143] Accordingly, except in the Netherlands, copyrights, like patents, offer little potential solace to an industry unsettled by the vulnerability of its most valuable intellectual property.

III. Fragrance and Trademark Protection

A. The Expanding Scope of Trademark Protection

Like the sphere of copyrightable expression in the United States, the range of commercial indicators protectable as trademarks increased dramatically in the latter half of the twentieth century. By the end of the century, trademark protection extended to sounds,[144] and even single colors.[145] Moreover, legally protectable visual trademarks were no longer limited to those comprising words and/or designs, but had been extended also to the “trade dress” of products and services.[146] Although U.S. law now protects colors, scents, and sounds, international conventions and other national trademark regimes are typically less accommodating than the United States of these non-traditional marks. For example, the World Trade Organization’s multilateral Agreement on Trade-Related Aspects of Intellectual Property establishes only that combinations of colors may be eligible for trademark protection, and that registration may be conditioned on the mark having already acquired distinctiveness in the market through use.[147] There is considerable variation among national trademark regimes on trademark protection for color marks, and even relatively liberal regimes, like that of Germany, may protect only those single-color marks that have acquired secondary meaning.[148] There is a similar lack of consistency among national trademark regimes with respect to sound marks, stemming in part from the fact that some nations, like Mexico and Brazil, permit registration only of marks that are visually perceptible.[149] While sound marks are not visually perceptible, they can be verbally documented quite accurately. Many sounds, like the iconic NBC chimes comprising the intervals of a rising sixth followed by a falling third, can be represented visually through music notation. Similarly, scent marks can be accurately visually represented using the standard chemical symbols and verbal instructions of formulas, but are less tractable to verbal description than sound marks. Courts have been reluctant to recognize colors, scents, and flavors as trademarks because doing so could allow the initial user to unfairly monopolize the viable marks for particular categories of goods and services.[150] A fundamental purpose of trademark protection is to protect consumers from deceptively labeled goods.[151] However, courts have held that this objective should not limit competition in a market by rewarding early entrants with potentially perpetual exclusive use of a limited number of marks.[152] In particular, scent, flavor, and single-color marks are more readily depleted than design and word marks. This is not because there are fewer potential marks in these classes, but rather because consumers are less able to distinguish among them than among design, word, and sound marks.[153] For example, there are innumerable shades of red, but it is difficult to distinguish readily between Stanford’s “cardinal” and Harvard’s “crimson” without additional verbal or visual information. On the contrary, it is impossible to confuse the different verbal mottos also branding these universities.[154] Consumers would not be served by a prohibition on other universities using the color red as part of their “brand” simply because Harvard was the first to do so. It is desirable, however, to curtail another university’s branding itself with the color red, the motto “Veritas”, and – obviously – the name Harvard, because it curbs potential confusion on the part of consumers, as well as unwarranted derogation that the original Harvard might suffer.[155] Like single colors, scents and flavors are problematic trademarks because the typical consumer cannot distinguish variations in scents and flavors as easily as those of designs and words.[156] While there is an infinite variety of scents and flavors, we tend to classify them, as we do shades of color, using relatively limited taxonomies: “royal blue”, “blood red”, etc. For instance, we use “floral” to describe the scents of hundreds of different flowers, and “spicy” to refer to the taste of hundreds of different piquant flavors. The greater the number of words or designs that are combined in a mark, the more complex and inherently distinctive it is likely to be.[157] This is not true for scents and flavors. Combinations of different scents and of different flavors may yield more complex chemical compounds, but we tend to perceive these not as distinctive new scents and flavors but rather as muzzy blends of the relatively few existing generic categories: “this wine has a citrusy flavor;” “this moisturizer has a vegetal scent.”[158] There are, of course, thousands of varieties of vegetal scents and citrusy flavors but humans cannot readily distinguish among them using a broadly shared taxonomy. Accordingly, because this perceptual inadequacy presents a risk for trademark depletion, most trademark regimes do not accommodate scent marks.[159] International conventions touching on intellectual property, like the TRIPS Agreement and the European Union Trademark Directive, do not expressly permit or prohibit trademark protection for scent marks. However, international courts and national laws have made it difficult or impossible to register them.[160] On the national level the trademark statutes of France and Germany implicitly preempt scent mark registrations by limiting protection to marks that can be visually represented.[161] In 2002 the European Court of Justice (ECJ) considered the question of what constitutes graphical representation of a scent mark.[162] The dispute involved an appeal from the ruling of a German court that upheld the national trademark office’s refusal to register a scent.[163] The applicant, Ralf Sieckmann, had submitted a fragrance claiming it as a mark denoting a range of professional services.[164] As graphical representations of the mark Sieckmann provided a sample of the fragrance in a liquid, the chemical composition of the fragrance (C6H5-CH = CHCOOCH3), and a description of it as “balsamically fruity with a slight hint of cinnamon.”[165] The ECJ determined that none of these submissions, alone or combined, were an effective graphical representation of the scent as a trademark.[166] Although the submissions were visual, they did not enable consumers “… to guarantee the identity of the origin of the marked product or service…by enabling him, without any possibility of confusion, to distinguish that product or service from others…”[167] In other words, these representations are ineffective graphical representations because the appearance of an amber liquid, and the words “balsamically fruity” are so commonplace that when consumers encounter them they will conjure any number of scents (and tastes). Moreover, only a narrow sector of consumers who are trained as chemists might recognize “C6H5-CH = CHCOOCH3” as the fragrant compound Methyl Cinnamate. Consumers’ limited capacity to perceive Sieckmann’s mark through these visual representations would obscure the boundaries of protection trademark registration might provide. This would defeat the purpose of graphical representation, which is required “…specifically to avoid any element of subjectivity in the process of identifying and perceiving the sign,” and would provide overly broad protection that is potentially detrimental both to consumers and competitors of the trademark owner.[168] Moreover, even if consumers recognized the chemical formula for Methyl Cinnamate, as readily as we recognize H2O as that for water, the fragrance mark itself is inherently unstable “…because of different factors which influence the manner in which it can actually be perceived, such as concentration, quantity, temperature or the substance bearing the odor.”[169] Unlike the European Union, the United States does not require that trademarks be represented graphically to be registered. In fact, the United States Patent and Trademark Office (USPTO) has accommodated registration for such marks by establishing a classification for them: Mark Drawing Code 6, “for situations for which no drawing is possible, such as sound.”[170] Nevertheless, while there are many United States trademark registrations for sounds there are remarkably few for scents.[171] The first scent mark registration was issued in 1991, after the Trademark Trial and Appeal Board overturned the USPTO’s initial denial of a trademark registration for a floral scent used to brand sewing thread.[172] Since then, only fifteen additional fragrance marks have been registered, and only four are still “living”.[173] These include the scents of vanilla and citrus for cosmetics and fuel oil respectively, and that of coconut to mark the retail space of a beach apparel seller.[174] Most of the now expired fragrance marks were registered by Smead, Inc., a Minnesota office supplies manufacturer, to protect their branding of their paper products with fragrances like peppermint and peach.[175] The fact that Smead allowed their scent trademark registrations to lapse suggests that the company determined that consumers did not effectively associate the scents with their particular source. Smead might have more successfully developed consumer association between the scent of the paper products and their manufacturer if it had deployed a single fragrance across its entire line of products. While Smead could not monopolize the idea of scenting paper, which stationers have been doing for centuries, it might have monopolized the use of a particular fragrance for paper products. Smead’s use of various fragrances to mark interchangeable products, likely led consumers to associate the fragrance more with the particular variants – scent of peaches for the peach colored file folders, etc. – than with the manufacturer.

B. Growing Significance of Multisensory Trade Dress

Retail sales of consumer goods have always been driven in part by visual cues in the presentation of the merchandise, which comprise not only the distinctive packaging that we identify with specific products, but also the overall visual ambiance in which the products are displayed and sold. While one shops for apparel at Saks, the flattering lighting and carpeted dressing rooms promote lingering, and palliate sticker shock.[176] In contrast, the buzzing fluorescence and concrete floors of Costco generate brisk efficiency for the grim acquisition of huge quantities of “house brand” paper products and similar utilitarian goods sold at “volume discount” prices.[177] In general, the more a product’s economic value stems from its purported refinement and exclusivity (e.g., perfume, jewelry, haute couture) the more its retail sales are tied to visual presentation and other factors extrinsic to the product itself. The cost of presentation and other less immediately apparent external factors, like polite salesclerks, clean washrooms, or a forgiving returns policy, are built into the retail prices of these goods. Accordingly, while Target sells for $80 a two-and-a-half ounce bottle of the eau de toilette of Patou’s Joy, Nordstrom charges $130 for the same item.[178] Retailers also use sound to boost sales, which is not a recent phenomenon. Even before the advent of technologies for recording and broadcasting musical works, department stores engaged musicians whose live performances promoted not only sales of sheet music, but also the sale of other merchandise, once the music had lured customers into the store.[179] Some retailers, restaurants, and hotels still enhance their public spaces with live music performances.[180] However, with the development of recording and broadcasting technologies, this goal is now met mainly through recorded performances of musical works.[181] “Background music” is now so prevalent in commercial spaces in the United States that it is disconcerting to enter a shop or restaurant blessedly free of it.[182] Like the visual décor of commercial spaces, these aural ambiances are now artfully developed and deployed to promote sales of goods and services to specific populations of consumers.[183] This deployment is now so pervasive in retail spaces that the withholding of music and all visual adornment in “big box” stores like Costco, appears to be a deliberate reassurance to customers seeking to purchase goods at “no-frills” prices. Because the sound of recorded music is now so commonplace in retail locations, customers do not listen to it as they might have a century ago. Customers still hear the music, however, and the character of the music becomes a familiar and predictable attribute of a particular commercial milieu.[184] In other words, regardless what music being played, it is being deployed as “muzak”, i.e. an aural enhancement of an environment in which retailers anticipate that we will focus on something else, namely purchasing their goods or services.[185] This ambient music affects the experience and behavior of customers in these retail spaces.[186] Retailers play music to entice customers to linger over merchandise by providing familiar and affirmative “sound tracks”.[187] They tailor the music to appeal to consumers based on factors like age, race, gender, and economic class.[188] While a misogynist rap number broadcast in a Nike store might subliminally flatter the egos of young male customers, it would likely alienate shoppers at Tiffany’s browsing engagement rings and bone china.[189] Accordingly, shopping malls and department stores broadcast varied “soundscapes” in which the disparate “sound tracks” comprising them are regularly updated to conform to the evolving musical tastes of their targeted consumers.[190] Like music, scents have long been used in the marketing of goods and services. The aroma of baking cookies, with its homey associations, has often been used to market houses; the alarming odor of burnt wood has been used to sell fire insurance policies to protect such property.[191] Moreover, developers of scent delivery systems have capitalized upon existing technologies like ink-jet printing and smartphones to create new means of communicating scents, particularly in connection with advertising and entertainment.[192] Marketers, however, increasingly use fragrances not to conjure a narrowly defined good or service, but to mark an overall environment in which retail customers purchase the goods or services of a particular seller.[193] This practice is not limited to sellers of luxury goods and services: today “… marketing using scent is catching on among retailers and in car showrooms, sports stadiums, airports, banks and apartment buildings that seek to distinguish themselves with customers via the deeply influential sense of smell.”[194] For example, the air in the lobbies of all Omni hotels is infused with the scent of lemongrass.[195] Omni anticipates that its repeat customers will learn to associate this scent with comfortable accommodations, and specifically those offered by Omni. Hyatt’s objectives in their use of fragrance are more nuanced than Omni’s. Hyatt scents the air of the public spaces of each of its sumptuous “Park Hyatt” properties with a different and unique bespoke fragrance, which is also used to scent the lotions, soaps, and candles liberally sprinkled throughout the particular hotel’s public and private spaces.[196] Hyatt hopes that because customers will encounter a particular bespoke fragrancs only at the hotel in which it is deployed; they will associate it not with “Hyatt” – which has properties ranging from relatively austere to luxuriant – but with a particular top-tier Hyatt hotel that is purposefully distinct from all other Hyatt properties.[197] Retailers scent their salesrooms hoping to imprint customers’ memories with a positive association between a fragrance and the experience of shopping at their stores. Retailers may do so also to promote sales of the fragrance itself, like Abercrombie & Fitch’s “Fierce.” The distinctive environment of this retailer is a carefully manufactured mix: visual elements like dim lighting and exiguously dressed young clerks; a loud soundtrack of popular music that teenagers find appealing; and an atmosphere constantly infused with the scent of “Fierce,” the retailer’s “signature” scent.[198] As Abercrombie’s website proclaims, “Fierce”, sold as a cologne, body wash, and candles, is “[k]nown as the world’s hottest fragrance…a symbol of masculinity and great American achievement.”[199] In other words, the retailer’s hypertrophic sensory ambiance is intended to evoke that of attractive nuisances like Los Angeles’s Sunset Strip music “clubs” that entice the same customers that Abercrombie targets: just-legal teenagers untethered from their parents, but in possession of their credit cards.[200]

C. U.S. Trademark Protection for Trade Dress

Trade dress is the “total image and overall appearance” of a product, or the totality of elements that “may include features such as size, shape, color or color combinations, texture, graphics.”[201] A product’s trade dress may be a concatenation of elements that are not separately protectable as trademarks, but the amalgamation of these elements is protectable because of its capacity to identify the source of a product or service.[202] In this respect, the whole is greater than the sum of its parts. The trade dress of a product or service might also be the combination of separately protectable marks. If, for instance, a computer manufacturer tagged its products with an image of a lemon, colored them a bright yellow, and imbued their laptops and peripherals with a lemon scent, all three tags contribute to the products’ trade dress despite the fact that each of them might be separately protected as a trademark when applied to computing machinery. Trade dress can now be registered as trademarks on the Principal Register in the United States.[203] Nevertheless, the trade dresses for most products and services have not been registered.[204] This may be because a trade dress, like a trade secret, is typically developed incrementally over time, and its value often becomes apparent to the owner only when another tries to capitalize upon it.[205] Also, trade dress tends to be more protean than word and design marks. Whereas the hairstyle and apparel of Betty Crocker or the Morton Salt Girl needs to be updated only every few decades, the soundscape of a department store must be adjusted to evolving markets far more frequently to retain its potency.[206] Moreover, trade dress often combines non-traditional marks like sounds, colors, and scents that retailers use to create a deeper emotional response in consumers than that engendered by purely visual marks.[207] In recent decades, as these non-traditional marks and trade dress have become more prevalent, U.S. courts have recognized their capacity, like that of traditional word and design marks, to enable consumers to distinguish among sources of goods and services. Two Supreme Court cases, in particular, have established a vastly larger sphere of protectable trademarks than that of half a century ago.[208] In Two Pesos, Inc. v. Taco Cabana, Inc., the Supreme Court determined that trade dress was entitled to the same protection as that afforded to word and design trademarks.[209] The parties were small Mexican fast food chains that deployed a similar décor in all of their restaurants. Taco Cabana was established seven years earlier than Two Pesos, and claimed that Two Pesos infringed its trade dress by using interior decoration confusingly similar to theirs.[210] The plaintiff’s description of their trade dress was more discursive than those typically used for word or design marks, comprising not only specific ornamental objects like “artifacts, paintings, and murals,” but also color schemes and architectural features like “a stepped exterior,” and “bright awnings and umbrellas.”[211] A district court jury found Two Pesos liable for trademark infringement, having determined that that Taco Cabana’s décor was protectable trade dress because it was inherently distinctive, despite the fact that the plaintiff had not demonstrated that its trade dress had acquired secondary meaning.[212] The Fifth Circuit upheld the lower court’s judgment and Two Pesos appealed. The Supreme Court granted a writ of certiorari because of a circuit split; the Second Circuit precedent conflicted with that of the Fifth, holding that, unlike a registered trademark, unregistered trade dress like the plaintiff’s could obtain protection under the Lanham Act only if it were inherently distinctive and had acquired secondary meaning through use in commerce.[213] In Two Pesos, the Supreme Court rejected the Second Circuit’s narrower approach, and upheld the Fifth Circuit’s broader understanding of protection available to trade dress under the Lanham Act. Because trade dress serves exactly the same function as trademarks, the Court reasoned, there is no reason why trade dress should be subject to more stringent standards than traditional registered marks like words and designs, to obtain protection under the federal statute.[214] Three years later, in Qualitex Co. v. Jacobson Prods. Co., the Supreme Court further accommodated trade dress as protectable information, determining that a single color could constitute a valid trademark.[215] The single color at issue was a bilious green/gold hue that Qualitex used on the dry cleaning pads they manufactured.[216] When Jacobson appealed the district court’s ruling that it was liable for infringing Qualitex’s single-color trademark, the Ninth Circuit court overturned this decision, finding that color alone could not qualify for trademark protection.[217] Qualitex, in turn, appealed to the Supreme Court, which granted a writ of certiorari to resolve contradictory precedent from various federal circuit courts on the validity of single-color trademarks.[218] The Supreme Court overruled the Ninth Circuit and upheld the validity of Qualitex’s single-color trademark registration.[219] Like the multifaceted décor indicating Taco Cabana’s restaurant services, the particular shade of green with which Qualitex “dressed” its products distinguished them from those of others, and served no other practical purpose.[220] Moreover, Qualitex had used the color in this capacity for over thirty years, during which it acquired distinctiveness as consumers came to associate it with this company’s product.[221] Qualitex’s trade dress gradually matured into a protectable trademark as it developed secondary meaning through ongoing use in the marketplace.[222] In other words, it is only through ongoing exposure to a single color used to dress a product that consumers will begin to link that particular color with that product’s manufacturer. This gradually built association renders the color protectable despite the fact that consumers may have encountered or associated it with unrelated products. Taco Cabana’s décor, on the other hand, was found to be inherently distinctive and therefore did not require a similar incubatory period in which to establish its eligibility for trademark protection.[223] Such complex trade dress tends to be unique, and consumers are likely to immediately associate it with the source of the product or service it is “dressing”.[224]

D. Fragrance as Trade Dress

Trademark, and specifically trade dress, might ultimately provide the only legally protectable intellectual property to fragrances and their manufacturers. The earlier discussion of patents and copyrights concluded that neither form of legal protection held much potential for safeguarding the fragrance industry’s intellectual property.[225] The term for a patent is only twenty years – shorter than the market lifespan of a successful perfume.[226] Moreover, fragrances must be “useful” to be patented – not a designation that fragrance manufacturers would want applied to most of their products, particularly fine fragrances.[227] As French courts ultimately determined, copyright is not a viable form of protection for fragrances, despite the creative thought their creation may involve, because this intellection cannot be communicated or perceived in an effective and consistent manner.[228] Given these shortcomings of other forms of legal protection for intellectual property, trademark and trade dress emerge as the most viable options for protection. However, a number of idiosyncrasies particular to fragrance trade dress may condition its trademark protection. Recorded music is one of the most commonly used elements to create a distinctive trade dress, but recorded music is more tightly constrained by intellectual property law than is fragrance. Retailers typically do not own the copyrights of musical works that they broadcast in their stores, and therefore must pay royalties to their owners.[229] However, the legitimate use of another’s intellectual property might become part of the user’s legally protectable multisensory trade dress.[230] For instance, if one retailer were to consistently broadcast Bee Gees songs in its shops, it might be able to prevent other retailers from using these songs by claiming that consistent use and consumer association had transformed the music into a protectable component of its trade dress. Such use of another’s copyrighted material does not provide a retailer any legal interest in the music or the recordings themselves. If a retailer owns the underlying musical work being performed in connection with their goods, it might also seek trademark protection for the music itself. While sound marks typically comprise merely a few non-musical sounds or notes – like NBC’s – a larger musical work, like a jingle, can also function as a trademark.[231] The efficacy of such marks, however, may ultimately depend upon consumers’ ongoing familiarity with the words of the jingle that identify the retailer or brand. Accordingly, while NBC’s sound mark continues to be effective, that of Mr. Softee, Inc. for instance, has become genericized.[232] Today the tinkling sound of the Mr. Softee jingle, invariably heard without its long-forgotten words, conjures the sale of aerated ice cream from itinerant trucks by any number of purveyors.[233] In fact, any singsong tune, rendered in a chimey timbre, and sounding from a truck on a city street during the summer will elicit memories of ice cream and popsicles.[234] Because fragrance enjoys relatively limited intellectual property law protection, retailers are less legally constrained in their deployment of fragrance in their public sales spaces.[235] Fragrances are not copyrightable works, so the copyright performance and display rights applicable to musical and artistic works do not regulate their release into communal spaces, even commercial spaces.[236] Many small retailers and service providers like spas routinely scent their atmospheres, delivering established brands of fragrances through various delivery methods, such as aerosols, scented candles, and potpourris. For example, there is nothing objectionable to Diptyque or Shoyeido about a clothing boutique owner lighting a Diptyque candle in his shop, or a spa manager burning a stick of Shoyeido incense on the premises, because doing so promotes sales of these products among customers who inquire about what they smell. Fragrance brands would likely object, however, if a large retailer used their air conditioning system to disseminate one of their fragrances consistently, and without authorization. With scant intellectual property protection fragrance producers possess little control over use of their products, but trademark protection may offer an untapped compensatory benefit. Once Chanel sells a bottle of its well-known No. 5 the company has virtually no legal means of controlling how the buyer uses it. However, trademark may allow it to regain some authority over sales and use of this product. For instance, if Omni Hotels began to scent all of their properties with No. 5, rather than using their proprietary lemongrass fragrance, Chanel would likely assert a claim under the Lanham Act of unfair competition, and “passing off”.[237] Even if Omni used no visual evidence of the brand, Chanel would argue that by “marking” their air with the well-known No. 5 Omni was attempting to lead consumers to believe that its mid-range hotels are legitimately associated with this purveyor of top-tier luxury products. This association would potentially sully Chanel’s image.[238] Therefore, even though no other business uses No. 5 as trade dress, Chanel could prevent such use if it can demonstrate that consistently scenting the air of a commercial space would likely create confusion “as to the source, sponsorship, or association between goods or services.”[239] Fragrance’s relatively brief period of perceptibility affects its trademark capacity. While consumers may “tune out” the sounds and images of a retailer’s trade dress they cannot stop hearing and seeing them unless they block or replace the sounds and images.[240] In contrast, humans become habituated to scent relatively swiftly.[241] Once we have perceived an odor, our awareness of it rapidly wanes even though we continue to be exposed to the same concentration of it in the atmosphere.[242] Because of its brief period of perceptibility, fragrance is more likely to be protected as trade dress when used in a multisensory combination of various stimuli like colors, images, and sounds. When one first enters an interior space, a signature fragrance by itself may be the first confirmation that one is at a particular retailer, hotel, or spa. As that perception wanes, however, visual and aural stimuli will play a more prominent role in consumers’ awareness of a particular vendor. In fact, the consistent combination of a particular fragrance with other visual and aural stimuli will likely strengthen through such amalgamation, the association of that fragrance with a specific retailer. The greater the number of sensory elements comprising a trade dress, the more likely it is to be protectable as a distinctive indicator of a particular source of goods or services. Taco Cabana’s trade dress was protectable only because it combined a number of architectural and decorative elements; the murals, umbrellas, and interior design only become distinctive when combined.[243] The greater the complexity of the trade dress, however, the narrower the scope of its protection. Accordingly, if Abercrombie were to claim a trade dress that comprises visual elements as well as sounds and scents, it would be difficult for this retailer to establish unfair competition based upon a competitor’s use of a discrete element of Abercrombie’s multisensory trade dress. Fragrance’s capacity for trademark protection may directly benefit retailers and fragrance brands, but not the industrial manufacturers of fragrance blends. A fragrance manufacturer cannot protect a fragrance as trade dress for its products or services because fragrances are its products and services. As such they cannot acquire the distinctiveness, or secondary meaning necessary to obtain trademark protection. Manufacturers sell even fine fragrances to retail products manufacturers like couture houses and cosmetic companies that package and label commissioned blends under their own brands.[244] These brands, like retailers and hotel chains that scent their atmospheres, do not publicize the name of the manufacturer that produced the scent. In short, major fragrance manufacturers are akin to ghostwriters who cede copyright in their writings in exchange for fees from the commissioning parties to whom the published works are attributed.[245] Although they are not the primary beneficiaries of fragrance trade dress protection, fragrance manufacturers may indirectly capitalize on the growing accommodation under U.S. trademark law for the protection of non-traditional trade dress. Ghostwriters are aware of the economic value of the copyrights that they assign to the attributed author of their works, and this value is reflected in their fees. The fact that a commissioned fragrance may now be deployed and protected as trade dress adds economic value greater than that of fragrances distributed exclusively as retail products. Given that fragrances increasingly function as valuable and legally protectable branding agents, fragrance creators should factor this development into the fees they charge for the creation and production of products deployed in this innovative manner.


Over the past twenty-five years advances in analytic technologies, and increasingly stringent government disclosure regulations, have challenged fragrance manufacturers’ efforts to maintain exclusive control over their most valuable assets: proprietary information relating to the creation and manufacture of fragrances.[246] As discussed earlier, once this information has been disseminated there is little recourse under copyright or trade secret law to check its distribution or implementation.[247] Patents also are of limited efficacy to fragrance manufacturers, effectively protecting only newly discovered “captive” molecules.[248] Johann Böttiger’s enterprise (Meissen Porcelain) lost its most valuable, and seemingly mission-critical, trade secret within decades of its establishment. Yet, hundreds of years after what would appear to have been a devastating loss, Meissen Porcelain is flourishing not only because it diversified its merchandise, but more importantly because it capitalized upon the prestige associated with the quality of its porcelain, and also its market lead, made possible by the fact that it was the first European enterprise capable of producing it. Like Meissen Porcelain, major fragrance houses have long histories, most having existed for well over a century.[249] Unlike Meissen, however, these fragrance companies have functioned as ghostwriters, creating and manufacturing products ultimately sold as emanations of their customers, including couture houses, retailers, and consumer products companies. To an increasing extent, these fragrances are not merely sold by retailers as consumer products but are also used, like broadcast recordings of musical works, as a component of a larger trade dress by which consumers identify a particular retailer or service provider. The economic potential in fragrance trade dress can be analogized to that inherent in the public performance of copyrighted musical works. Copyright owners of musical works cannot control, or derive financial benefit from, private performances of their songs. When their copyrighted compositions are performed in public, however, they are legally entitled to receive royalties for these uses of their works.[250] Like copyright owners, fragrance manufacturers and retailers cannot control, or derive financial benefit from, private uses of their legitimately acquired products. Nonetheless, as these products are increasingly deployed in public – especially commercial – spaces, they acquire value beyond that ascribable to their hedonic attributes, by contributing to the branding of a commercial product or service. In other words, just as a copyrighted song realizes greater economic value through public performances, a fragrance acquires greater economic value when it becomes part of a legally protectable trade dress through its deployment in a commercial public space. The legal protections for fragrance remain limited. Like the Meissen Porcelain enterprise in the 1720s, individual fragrance manufacturers today can never recapture once-secret information that is now widely known, or readily and legitimately ascertained by others through reverse engineering. Accordingly, trade secret protection is an increasingly elusive quarry for this industry. Patent protection is similarly inefficacious, due to its limited duration and requirement of usefulness — a characterization unpalatable to luxury purveyors. As works of fragrance are not copyrightable, their diffusion in public spaces cannot be regulated as performances of them. When such use results in the establishment of legally protectable trade dress, however, fragrance creators could acquire a portion of the economic value of this interest, in the creation of which they have played a significant role. By capitalizing on the value stemming from the increasing use of fragrances in establishing protected brands, fragrance houses might find through trademark law partial compensation for this loss of intellectual property.
* B.M. Oberlin; J.D. American Univ.; M.A., Ph.D. Stanford; M.I.M.S., Berkeley; Lecturer, USC Law School ( Un grand merci to Claire Guillemin, collègue hambourgeoise, for sustaining my interest in this curious industry.
[1] See Richard Stamelman, Perfume: Joy, Obsession Scandal Sin; A Cultural History of Fragrance from 1750 to the Present 94 (2006).
[2] Id.
[3] See Eugénie Briot, From Industry to Luxury: French Perfume in the Nineteenth Century, 85 Bus. Hist. Rev. 273, 277 (2011).
[4] Id.
[5] See id. at 277–79.
[6] See generally Sue Minter, Fragrant Plants, in The Cultural History of Plants ch. 13 (2005). The small city of Grasse is situated a few miles north of Cannes on the Ligurian sea.
[7] See id. See also Stamelman, supra note 1, at 95 (noting that the production of jasmine in Grasse peaked in the 1920s and 1930s but subsequently declined dramatically).
[8] See Leffingwell & Associates, 1999–2002 Flavor and Fragrance Industry Leaders, (last visited Jan. 12, 2016).
[9] See id.
[10] See infra text accompanying note 57.
[11] See infra text accompanying note 38.
[12] In the eleventh century , on a mountainside north of Grenoble, Saint Bruno of Cologne established the Carthusian monastic order. See John F. Fink, 100 Important Events in Catholic History: From Pope Peter to Pope Francis 43 (2013). The region’s harsh climate, however, accommodated the cultivation of little more than medicinal herbs. See La Vallée du Secret, VSD Magazine (Jun. 18, 2014). Faute de mieux, the Carthusians grew herbs that they eventually used in the manufacture of the “elixir” that in the eighteenth century they began to distribute beyond the monastery as “Chartreuse”. See id.
[13] Their expulsion appears to have been motivated by the French government’s resentment of the order’s financial prosperity. See France Banishes Carthusian Monks, N.Y. Times, Mar. 27, 1903, at 3.
[14] See History of the Chartreuse Liqueurs,, (last visited Jan. 12, 2016).
[15] See id. Meanwhile, the Carthusians in Spain continued to produce the original liqueur that they then called “Tarragone” because the French government forbade their use of the trademark “Chartreuse” on products sold in France. Id. Accordingly, until 1929, when the trademark “Chartreuse” was restored to the Carthusians, those drinking Chartreuse in cafes and restaurants in France would order “a Tarragone”. Id.
[16] See generally, Into Great Silence (Zeitgeist Films 2005).
[17] See The Carthusian Order, Quick Presentation, (last visited Jan. 11, 2016). Carthusian monks avow a life of silence, which is spent at one monastery. Many aspects of their life are hermetic, but the Carthusians are a coenobitic order in which all monks work for the communal good, with no ambition for personal financial gain. Id. Only two monks know the formula for Chartreuse at any time. See History of the Chartreuse Liqueurs, supra note 14. Like wine, Chartreuse is a volatile drink, the quality of which typically improves over time. Accordingly, even if one successfully reverse engineered the molecular composition of Chartreuse, one would need also to discover the techniques by which to nurse the liqueur to maturity over many years, to match the quality of that of the monks’. See id. Moreover, there are over 100 ingredients in Chartreuse, many of which grow only in Alpine locations like that of the Carthusians’ motherhouse. See Christina Rebuffet-Broadus, Chartreuse, France Today (Apr. 9, 2013), Over time there have been a number of attempts to market counterfeit Chartreuse, the bottles of which are displayed at the museum of the Caves de la Chartreuse in Voiron, located near La Grande Chartreuse. WikiMedia Commons, File:Chartreuse-fake.jpg, (last visited Jan. 22, 2016).
[18] See Janet Gleeson, The Arcanum 56 (1998). After Vasco da Gama discovered a sea route between Europe and the Orient in the late-fifteenth century, Europeans began to collect Chinese porcelain. As porcelain became popular in Europe, huge quantities were obtained at great cost from China because Europeans did not possess the information needed to manufacture it. Id. at 45 (noting porcelain’s imperviousness to the water damage that threatened other luxuries like tea, silk, and spices shipped from the Orient).
[19] See text panels for exhibition: The Arnhold Collection of Meissen Porcelain, 1710 – 50 (Frick Collection, 2008) (copy on file with author). I am grateful to Frick curator Charlotte Vignon who provided me a copy of these text panels.
[20] See Gleeson, supra note 18, at 295 (noting how “the efforts of discontented employees and wandering arcanists had demolished its monopoly and spread the secret arcanum for porcelain far and wide”). Moreover, the manufacture of Böttiger’s porcelain depended upon a rare white clay called kaolin, available at the time from a sole provider in the Saxon town of Aue. When the Aue clay merchant realized that Böttiger no longer monopolized porcelain manufacture, he increased the price for clay that he charged Böttiger. He also began to sell his clay to Böttiger’s competitors, despite the fact that he was contractually bound to supply only Böttiger’s enterprise. See id. at 106.
[21] Id. In the eighteenth century the nation we now call Germany was comprised of many independent states like Saxony, Bavaria, and Württemberg. Each state had its own legal regime and there was little chance of being prosecuted for a malfeasance like trade secret misappropriation outside the jurisdiction of the owner of the trade secret. See James Sheehan, German History, 1770 – 1866 14 (1989) (noting that the “Reich came from a historical world in which nationality had no political meaning and states did not command total sovereignty.”).
[22] See Meissen Couture, Our Famous Brands, (last visited Jan. 11, 2016).
[23] Id. Martin Lindstrom refers to this association as the “Organizational Selling Proposition” in which not a physical product, but rather “the organization or corporation behind the brand in fact became the brand.” See Martin Lindstrom, Brand Sense: Build Powerful Brands Through Touch, Taste, Smell, Sight, and Sound 4 (2005).
[24] The quality of a fragrance, like a culinary dish, depends upon not only the formula or recipe, but also the particular ingredients used to instantiate it. If Chanel were to use jasmine from India rather than that from France to manufacture No. 5 the scent of this perfume would differ slightly from that of No. 5 – made from jasmine grown in the south of France – despite the fact that the same chemical formula would be used in creating it. Accordingly, even if one successfully reverse engineers the formula for a fragrance, one still may not be able to replicate exactly the original without access to the same source of ingredients used in the original. Cathy Newman offers a bird’s-eye view of fragrance manufacturing in her book Perfume: The Art and Science of Scent. (1998).
[25] See Int’l Fragrance Ass’n, The Socio-Economic Impact of Fragrance in Europe 7 (2012).
[26] The tension between employees’ rights of free expression and employment mobility, and employers’ right to control the dissemination of information they consider proprietary underlies most trade secret misappropriation claims today. See Roger Milgrim, Milgrim on Trade Secrets § 5.01 (1994) (noting that “as reliance on trade secret protection increases, controversies between former employers and ex-employees … will not only increase in number, they will be contested for significantly higher stakes”).
[27] See Meissen Couture, supra note 22.
[28] See infra Part I.B.
[29] See infra text accompanying note 189.
[30] Most fragrance manufacturers do not produce retail products but rather develop and produce proprietary blends that are sold to consumer goods producers like Unilever, Proctor & Gamble, and brands like Dior. See infra note 75 and accompanying discussion.
[31] See Tracy Lewis & Jerome Reichman, Using Liability Rules to Stimulate Local Innovation in Developing Countries: Application to Traditional Knowledge, in International Public Goods and Transfer of Technology Under a Globalized Intellectual Property Regime 342 (Keith Maskus & Jerome Reichman eds., 2005).
[32] See infra notes 44, 46 and accompanying text.
[33] See Unif. Trade Secrets Act (amended 1985), 14 U.L.A. 433 (1990). This Act has now been adopted by 47 of the United States. Unif. Law Comm., Legislative Fact Sheet – Trade Secrets Act, The same criteria define trade secrets in the pending European Union trade secret legislation. See Commission Proposal for a Directive of the European Parliament and of the Council on the Protection of Undisclosed Know-How and Business Information (Trade Secrets) Against their Unlawful Acquisition, Use and Disclosure, COM (2013) 813 final (Nov. 28, 2013) [hereinafter Commission Proposal].
[34] The United States Economic Espionage Act criminalizes the misappropriation of trade secrets on behalf of a foreign government. See 18 U.S.C. §§ 1831–1839 (2012). In 2015, for the third time, United States legislators have introduced a bill that would create allow civil trade secret claims to be brought in federal court. See Defend Trade Secrets Act of 2015, S. 1890, 114th Cong. (2015). The United States Patent & Trademark Office identifies trade secrets as “a fourth type of intellectual property, in addition to patents, trademarks, and copyrights.” See USPTO, Trade Secrecy Policy, (last visited Jan. 12, 2016).
[35] “Although not protected as a classical IPR, trade secrets are nevertheless a key complementary instrument for the required appropriation of intellectual assets that are the drivers of the knowledge economy of the 21st century. The holder of a trade secret does not have exclusive rights over the information covered by the trade secret.” Commission Proposal, supra note 33, at 3.
[36] Under U.S. law the rights of copyright owners are limited by statutory provisions allowing for unauthorized uses of copyrighted information by journalists, educators, et al. See 17 U.S.C. §§ 107, 108 (2012).
[37] Many manufacturers anticipate and avert such losses through sales contract provisions that prohibit customers from reverse engineering products acquired from the manufacturer.
[38] See Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 490–493 (1974) (distinguishing patent protection that operates “against the world” from trade secret protection that does not protect against independent creation or reverse engineering).
[39] See Chicago Lock Co. v. Fanberg, 676 F.2d 400, 405 (9th Cir. 1982) (finding that federal patent law would preempt any state-conferred monopoly through absolute protection of a trade secret).
[40] See Semiconductor Chip Protection Act of 1984, 17 U.S.C. §§ 901–914. (2012) (providing ten years protection for registered computer chip topographies); Digital Millennium Copyright Act, Pub. L. No. 105-304, 112 Stat. 2860 (codified in scattered sections of 5, 17, and 28 U.S.C.) (restricting the production and use of devices whose purpose is to circumvent digital rights management technologies). See Vessel Hull Design Protection Act, 17 U.S.C. §§ 1301–1332 (2012). The fashion industry has attempted, so far unsuccessfully, to obtain similar sui generis federal protection for apparel designs. See Innovative Design Protection and Piracy Prevention Act, H.R. 2511, 112th Cong. (2011).
[41] See J.H. Reichman, Legal Hybrids Between the Patent and Copyright Paradigms, 94 Colum. L. Rev. 2432, 2443 (1994) (discussing the economic and social underpinnings of sui generis intellectual property protection enacted in the United States and Europe “owing to the advent of new, information-based technologies, including computer science … whose industrial applications were costly to develop but vulnerable to rapid duplication.”).
[42] Article 4 of the Directive states: “The acquisition of trade secrets shall be considered lawful when obtained by any of the following means: (a) independent discovery or creation; (b) observation, study, disassembly or test of a product or object that has been made available to the public or that it is lawfully in the possession of the acquirer of the information; (c) exercise of the right of workers representatives to information and consultation in accordance with Union and national law and/or practices; (d) any other practice which, under the circumstances, is in conformity with honest commercial practices.” Commission Proposal, supra note 33. The underlying motivation for the proposed Directive was the lack of harmonization among the trade secret laws of the twenty-eight member states of the European Union. See Study on Trade Secrets and Confidential Business Information in the Internal Market, at 15-16 (Apr. 2013), In June 2015 the EU Parliament approved the key features of the draft prepared by the EU Commission and EU Council. See Report on the Proposal for a Directive of the European Parliament and of the Council on the Protection of Undisclosed Know-How and Business Information (Trade Secrets) Against Their Unlawful Acquisition, Use and Disclosure, COM (2013) 813 final (Jun. 25, 2015), – title1. The Directive is now being reviewed by the EU Council that will finalize a draft that will be become EU law and ultimately the basis of member states domestic legislation for trade secrets. See Mark P. Wine, One Step Closer: European Parliament Legal Affairs Committee Approves Trade Secret Directive, Orrick Trade Secrets Watch (Jun. 26, 2015),
[43] See Roland Knaak, et al., Comments of the Max Planck Institute for Innovation and Competition On the Proposal of the European Commission for a Directive on the Protection of Undisclosed Know-how and Business Information (Trade Secrets) Against Their Unlawful Acquisition, Use and Disclosure of 28 November 2013, COM (2013) 813 Final, Max Plank Institute for Innovation and Competition Research Paper No. 14-11, 11 (2014) (noting that the Directive does not consider trade secret protection an exclusive right but rather one obtained under unfair competition law, whereas Italian law considers trade secrets to be an intellectual property right).
[44] Id.
[45] See id. ¶ 37.
[46] Id.
[47] “Perfumers work in the strictest secrecy, jealously guarding the mysteries of their art. Since the beginning of perfume, formulas have been kept hidden from prying eyes….” Elizabeth Barillé & Catherine Laroze, The Book of Perfume 45 (1995).
[48] The still-secret formula for the liqueur Chartreuse was originally used to create a more potent potable used as medicine. See History of the Chartreuse Liqueurs, supra note 14.
[49] Napoleon III, Emperor of France between 1852 and1870, was perhaps indirectly responsible for the separation of the fragrance and pharmaceutical industries by promulgating a regulation requiring makers of pharmaceuticals to disclose on the labels of their products the ingredients they contained. To preserve the secrecy of their formulas, perfume manufacturers disassociated themselves and their products with pharmacists and pharmaceuticals. See Stamelman, supra note 1, at 95.
[50] Id.
[51] “Hubigant legend … is that Marie Antoinette, in disguise on her flight to Varennes, was wearing a Houbigant fragrance, which caused her to be identified as royalty when her coach was stopped, because none but royalty would have possessed such a magnificent perfume!” Lightyears Collection: Houbigant,, (last visited Jan. 19, 2016).
[52] See Briot, supra note 3, at 276 (reporting that by the end of the nineteenth century there were over 300 perfume producers in France and around 2000 small perfume shops in Paris).
[53] See, e.g., Anna Chesters, A Brief History of Guerlain, The Guardian (Mar. 26, 2012), (discussing the origins of Guerlain, long a family enterprise).
[54] Eighty percent of the fragrances sold today are incorporated into personal care and household care products. See Int’l Fragrance Ass’n, supra note 25, at 14-17.
[55] In 2013, Givaudan, the world’s largest producer of flavors and fragrances had 9,331 employees distributed throughout eighty-eight locations in five continents. Givaudan, Annual Report 3 (2013).
[56] The same factors have simultaneously challenged many other high technology industries that rely on trade secrets, resulting in an enormous increase in trade secret litigation in federal courts. See David Almeling, et al., A Statistical Analysis of Trade Secret Litigation in Federal Courts, 45 Gonz. L. Rev. 291, 293 (2009/10).
[57] One such example is IPRA Fragrances’ (France) claim in 2012 against employees who joined a competitor. See Mathilde Tranoy, Deux Salairiés Accusés d’avoir Vendu des Formules Aromatiques Secrètes, Nice Matin (Mar. 30, 2012, 7:37 AM), Another example is Estée Lauder’s claim against ex-employee Shashi Batra for sharing its trade secrets with a direct competitor. See Estée Lauder Cos. Inc v. Batra, 430 F. Supp. 2d 158, 160 (S.D.N.Y. 2006). A third example is, Intarome Fragrance & Flavor’s prosecution of fired employee Michael Zarkades for sharing trade secrets with E.T. Horn, a flavor manufacturer. See Intarome Fragrance & Flavor Corp. v. Zarkades, No. 07-873, 2007 U.S. Dist. LEXIS 22780, *4 (D.N.J. Mar. 29, 2007).
[58] Givaudan Fragrances Corp. v. Krivda, No. 08-04409, 2013 U.S. Dist. LEXIS 153437, *1 (D.N.J. Oct. 25, 2013).
[59] See id.
[60] Id. at *4.
[61] Id. at *6.
[62] See id.
[63] See generally Motion for New Trial by Givaudan Fragrances Corporation, Givaudan Fragrances Corp. v. Krivda, No. 08-04409 (D.N.J. Mar. 7, 2014).
[64] Apparently Givaudan did not trust the efficacy of the court’s protective order that would have purportedly prevented the disclosure of over 600 Givaudan formulas through their inclusion in the docket for this litigation. Because Givaudan would not fully disclose these formulas to the defendant and the court, the court dismissed the case based on its determination that the plaintiff failed to provide defendant adequate notice of the allegedly misappropriated trade secrets. See id. at 7–8.
[65] See generally Patrick Süskind, Perfume: The Story of a Murderer (1985).
[66] See generally Tim Robbins, Jitterbug Perfume (1984).
[67] See Arian van Asten, The Importance of GC and GC-MS in Perfume Analysis, 21 Trends in Analytical Chemistry 698, 699 (2002) (noting that there is little discussion in the open literature about the work of analytic chemists in the fragrance industry because “that is usually not in the best interest of the companies operating in this highly competitive market.”).
[68] See John Daintith, A Dictionary of Chemistry (6th ed. 2008).
[69] See id.
[70] See Arian van Asten, supra note 67, at 701.
[71] See Int’l Fragrance Ass’n, supra note 25, at 22-23.
[72] For example, Pirate Parfum, the self-proclaimed producer of “[t]he greatest perfumes, at impertinent prices,” is a significant player in this industry. It does not sell counterfeits of well-known perfumes, but rather copies of them that are branded with different names, and sold in uniform and non-descript packaging with no resemblance to that of the original products. See Pirate Parfum: Master Perfumer, (last visited Jan. 19, 2016).
[73] See Leffingwell & Associates, 2010–2014 Flavor and Fragrance Industry Leaders, (last visited Jan. 12, 2016). International Flavors & Fragrances (IFF) is headquartered in the United States, but originated in the Netherlands. See IFF, History Timeline, – /timeline (last visited Jan. 19, 2016).
[74] See Claire Guillemin, Law and Odeur: Fragrance Protection in the Fields of Perfumery and Cosmetics Part I, § 3 (forthcoming 2016) (discussing the origins of this understanding in the quasi-familial ethos of the fragrance industry through the early twentieth century).
[75] See Nicole Vulser, Le Groupe LVMH se Réapproprie la Fabrication de ses Parfums, Le Monde 16 (May 28, 2011) (noting that LVMH owns the Sephora chain, one of the largest retail outlets for perfume). See also, Fabien Pigalle, Grasse: Louis Vuitton se (Re)met au Parfum, Nice-Matin 36 (Apr. 13, 2012) (discussing LVMH’s plans to open in 2014 a workshop for fragrance development in Grasse).
[76] Vulser, supra note 75.
[77] Id.
[78] Id.
[79] Id.
[80] “We were presented with a fait accompli. Dior did not warn us that it would no longer market one of our flagship products,” said Frédéric Rivoire, CEO of Givaudan Fine Fragrances Europe. The shortfall for the company, even though it is working for other brands, amounts to several million euros of turnover.” Id.
[81] See Int’l Fragrance Ass’n, supra note 25, at 44.
[82] See LVMH, Perfumes & Cosmetics, (last visited Jan. 19, 2016).
[83] See, e.g., Commission Regulation 1367/2006 of Sept. 6, 2006, The Application of the Provisions of the Aarhus Convention on Access to Information, Public Participation in Decision-making and Access to Justice in Environmental Matters to Community Institutions and Bodies, 2006 O.J. (L 264) 13, 14 (granting a public right of access to information held by EU agencies relating to “emissions into the environment”).
[84] See 35 U.S.C. § 154 (2012); Convention on the Grant of European Patents, Oct. 5, 1973, 1065 U.N.T.S. 199.
[85] See 35 U.S.C. § 112 (2012).
[86] Within the fragrance industry “fine fragrances” refers to stand-alone perfumes that are worn on the body for aesthetic purposes.
[87] See 35 U.S.C. § 101 (2012).
[88] The appeal and economic value of a perfume are actually enhanced by its lack of utility. Lysol underscores the utility of its stolidly named aerosol Sanitizing Spray as an economical and effective product to control “bathroom, pet, garbage, and diaper odors,” whereas Patou fosters an attitude of hedonic and heedless extravagance in advertising Joy as “the costliest perfume in the world.” See Lysol, Lysol Neutra Air Sanitizing Spray, (last visited Jan. 22, 2016); Two Sides of Dear: Demystifying Patou’s Joy Perfume Promotion, Perfume Shrine (May 13, 2014), Economist Thorstein Veblen (1857-1929) pointed out the value associated with conspicuous wastefulness and lack of utility of certain products. “The superior gratification derived from the use and contemplation of costly and supposedly beautiful products is, commonly in great measure a gratification of our sense of costliness masquerading under the name of beauty.” Thorstein Veblin, The Theory of the Leisure Class 128 (Prometheus Books 1998) (1899).
[89] The prestige and prices of Chanel’s No. 5 and Patou’s Joy for instance, are bolstered by the fact that both products have endured since 1929 and 1921 respectively, in a market in which hundreds of new fragrances are launched (and typically fail) annually. On the other hand, most new seasonal or “celebrity” fragrances have such a limited shelf life that they need no IP protection whatever.
[90] See USPTO, USPC Index: Class 512 Perfume Compositions (2011),
[91] Id. (providing an interactive index of applications).
[92] E.g., U.S. Patent No. 7,169,746 col 1 l. 6–7 (filed Mar. 26, 2002) (Shiseido’s patent for a “perfume for effecting mental control through psycho-sedation or psycho-stimulation.”). United States regulation of fragrances has been relatively sparing. The United States Fair Packaging and Labeling Act requires disclosure of ingredients used in consumer products unless such disclosure would reveal trade secrets. 15 U.S.C. § 1453 (2006). Because fragrance manufacturers assert that their compositions constitute trade secrets, their incorporation into a product is typically indicated not with a list of constituent chemicals but rather simply the term “fragrance”. The United States Food and Drug Administration (USFDA), however, classifies fragrances that are marketed for their aroma-therapeutic capacities as drugs that are subject to the Administration’s more stringent regulation. See USFDA, Aromatherapy (2015),
[93] See Erin McAvoy, Chemical Romance: How did Chemists Become the Greatest Force in Fragrance? The Independent (Dec. 10, 2010), (noting that “synthetic raw materials, usually single molecules, enable perfumers to create entirely new smells….”).
[94] See id. See also Emma Davies, The Sweet Scent of Success, 2009 Chemistry World 40 (Feb. 2009) (discussing the deployment of several of the most significant proprietary perfume molecules in fragrances like Dior’s Poison and Donna Karan’s Be Delicious).
[95] See Leffingwell & Associates, supra note 73.
[96] See generally Wendy Wolfson, In the Fragrance Business, the Right Molecule Smells like Money, 12 Chemistry & Biology 857 (2005) (discussing Flexitrol’s attempt to become a clearinghouse for scent molecule licensing).
[97] A patent is costly and difficult to obtain because the claimant must establish that his invention effectuates new information for a useful purpose. See 35 U.S.C. §§ 101–103 (2012). Patent prosecution is costly in large part because it involves the work of highly trained examiners who investigate claims of novelty against the state of the art in a given class of goods. See MPEP § 2131 (9th ed. 2014).
[98] One must register an invention to obtain a patent for it; an author automatically obtains a copyright, however, simply by recording his original expression as text, sound, images or other copyrightable content. 17 U.S.C. § 102 (2012).
[99] In the timeworn words of Judge Learned Hand: “if by some magic a man who had never known it were to compose anew Keat’s Ode on a Grecian Urn, he would be an ‘author,’ and, if he copyrighted it, others might not copy the poem, though they might of course copy Keats’s.” Sheldon v. Metro-Goldwyn-Mayer Pictures Corp., 81 F.2d 49, 54 (2d Cir. 1936).
[100] See 17 U.S.C. § 106 (2012); WIPO, Summary of the Berne Convention for the Protection of Literary and Artistic Works (1866), (last visited Jan. 19, 2016).
[101] See 17 U.S.C. §§ 107-110 (2012). Germany’s copyright law, for instance, has no general fair use provision, and specifies all limitations on authors’ rights under the statute. See Wencke Bäsler, Technological Protection Measures in the United States, the European Union and Germany: How Much Fair Use do We Need in the “Digital World”?, 8 Va. J.L. & Tech. 13, ¶ 4 (2003).
[102] 8 Anne, c. 19 (1710).
[103] Code de la Propriété Intellectuelle [Intellectual Property Code] Art. L. 112-1 (the last of the fourteen categories covers articles of haute couture).
[104] The House Report on the Copyright Act of 1976 states: “The bill does not intend either to freeze the scope of copyrightable subject matter at the present stage of communications technology or to allow unlimited expansion into areas completely outside the present congressional intent. Section 102 implies neither that the subject matter is unlimited nor that new forms of expression within that general area of subject matter would necessarily be unprotected.” H.R. Rep. No. 94-1476, at 51 (1976), as reprinted in 1976 U.S.C.C.A.N. 5659, 5664.
[105] Theoretically, another fragrance manufacturer could independently and legitimately re-create the copyrighted fragrance as long as this effort were done without access to the original fragrance. See discussion supra note 99.
[106] See 17 U.S.C. § 302 (2012).
[107] The most famous example of such a transcription from memory is Mozart’s of the score of Gregorio Allegri’s “Miserere” after hearing two performances of it at the Vatican in 1770. The Vatican owned the only score and parts to Allegri’s work that was performed only twice a year, during Holy Week, in the Sistine Chapel. Performers with access to the score and parts were threatened with excommunication if they were found to have copied or distributed the work outside the Vatican. “The Papacy, realising that it owned a composition of exceptional appeal, shrewdly heightened its reputation by refusing to allow any copy to leave the Sistine Chapel. This ban was supported by threats of severe punishment.” Peter Phillips, Brochure Notes to The Tallis Scholars Recording of Allegri’s Miserere (Gimell Records 1985).
[108] Makers of Sibelius music notation software, for instance, claim that their program can convert the sound of “up to 16 instruments or notes at a time into multiple staves, with up to four voices per staff.” AudioScore Ultimate 8, Sibelius, (last visited Jan. 13, 2016).
[109] See generally U.S. Copyright Office, Circular No. 50, Copyright Registration for Musical Compositions 1 (2012).
[110] See infra note 120 and accompanying text.
[111] This difficulty is nicely illustrated in one of the opening scenes of the film adaptation of Patrick Süskind’s novel Perfume: The Story of a Murderer, in which the hapless perfumer played by Dustin Hoffman struggles vainly to analyze a popular fragrance of one of his competitors. Perfume: The Story of a Murderer (Dreamworks Pictures 2006).
[112] The 1909 Copyright Act that was effective until 1978, provided an initial term of twenty-eight years, which could be renewed. Act of Mar. 4, 1909, Pub. L. No. 60-349, 35 Stat. 1035, as amended (formerly codified at 17 U.S.C. § 1 et seq.). When copyright (and patent) terms expire, the once-protected work enters the public domain and can be used by anyone.
[113] See Sid & Marty Krofft Television Productions, Inc. v. McDonald’s Corp., 562 F.2d 1157 (9th Cir. 1977) (finding that characters used in a McDonald’s television commercial copied not merely the plaintiffs’ idea of fanciful characters in action, but substantially also their specific means of conveying the idea); Apple Computer, Inc. v. Franklin Computer Corp., 714 F.2d 1240 (3d Cir. 1983) (finding that if “other programs can be written or created which perform the same function as an Apple’s operating system program, then that program is an expression of the idea and hence copyrightable”).
[114] See About Us, Editions de Parfums-Frédérick Malle, (last visited Jan. 13, 2016).
[115] E.g., the American company International Flavors & Fragrances designated the French-educated Carlos Benaim as its first “master perfumer”. IFF Names Carlos Benaim Master Perfumer, Perfumer & Flavorist (Jan. 18, 2013),
[116] See, e.g., Perfumery School, Givaudan, (last visited Jan. 13, 2016); Perfumery School in Grasse, Grasse Inst. of Perfumery, (last visited Jan. 13, 2016).
[117] See generally Perfumery School, supra note 116; Perfumery School in Grasse, supra note 116.
[118] See Guillemin, supra note 74, at Part IV (providing exhaustive coverage of French and Dutch copyright litigation involving fragrances).
[119] Olivia Su, Odor in the Courts! Extending Copyright Protection to Perfumes May Not Be so Nonscentsical: An Investigation of the Legal Bulwarks Available for Fine Fragrances Amid Advancing Reverse Engineering Technology, 23 S. Cal. Interdisc. L.J. 663 (2014).
[120] See Rochas v. de Laire, Cour d’appel [CA] [regional court of appeal] Paris, 4e ch., July 3, 1975, Gaz. Pal. 1976.
[121] See id.
[122] “Oeuvres de l’esprit.” Id.
[123] Mugler v. Molinard, Tribunal de commerce [T. Com.] [court of commerce] Paris, 15e ch., Sept. 24, 1999, Gaz. Pal. 2001, 17-18.01, at 5. Quest International was subsequently acquired by Givaudan in 2006. See Sam Cage, Givaudan Buys Quest from ICI, Reuters Business News (Nov. 22, 2006),
[124] See Mugler v. Molinard, Gaz. Pal. 2001, 17-18.01, at 5. Historically, in civil law regimes, like that of France, courts pay less obeisance to case law precedence than their common law counterparts; see generally Vincy Fon & Francesco Parisi, Judicial Precedents in Civil Law Systems: A Dynamic Analysis, 26 Int’l Rev. L. & Econ. 519 (2006). Accordingly, the Mugler court evidently felt no compunction about rendering a decision incompatible with that of the earlier Rochas decision by a higher court.
[125] See Mugler v. Molinard, Gaz. Pal. 2001, 17-18.01.
[126] See Bellure v. L’Oréal, Cour d’appel [CA] [regional court of appeal] Paris, 4e ch. A, Jan. 25, 2006, D. 2006, at 580, J. Daleau, aff’g Bellure v. L’Oréal, Tribunal de Grande Instance [TGI] [ordinary court of original jurisdiction] Paris, May 26, 2004, D. 2004, at 2641, note Galloux.
[127] See id.
[128] Id.
[129] HR 16 juni 2006, NJ 2006, 585 m.nt. JHS (Lancôme/Kecofa) (Neth.).
[130] See id.
[131] Id.
[132] See Beauté Prestige Int’l v. Senteur Mazal, Cour d’appel [CA] [regional court of appeal] Paris, 4e ch., Feb. 14, 2007, D. 2007, at 735, J. Daleau.
[133] Id.
[134] Id.
[135] Id.
[136] See Bellure v. L’Oréal, Cour d’appel [CA] [regional court of appeal] Paris, 4e ch. A, Jan. 25, 2006, D. 2006, at 580, J. Daleau.
[137] See Charles Cronin, Genius in a Bottle: Perfume, Copyright, and Human Perception, 56 J. Copyright Soc’y U.S.A. 427 (2009) (discussing relative acuity of human senses and its relevance to intellectual property protection).
[138] The Cour de cassation is France’s highest appellate court that is separated into six subject-matter divisions, e.g., labor, criminal, civil, etc.
[139] Société Lancôme v. Patrice Farque, Cass. com., Dec. 10, 2013 [pourvoi n° 11-19872] available at
[140] See id.
[141] Id.
[142] Jean-Michel Bruguière has argued that the court’s rationalizing its decision on the imperceptibility of the intellectual investment in the creation of a fragrance is flawed: “[t]he olfactory notes of Chanel No. 5 or Eau Sauvage are as reliably and accurately identifiable as musical notes – the harmony of the Beach Boy’s “Good Vibrations” or the melody of the Rolling Stones’ “Satisfaction”. Jean-Michel Bruguière, Chroniques: Droit d’Auteur et Droits Voisins [News Column: Copyright and Neighboring Rights], 50 Propriétés Intellectuelles 51, 52 (2014). He also suggests that some contemporary works of [classical] music are no more intelligible to the public than are fragrances. See id.
[143] The U.S. Copyright Office accepts registrations for works of: literature, visual arts, performing arts, sound recordings, and single serials. See eCO Frequently Asked Questions, U.S. Copyright Office, (last visited Jan. 14, 2016).
[144] See generally Trademark “Sound Mark” Examples, U.S, Patent & Trademark Office, (last visited Jan. 14, 2016).
[145] See, e.g., Qualitex Co. v. Jacobson Products Co., 514 U.S. 159 (1995).
[146] See discussion infra Part III.C. “The ‘trade dress’ of a product is essentially its total image and overall appearance. It ‘involves the total image of a product and may include features such as size, shape, color or color combinations, texture, graphics, or even particular sales techniques.’” Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 765 n.1 (1992) (citing Blue Bell Bio Medical v. Cin Bad, Inc., 864 F.2d 1253 (5th Cir. 1989), and John H. Harland Co. v. Clarke Checks, Inc., 711 F.2d 966 (11th Cir. 1983)).
[147] See Daniel Zendel & Dennis Prahl, Making Sense of Trademarks: An International Survey of Non-Visual Marks, Trademark World, Issue 89 (1996),
[148] See id. (noting that while the German Marks Act of 1995 accommodates color marks, the German Patent Office internal guidelines require these marks to be combined with other distinctive features to be registrable).
[149] See id. (noting the “visual perceptibility” requirement of both the Mexican Industrial Property Law and Brazilian Industrial Property Code).
[150] In 1949, for instance, in Campbell Soup Co. v. Armour & Co., the Third Circuit rejected plaintiff’s claim to the exclusive right to use the color combination of red and white on food products: “If they may thus monopolize red in all of its shades the next manufacturer may monopolize orange in all its shades and the next yellow in the same way. Obviously, the list of colors will soon run out.” Campbell Soup Co. v. Armour & Co., 175 F.2d 795, 798 (3d Cir. 1949). Over forty years later, in NutraSweet Co. v. Stadt Corp., the Seventh Circuit rejected the plaintiff’s claim to the exclusive right to use pastel blue on its sugar substitutes packaging: “if each of the competitors presently in the tabletop sweetener market were permitted to appropriate a particular color for its product, new entrants would be deterred from entering the market.” NutraSweet Co. v. Stadt Corp., 917 F.2d 1024, 1028 (7th Cir. 1990). Ultimately, in Qualitex, the Supreme Court “concluded that the color depletion issue would rarely arise and could be resolved, if necessary, by applying the functionality doctrine to prevent anticompetitive results.” 1-2 Gilson on Trademarks § 2.11.
[151] See William Landes & Richard Posner, Trademark Law: An Economic Perspective, 30 J. L. & Econ. 265, 269 (1987) (discussing this benefit as reducing “consumer search costs”).
[152] See NutraSweet, 917 F.2d at 1028.
[153] See Cronin, Genius in a Bottle, supra note 137.
[154] “Die Luft der Freiheit weht” and “Veritas” respectively. The greater the number of different colors in a mark the greater the likelihood of its distinctiveness. Nevertheless, the particular perception of color marks depends to a greater extent than that of word marks on geographical location. San Franciscans will associate a combination of the colors blue and gold with the University of California at Berkeley while residents in St. Paul will think of their city’s Bethel University. Residents of both cities will associate the combination of red, white, and blue with the United States; those of Paris and Lyon, on the other hand, will think of France (though they will reorder the colors to that of their tricolore: blue, white, and red).
[155] See 1-2 Gilson on Trademarks § 2.11 (discussing case law establishing that school colors, when used with “other indicia” of the school, can acquire secondary meaning to qualify for trademark protection).
[156] See Douglas Churovich, Intellectual Property: Policy Considerations from a Practitioner’s Perspective: Scents, Sense or Cents? Something Stinks in the Lanham Act: Scientific Obstacles to Scent Marks, 20 St. Louis U. Pub. L. Rev. 293, 293–94 (2001) (claiming that “the landmark In re Clarke decision was ill-advised since it was founded upon a poor, if not non-existent, understanding of osphresiology and the misguided application of legal principles that fail to apply to scents”); Bettina Elias, Do Scents Signify Source? An Argument Against Trademark Protection for Fragrances, 82 Trademark Rep. 475 (1992) (claiming that “fragrances only rarely, if ever, function as trademarks in the marketplace and, in those few cases in which fragrances do arguably indicate a product’s source, their trademark protection remains doctrinally problematic and potentially impossible to implement and enforce in a consistent fashion.”).
[157] Visually or verbally complex marks, however, may be weaker than simple marks because they demand more intellectual effort on the part of consumers to decipher and recall their association with a particular product or service.
[158] Most notorious is the wine industry’s attempts to suppress the fact of weak human perception of tastes and smells, with the humbuggery it uses to market its products. See David Derbyshire, Wine-Tasting: It’s Junk Science, Guardian (June 22, 2013, 7:01 PM),
[159] On this question “[t]he door is firmly closed in … countries such as Brazil Japan, China and Taiwan, where scent marks are neither registrable and the courts do not appear to have considered the protection of unregistered scent marks under other legal theories.” See Zendel & Prahl, supra note 147.
[160] See First Council Directive to Approximate the Laws of the Member States Relating to Trade Marks, art. 2, O.J. L 40/1, at 2 (1989) (broadly defining trademarks as comprising “any sign capable of being represented graphically … provided that such signs are capable of distinguishing the goods or services of one undertaking from those of another undertaking”); Agreement on Trade Related Aspects of Intellectual Property Rights, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1C, Legal Instruments-Results of the Uruguay Round, 1869 U.N.T.S. 299, 33 I.L.M. 1197 (1994) (offering a similarly broad definition of trademarks, and the right of convention members to “require, as a condition of registration that signs be visually perceptible”).
[161] See Zendel & Prahl, supra note 147.
[162] Case C-273/00, Ralf Siekmann v. German Pat. & Tmk. Office , 2002 E.C. R. I- 11754.
[163] Id. at ¶14-15.
[164] Id. at ¶ 10-11.
[165] Id. at ¶11-13.
[166] Id. at ¶39, 45-48.
[167] Id. at 35.
[168] Id. at 54.
[169] Id. at 63.
[170] See TMEP § 807.09 (8th ed. Oct. 2014).
[171] See Trademark “Sound Mark” Examples, supra note 144.
[172] See In re Celia Clarke, 17 U.S.P.Q.2d 1238 (T.T.A.B. 1990) (finding that “the scent of a product may be registrable if it is used in a non-functional manner”).
[173] Sixteen records were obtained July 21, 2015 using the USPTO’s Trademark Electronic Search System searching the terms “for situations for which no drawing is possible, such as sound,” in the Mark Drawing Code field, and the term “fragrance” in the Description of Mark field. Trademark Electronic Search System (TESS), U.S. Patent & Trademark Office, (last visited Jan. 15, 2016).
[174] CITRUS FRAGRANCE, Registration No. 3,726,789 (issued 2009 for biofuel); the mark consists of a coconut scent or fragrance, Registration No. 4,113,191 (issued 2012 for retail sales); the mark consists of a high impact fragrance primarily consisting of musk, vanilla, rose, and lavender, Registration No. 4,057,947 (issued 2011 for cosmetics).
[175] The specific paper products they sought to protect were hanging file folders. See, e.g., APPLE CIDER SCENT, Registration No. 3,140,701 (cancelled in 2013).
[176] See generally, Martin Lindstrom, Buyology: truth and Lies About Why We Buy (2008) (discussing tactics by which retailers and consumer products companies sell products through the use of various sensory stimuli).
[177] See generally id.
[178] These prices were found on websites of these retailers on July 28, 2015. Target does not sell Joy in its physical stores, so its presentation costs for this product are minimal. Target, (last visited July 28, 2015); Nordstrom, (last visited July 28, 2015). Target’s decision not to sell Joy in stores may have been motivated by fears that the small but relatively expensive item would be attractive to shoplifters. It may also have contracted with Patou not to do so based on Patou’s concern about the pollution of its brand through association with Target. The existing arrangement benefits both companies because underlying it is a conspiratorial understanding between them and consumers who would be embarrassed by purchasing a high-end product at a brick-and-mortar Target bargain-counter. Eau de toilette is the most diluted version of a fine fragrance, and the only version of Joy sold by Target. Nordstrom also sells Eau de parfum that is less diluted, and more expensive, than the Eau de toilette. Nordstrom does not sell the parfum, a half ounce of which is sold only at top-tier shops like Neiman Marcus for about $350. Target, (last visited July 28, 2015); Nordstrom, (last visited July 28, 2015); Nieman Marcus, (last visited July 28, 2015).
[179] See David Suisman, Selling Sounds: The Commercial Revolution in American Music 66 (2012).
[180] These venues include not only restaurants featuring the dreaded “strolling musicians” but also Nordstrom department stores where former U.S. Secretary of State Condoleezza Rice once anticipated making a living as a musician. See Elisabeth Bumiller, Condoleezza Rice: An American Life 56 (2007).
[181] See Suisman, supra note 179.
[182] “[M]usic, any music at all, is so welcome to the weak of mind and so readily supplied by their commercial manipulators that almost all the music you hear, at least all you hear inadvertently, is BAD.” Paul Fussell, Bad: Or, The Dumbing of America 126 (1991).
[183] See generally George Prochnik, In Pursuit of Silence: Listening for Meaning in a World of Noise (2010) (discussing soundscapes developed on behalf of Abercrombie & Fitch).
[184] See Lindstrom, Brand Sense, supra note 23, at 72 (observing that “while hearing involves receiving auditory information through the ears, listening relies on the capacity to filter, selectively focus, remember, and respond to sound”).
[185] Broadcasting music recordings has also been used effectively to alienate undesirables from commercial spaces. See Twilight of the Yobs: How Classical Music Helps Keep Order, Economist (Jan. 6, 2005),
[186] See Lindstrom, Brand Sense, supra note 23, at 74 (noting that in Disney World “carefully choreographed sound is piped through the entire park. Even the bird sounds are controlled. It’s a whole environment designed to capture the hearts of children and waken the child within each adult.”).
[187] A Gap store in San Francisco enables customers to use smart phones to select the muzak they hear while shopping. See Gap Pilots In-Store DJ System, Lets Customers Pick and Play Music, VentureBeat (Nov. 21, 2011, 9:13 AM), It seems unlikely, however, that retailers would ever provide customers the wondrous capacity simply to turn off a soundtrack.
[188] See generally Prochnik, supra note 183.
[189] Marketing researchers have established that classical music played in a commercial setting increases the “quality sensation”. See Annabel Elliott, The Buy-ology of a Shopping Spree: How Stores like Apple and Victoria’s Secret Use Scent, Sound and Color to Make You Spend More Money, Daily Mail (Dec. 5, 2014), Accordingly, one is more likely to hear classical music in a high-end restaurant or art gallery catering to discerning customers than in a fast-food joint or sneaker shop targeting teenagers. See id.
[190] See id.
[191] In the 1930s, a Connecticut home insurance firm impregnated their advertising brochures with the scent of charred wood. See Marston Bogert, Your Nose Knows, 39 Sci. Monthly 345 (1934). Such uses of scents capitalize on their potential to conjure powerful memories instantaneously, a phenomenon known as the “Proustian effect”. See Sarah Dowdey, Does What You Smell Determine What You Buy?, How Stuff Works, (last visited Jan. 14, 2016). “When you first perceive a scent, you connect it to an event, person or thing. When you smell the scent again, it often triggers memory in the form of a conditioned response . . . smell can also activate the subconscious and influence your mood. Instead of reminding you of specific details from [a] vacation, [an] ocean scent might make you feel content or happy.” Id.
[192] See Roxie Hammill & Mike Hendricks, Scent Received, With a Tap of a Smartphone, N.Y. Times (July 8, 2015), (discussing “Scentee”, “oPhone Duo” and other mechanisms developed to generate specific scents in response to digitally communicated instructions).
[193] “The real action, however, lies in projecting olfactory character into indoor commercial spaces. This application has been fully embraced in one large business sector: the gaming industry. Las Vegas is the trend’s epicenter; half the major properties on the Strip have scent systems. The MGM Grand has deployed as many as nine scents simultaneously around its property and the Venetian features a corporate logoscent called ‘Seduction’.” Lindstrom, Brand Sense, supra note 193, at 171.
[194] Alexia Elejalde-Ruiz, For Branding, Many Places Adopt Signature Scents, L.A. Times (April 14, 2014, 7:05 PM), at – page=1. Even Goodwill Industries now infuse the air of their retail shops with a bespoke fragrance. See id. Goodwill’s motivations for doing so, however, differ from those of a retailer like Bloomingdales. Bloomingdales scents its air to promote an atmosphere of luxury and exclusivity; Goodwill uses fragrance to counter an atmosphere of frugality, if not penury, associated with the sale of used apparel of questionable cleanliness.
[195] See L. Aruna Dhir, Scent of a Hotel, 4 Hoteliers (Dec. 3, 2013),
[196] See Caroline Cerny, A New Scent at Park Hyatt Zurich, Hyatt (May 1, 2008), (discussing how parfumeur Blaise Mautin creates different scents for Hyatt depending upon the location of the hotel; Zurich’s commercial vibe is captured in an astringent scent).
[197] “Global Hyatt offers… more than 750 hotels in more than 45 countries.” Id.
[198] In one branding study, a teenager expressed confidence that the Abercrombie jeans she had been handed were authentic, and not a knockoff pair bought from a sidewalk vendor, because they were imbued with Abercrombie’s signature scent. See Lindstrom, Brand Sense, supra note 23, at 2.
[199] Fierce Cologne, Abercrombie & Fitch, (last visited Jan. 14, 2016). “Fierce” was created by Christophe Laudamiel, a gay parfumeur from France. See Serguey Borisov, Intervew with Christophe Laudamiel, Fragrantica (Apr. 13, 2014, 7:03 AM),
[200] Abercrombie & Fitch is a “retailer of men’s and women’s casual clothing, such as t-shirts, outerwear, sweatshirts, woven shirts, sweaters, jeans, khakis, shorts, baseball caps, belts, socks, and other accessories . . . designed primarily to appeal to young men and women of college age.” Abercrombie & Fitch Stores, Inc. v. Am. Eagle Outfitters, Inc., 280 F.3d 619, 624 (6th Cir. 2002). See generally Prochnik, supra note 183, at 89 – 106 (2010) (documenting the marketing tactics of Abercrombie and similar retailers to attract young customers by creating alluringly risqué environments)
[201] Tmep, § 1202.02 (8th ed. Oct. 2014).
[202] See Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763 (1992) (finding that the overall décor of the respondent’s restaurant was inherently distinctive, and therefore protectable trade dress, even though respondent had not demonstrated that this décor had acquired secondary meaning).
[203] Marks that are not inherently distinctive, but otherwise meet registration requirements, may be registered on the Secondary Register. See Glynn S. Lunney, The Trade Dress Emperor’s New Clothes: Why Trade Dress Does Not Belong on the Principal Register, 51 Hastings L.J. 1131 (2000) (discussing significance of registration on the Principal rather than Supplemental Register).
[204] See 1-2 Gilson on Trademarks § 2A.01 (noting that “although there is no empirical evidence, it appears that most trade dress is not registered and may instead be judicially protected under Section 43(a) of the Lanham Act”).
[205] This was true, for example, of the trade dress at issue in Qualitex Co. v. Jacobson Prods. Co., 514 U.S. 159 (1995). Qualitex had been using the contested trade dress for about thirty years and only attempted to register it as a trademark when a competitor began using a similar trade dress.
[206] General Mills, History of Innovation: The History of Betty Crocker,; History of the Morton Salt Girl, Morton Salt, (last visited Jan. 14, 2016).
[207] See 1-2 Gilson on Trademarks § 2.11 (citing Martin Lindstrom’s Brand Sense: Build Powerful Brands Through Touch, Taste, Smell, Sight, and Sound, in which the author argues that given the overload of information in today’s marketplace, retailers must develop multisensory brands to reach consumers). Lindstrom, Brand Sense, supra note 23.
[208] See generally Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763 (1992); Qualitex Co. v. Jacobson Prods. Co., 514 U.S. 159 (1995).
[209] See Two Pesos, 505 U.S. at 767.
[210] See id. at 766.
[211] See id. at 764.
[212] See id. at 766. The plaintiff’s exhibits of images of the competing restaurants suggest that the district court jury was likely swayed in their verdict of infringement by the fact that the defendant not only copied the plaintiff’s colors, and architectural features, but also the size and arrangement of these features, down to the flower pots along an exterior wall. Photos of both restaurants are provided in the slides of Michael Atkins’ presentation Trade Dress Protection in the United States, given at the University of Washington School of Law. Michael Atkins, Trade Dress Protection in the United States, SlideShare, (last visited Jan. 14, 2016).
[213] See Two Pesos, 505 U.S. at 767.
[214] See id. at 766.
[215] Qualitex Co. v. Jacobson Prods. Co., 514 U.S. 159 (1995).
[216] See id. The pads, sold to dry cleaning establishments, look like ironing board covers. See SunGlow Press Pads, Qualitex, (last visited Jan. 15, 2016).
[217] See Qualitex Co. v. Jacobson Prods. Co., 13 F.3d 1297 (9th Cir. 1994).
[218] See Qualitex, 514 U.S. at 161.
[219] See id. at 174.
[220] See id. at 164.
[221] See Qualitex Co. v. Jacobson Prods. Co., 13 F.3d 1297, 1299 (9th Cir. 1994).
[222] See id.
[223] See Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 775 (1992).
[224] See, e.g., AmBrit, Inc. v. Kraft, Inc., 812 F.2d 1531, 1536 (11th Cir. 1986) (basing its finding that plaintiffs owned protectable trade dress in their packaging of Klondike ice cream bars on the fact that the images were not “a basic shape or common design [but] . . . [r]ather . . . a complex composite of size, color, texture and graphics . . . [creating] a distinctive visual impression”).
[225] See discussion supra Part I.
[226] See discussion supra Part I.
[227] See discussion supra Part I.
[228] See supra note 139 and accompanying text.
[229] See Copyright Act of 1976, 17 U.S.C. §106 (2012). “Mom & Pop” restaurants and retail establishments are exempt from this requirement. See id. at § 110. These royalties are managed by performing rights organizations, the largest of which is ASCAP. See Am. Soc’y of Composers, Authors & Publishers, (last visited Jan. 14, 2016).
[230] See Kimberlianne Podlas, I Do Not Endorse This Message! Does a Political Campaign’s Unauthorized Use of a Song Infringe on the Rights of the Musical Performer?, 24 Fordham Intell. Prop. Media & Ent. L.J. 1 (2013) (arguing that performers of copyrighted musical works should not be able to prevent uses of their performances that are legitimate under copyright law, based on trademark claims).
[231] See id.
[232] Mr. Softee, Inc. could still assert copyright to control performances of their registered song, written by jingle composer Les Waas in 1958. See Daniel Neely, Ding, Ding!: The Commodity Aesthetic of Ice Cream Truck Music, in Mobile Music Studies, Vol. II 155 (2014). In fact, Les Waas’s melody is highly derivative of the English Folk tune, “How many miles to Babylon?”. See Eloise Hubbard Linscot, Folk Songs of Old New England 18 (1939) (providing words and music notation of the song).
[233] The same fate did not befall Coca Cola Company’s jingle “I’d Like to Buy the World a Coke” despite the fact that Coca Cola later authorized the use of its melody in the pop song “I’d Like to Teach the World to Sing,” recorded by The Hillside Singers. See The “Hilltop” Ad: The Story of a Commercial, Libr. of Congress, (last visited Jan. 15, 2016). This is because whenever Coca-Cola advertised its products using the jingle, the words identifying Coca-Cola were always heard along with the melody.
[234] See Neely, supra note 232, at 146 (noting that the sound of ice cream truck music is not Pavlovian – i.e. stimulating a reflexive response to want ice cream – but rather plays on an “anamnesis” conditioned on the ability to recognize a specific product through sound).
[235] Health regulations rather than intellectual property rights are more likely to affect a retailer’s use of fragrances in commercial and public spaces. See Stuart Elliott, Joint Promotion Adds Stickers to Sweet Smell of Marketing, N.Y. Times (Apr. 2, 2007), (discussing the San Francisco’s Municipal Transportation Authority’s order to the California Milk Processor Board to remove chocolate-scented advertisements posted near public bus stops); Rachel Herz, The Scent of Desire: Discovering Our Enigmatic Sense of Smell 14 (2007) (noting that Halifax, Nova Scotia has enacted regulations making illegal the wearing of fragrances in public).
[236] See Copyright Act of 1976, 17 U.S.C. §106 (2012). The possibility of copyright protection for fragrances raises the question how moral rights of attribution, reputation, etc. would be applicable to them under copyright regimes like that of France, that provide such protection. See Code de la Propriété Intellectuelle [Intellectual Property Code] L. 121-1- L. 121-4 (1992).
[237] Lanham Act § 43(a), 15 U.S.C. § 1125 (2012) (prohibiting conduct “likely to cause confusion, or to cause mistake, or to deceive as to the affiliation, connection, or association of such person with another person, or as to the origin, sponsorship, or approval of his or her goods, services, or commercial activities by another person. . . .”).
[238] See John Tagliabue, Why European Computer Makers Flop, N.Y. Times (Oct. 7, 1996),
[239] Lanham Act § 43(a), 15 U.S.C. § 1125 (2012).
[240] Such efforts would involve wearing unwieldy headphones, blinders, or similar paraphernalia, the use of which may elicit ambivalence by spectators as to the mental stability of the wearer.
[241] See Jennifer Chen, Human Olfactory Perception: Mechanism, Characteristics, and Functions (May, 2013) (unpublished Ph.D. dissertation, Rice University) (on file with author); see also Avery Gilbert, What the Nose Knows: The Science of Scent in Everyday Life 85 (2008) (positing that “[t]he longer you are exposed to an odor, the more you adapt to it. Step into a garlic factory and the reek will overwhelm you. A few minutes later its intensity fades, and after an hour you might not be able to smell garlic at all, no matter how hard you try. Work there a few months and this adjustment will happen almost as soon as you step in the door”).
[242] See Gilbert, supra note 241, at 85. It is this habituation that prompts retailers of perfumes to keep at hand a saucer of coffee beans, the odor of which contrasts sharply with that of fine fragrances, thereby enabling customers to “reset” their olfaction, allowing them to perceive fragrances anew. Nevertheless, Gilbert notes that: “the bean meme is now a fixture in perfume retailing… The Jo Malone display in Saks had them [coffee beans] in an apothecary jar with a metal lid. It’s all good fun and marketing, but there is not a jot of science behind it. (There are twenty-seven aroma impact molecules in roasted Arabica coffee – how could smelling all these help clear the nose?)” Id. at 108.
[243] See Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763 (1992).
[244] Dior’s fragrances like Miss Dior and Diorissimo, for instance, were developed, and manufactured by Givaudan. See supra note 77 and accompanying text
[245] See Roberta Kwall, The Soul of Creativity: Forging a Moral Rights Law for the United States 91 (2009) (suggesting an inherent “degradation” of both attributed author and ghostwriter when the ghostwriter creates most of a literary work but without attribution).
[246] See, e.g., United Nations Economic Commission for Europe, Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters, June 25, 1998, 2161 U.H.T.S. 450 (granting the public right of access to any information held by European Union agencies relating to emissions into the environment); International Fragrance Association, North America, Federal Regulatory Chart, (identifying seven federal agencies involved in regulating fragrance ingredients and manufacturing in the United States).
[247] See discussion supra Parts I & II.
[248] See discussion supra Part II.A.
[249] Givaudan, the largest fragrance manufacturer, dates its origins to 1786. See A Rich Heritage of Growth, Givaudan, (last visited Jan. 14, 2016). IFF was established in 1889. See History Timeline, Int’l Flavors & Fragrances, Inc., (last visited Jan. 14, 2016). Firmenich was established in 1895. See Company, Firmenich, (last visited Jan. 14, 2016). Symrise merged Haarmaan & Reimer (1874) and Dragoco (1919). See History, Symrise, (last visited Jan. 14, 2016). Only Takasago was established in the twentieth century. See History of Taskago, Takasago, (last visited Jan. 14, 2016).
[250] See generally, Podlas, supra note 230 and accompanying text.

Bridging the Gap: Improving Intellectual Property Protection for the Look and Feel of Websites

Bridging the Gap: Improving Intellectual Property Protection for the Look and Feel of Websites
By Liz Brown* A pdf version of this article may be downloaded here.  


Every company strives for a unique and memorable website. There is a growing threat, however, that this valuable investment in website design will be copied by competitors without effective legal remedy and with potentially devastating consequences. The “look and feel” of a website—the immediate impression that makes a website recognizable, easy to use, and deserving of consumer trust—is not adequately protected by copyright, trademark, or any other intellectual property doctrine. No intellectual property doctrine currently provides companies with a predictable way to ensure that a competitor does not copy their websites’ look and feel. Website look and feel protection falls into a chasm between copyright on one hand and trade dress on the other, neither of which adequately addresses this modern problem. While copyright protects fixed texts, it cannot offer the scope and fluidity of protection needed to capture the look and feel of entire websites. Trade dress falls short because existing law does not adequately address the blend of form and function essential to website protection. This article proposes the adoption of a multi-factor test adapted from trade dress law in order to secure more effective protection for websites and clarify that trade dress is the proper doctrinal home of “look and feel” protection. In evaluating the potential legal protections for web sites, it is important to distinguish the parts from the whole. Some website elements are relatively easy to protect. These elements include the array of fonts, text size, colors and spacing mainly controlled by a cascading style sheet, or CSS. Copyright may protect CSS, like other code, along with specific images or text on any particular web page.[1] The business name and logo qualify for trademark protection. While copyright law protects a website’s source code and fixed elements and trademark law can protect logos, intellectual property law does not protect the more fluid look and feel of the site overall. A competitor may copy a website’s look and feel without infringing the original site’s copyright in any particular text. A website’s look and feel is different from, but partly dependent on, the specific code that defines how the text and images appear on the website. It is the composite effect of the technical elements, together with design features, that creates a unique user experience. The lack of effective protection for the look and feel of websites presents a serious legal and economic problem. Websites are essential to modern business, the electronic bridge between companies and their customers and the basis of all e-commerce. As companies as large as Facebook are discovering, those vital assets are susceptible to outright copying without permission. Because of gaps in intellectual property law, competitors can copy the overall look and feel of a trusted website, tricking customers into patronizing a different website, with few clear repercussions. Unless this gap is filled, the problem is likely to get worse over the coming years with potentially severe commercial impact. The cost to businesses, in the forms of revenue, trust and goodwill, of such copying is vast. The cost to consumers, who provide confidential data and payments based on misimpression of commercial source, has sparked investigations by the Department of Homeland Security[2]. Although look and feel protection is not a particularly new issue, courts and scholars continue to struggle with it. Some courts have suggested that website look and feel protection should stem from copyright law, which protects the expression of ideas.[3] Others have adopted trade dress, with its emphasis on protecting a firm’s impression on consumers, as a more appropriate basis for protecting website look and feel.[4] Scholars have noted the doctrinal confusion in the few cases addressing look and feel protection but have not proposed a workable solution.[5] Recent case law and technological developments make it clear that a comprehensive solution must be found soon. The ubiquity of website copying, combined with the increasing commercial value of website look and feel, creates a significant but ineffectively regulated threat to online commerce. As no satisfactory judicial approach to the problem of website copying has yet emerged, the questions remain: What is the best way to protect the look and feel of websites from illicit copying? What is the proper doctrinal basis for such protection? What is the optimal mechanism for putting it in place? The underlying thematic differences between copyright and trade dress law—in essence, the difference between protecting a website owner’s expression and protecting consumer impression—suggest that trade dress is the more appropriate basis for look and feel protection. While copyright law protects the expression of ideas, trade dress law essentially protects the consumer’s association between a given product or place (or perhaps a website) and a particular source. The latter is more relevant to the harms posed by website look and feel infringement. courts across the country, however, continue to struggle with this issue. This article explains the nature of the problem, analyzes recent case law, explores the limits of copyright and trade dress as bases for protecting website look and feel, and proposes a multi-factor judicial test for look and feel infringement. Part I discusses why the gap in website look and feel protection exists and its potential impact on businesses and consumers. Part II explores why neither copyright nor trade dress protects the look and feel of websites well enough, and the ways in which recent cases expose the gaps in this area of the law. In Part III, I propose a new judicial approach in the form of a multi-factor test for look and feel infringement, adapted from trade dress law and modified to address technological realities. Part IV concludes by offering additional advice for addressing these issues under current law and suggesting possible next steps in resolving these and related issues.

I. Copying Websites’ Look and Feel: A Growing Threat to E-Commerce

The “look and feel” of a website refers to the overall visual impression that allows a consumer to identify the website with a particular service provider or merchant, and confirms that the user is on the website she intended to visit.[6] As one court explained, “when a person visits a web site, she is comforted by the distinctive design, knowing that the look and feel is clearly associated with a specific brand name. This branding facilitates the association with a firm’s reputation.”[7] At least, it should. Website look and feel infringement disrupts the consumer’s trusted association of site with source, resulting in damage to both the merchant and the consumer.

A. The Nature of Look and Feel Copying

The following hypothetical illustrates the problem of copying look and feel. Assume there are two similarly worded websites: and Although their domain names are similar, they have different business models. offers discounted vouchers to a range of participating restaurants. Consumers can buy $25 vouchers for local restaurants at the discounted price of $10, or $5 during one of their frequent promotions. The main function of, in contrast, is to aggregate customers’ restaurant reviews. It also offers some discounts, but vouchers are not the heart of its business. These websites’ patrons may not always remember the domain name of the website they want to visit. While some consumers bookmark a favorite website enabling their computer to recall the specific site, many others rely on a search engine. They can do so because they know that when they click on the “right” website, it will look familiar. Familiarity and past positive experience engender the trust necessary to encourage consumers to purchase a restaurant voucher and for the website to profit. When two similar domain names like and come up near each other in a web search, the visual distinctiveness of a website is especially important.[8] When websites look different from each other, and are viewed on a large enough screen, this works well. A problem arises, however, when one company decides to mimic the look and feel of a competitor’s website. If were struggling, it might consider a new business and marketing strategy, perhaps focusing more on voucher sales. It could then redesign its website to look more like, a leading competitor in that field. By changing its font to mimic the site, adjusting its spacing, changing its colors, rearranging its layout, and redesigning the search box, it can make its site looks more like By doing so, it can siphon business from consumers who reached in error but do not realize their mistake before entering their credit card information.[9] If the copycat redesigns its website in this way, the original faces significant potential losses in what is, more or less, a zero-sum game. It loses customers, income, trust and good will. The original site,, can sue for infringement of the look and feel of its website. will respond that it is competing legally because it has not infringed any of’s copyrights or trade dress rights. Who wins? There is no reliable answer to this question. No published case has established a claim for protecting the look and feel of websites. Although there is a growing body of case law on this issue, the few reported decisions do not form a coherent precedent. This problem, however, is far from hypothetical. It is something that even the largest websites face, including Facebook. In 2008, Facebook sued a German rival, StudiVZ, for infringement of its website look and feel.[10] StudiVZ had developed a website that looked and operated like Facebook, although it used a red background instead of blue. The StudiVZ site launched in Europe, where Facebook’s popularity had begun to grow.[11] Facebook brought suit in Northern California and Germany but retreated a year later, settling with StudiVZ in the United States because it could not succeed on its legal claims.[12] Athleta, a Gap-owned brand, faced a similar problem in 2012. One of its competitors, Pitbull, gradually changed its website design over the course of three years to more closely resemble Athleta sued Pitbull for trademark infringement, unfair competition and false advertising, among other claims, in the Central District of California. It accused Pitbull of copying large parts of its website, misleading consumers into thinking that Athleta operated the website. Pitbull’s copying included the adoption of’s graphic layout and even extended to including consumer reviews “that were identical to the ones appearing on plaintiff’s site describing plaintiff’s products.”[13] As the court noted, the defendants were “generally adopting a similar look and feel to plaintiff’s website for their website.”[14] The court granted Athleta’s motion for a preliminary injunction.[15]

B. The Economic Impact of Look and Feel Infringement

Businesses increasingly rely on websites as their calling cards and on e-commerce as their main source of income. As the commercial importance of website look and feel increases, so has the ease with which websites can be copied in their entirety. Many video tutorials demonstrate how to copy a website in five minutes or less.[16] Alternatively, several companies offer website copying services. One example is, which offers to copy a website for a flat fee of $49. As the frequency of look and feel copying has grown, entrepreneurs have created profitable businesses premised on look and feel infringement. Rocket, the company that created Facebook’s competitor StudiVZ, is a prime example. It has created successful knockoffs of well known websites including Fab, Priceline, and others. One journalist noted that Rocket’s clone of Zappos “now dominates six European markets and is estimated to be worth $1 billion by Financial Times Deutschland.”[17] The financial impact of website look and feel copying is potentially vast, as there is a lucrative market to capture. Websites are a powerful source of revenue. According to some scholars, “E-commerce in the United States is predicted to grow by over 40% between 2011 and 2015, reaching more than $270 billion in sales accounting for 15% of total retail sales excluding food products. In 2010, more than three-quarters of the North American industrial businesses that reported growth said that websites made a significant contribution to that growth. Overall, global e-commerce revenue is forecasted to hit $960 billion by 2013 and $1.4 trillion in 2015.”[18] Because e-commerce is growing so rapidly, the look and feel of a website provides a competitive advantage. The overall visual impression a website creates is the primary source identifier for businesses on the Internet.[19] The look and feel of a website increases in importance as people use smaller screens to access the Internet. In 2013, the Pew Research Center reported that 63% of mobile phone owners use their phones to access the Internet, and 34% of them use their phones, rather than a laptop or home computer, as their primary Internet device.[20] Nearly half of respondents in the 18–29 age group said that their mobile phone is their primary device for the Internet.[21] As more consumers view websites on mobile devices, on which it is often impossible to read domain names, businesses and consumers rely increasingly on the website’s overall visual impression as a source identifier.[22] The rapid growth of consulting companies that specialize in designing mobile-friendly web sites, like Mobiquity, attest to the impact increased mobile usage is having on web site development.[23] The smaller the screen, the more important visual recognition of the website becomes to the user’s brand, because the domain name becomes illegible. Knockoff websites pose risks for consumers who are fooled into buying counterfeit goods from them. In July 2012, the Department of Homeland Security announced an investigation into 70 different knockoff websites that had sold counterfeit goods to American consumers. John Morton, director of the department’s U.S. Immigrations and Customs Enforcement unit, noted that the copycat websites were using increasingly sophisticated methods to pass themselves off as genuine. “The fake sites and the real sites are almost indistinguishable,” Morton said. “And the fake sites aren’t offering obvious knockoffs. They are trying to masquerade as the real deal.”[24] The fake web sites had confused consumers, who bought goods on the fake sites under the misimpression that they were legitimate businesses. When consumers lose money due to website copying, the original website owners suffer as well. When Athleta sued its competitor for mimicking the look and feel of its website, for example, the court noted that Athleta had established a likelihood of confusion by demonstrating actual consumer confusion.[25] Consumers who mistakenly paid for goods through a competitor’s knockoff website “have called or emailed Athleta with complaints that they have not received order status confirmation emails from Athleta, that products they ordered have not yet arrived, or that their Living Social discount codes that they purchased are not being accepted by Athleta’s website.”[26] The court agreed that it damaged Athleta in the form of lost goodwill and reputation. “These consumers have expressed frustration to plaintiff with regards to the untimely shipment of merchandise or the non-receipt of confirmation emails, when in fact these consumers have ordered from defendants’ website.”[27] The fact that Pitbull was selling poor quality copies of Athleta’s goods caused further damage. “[T]o the extent the parties’ goods differ in quality, consumers’ satisfaction with plaintiff’s products may decrease as a result of these consumers purchasing defendants’ products.” The resulting loss of goodwill and reputation in the marketplace led the court to rule that Athleta was being irreparably injured by the defendants’ copying.[28]

II. Using Copyright or Trade Dress to Protect Website Look and Feel

When victims of website copying have turned to the courts, they have brought claims under a variety of theories, including intellectual property infringement and unfair competition. The two most promising and commonly discussed intellectual property claims related to this kind of copying are copyright and trade dress. As described below, neither offers entirely effective protection.

A. The Ambiguous Nature of Look and Feel Protection

Although there is relatively little case law on look and feel infringement, a few key decisions illustrate the confusion often wrought. The lack of certainty as to which intellectual property doctrine best protects the look and feel of websites is obvious from the plaintiffs’ claims and the courts’ responses in these cases. In one of the most frequently cited look and feel cases, Blue Nile v., a 2007 case from the Western District of Washington, the court struggled with the legal nature of plaintiff’s look and feel infringement claim.[29] An initial problem was that the plaintiff’s claims were less than clear. Blue Nile had asserted copyright protection for certain elements of its website, including some search features, and had registered them with the Copyright Office. At the same time, Blue Nile alleged state law claims relating to the “look and feel” of the website, which it claimed were outside of the purview of copyright. In stating those claims, however, Blue Nile incorporated by reference its copyright claims. The court crossly pointed out that “Plaintiff cannot both expressly rely on the copyright allegations in all of its state law claims and assert that the state law claims are outside copyright’s subject matter for purposes of avoiding preemption.”[30] The ambiguity in the court’s own language is also striking. In some respects, the court’s reasoning is internally inconsistent. At one point, the court asserted that the look and feel of a website is a matter of copyright law rather than trade dress (“The look and feel of plaintiff’s website is within the subject matter of copyright.”).[31] At another point, the court allowed that the look and feel of a website might also be a matter of trade dress law rather than copyright (“[What] portions of the website relate to the look and feel of its trade dress claims require greater factual development”).[32] Had the litigation continued, presumably, the court could have evaluated more precise claims as to what specific elements comprised the “look and feel” of the websites at issue. Unfortunately for scholars, the Blue Nile litigation ended with dismissal two weeks after the court ruled on the motion to dismiss. [33] There was also confusion on both sides of the bench in Conference Archives, Inc. v. Sound Images, Inc., an unreported case from the Western District of Pennsylvania.[34] Although Sound Images admitted that it copied Conference Archives’ website and “the facts [were] largely uncontroverted,” the court denied both parties’ summary judgment motions.[35] The court acknowledged the unsettled nature of the issue it faced, noting that “while protection of source code and other electronic processes have found refuge in copyright or patent law, protection of the “look and feel” of a website remains unclear” and calling the matter “a case of first impression in this Circuit.”[36] The novelty of the issue apparently perplexed the plaintiff, too. The court observed that “Plaintiff’s motion for summary judgment is nebulous as to what ground of intellectual property law it relies on.”[37] In its complaint, Conference Archives referred to its claim as rooted in copyright law and yet acknowledged that “no federal copyright claim is asserted here.”[38] It stated that the “’look and feel’ of a web site can be ‘protected in multiple ways, such as by copyrights, trademark, [and] trade dress [sic].’”[39] Taking matters somewhat into its own hands, the court decided to consider the website copying claims under trademark, trade dress, trade secret and copyright theories.[40] The parties, however, apparently settled before trial. In addition to doctrinal uncertainty, another problem with existing case law is the number of decisions that have issued at a preliminary stage rather than after a full hearing on the merits. For example, the often-cited Blue Nile v. decision resolved only the defendant’s motion to dismiss on preemption grounds.[41] In most cases, the courts have evaluated the claims under the relatively minimal standards of pleading sufficiency, or to resolve basic questions of preemption, instead of issuing a more complete evaluation of the issues after a trial. The preliminary stage at which these decisions are issued means that many of the critical issues have not yet been played out, including the extent to which functionality might bar a trade dress claim. In Salt Optics v. Jand, for example, the defendant had asserted that the plaintiff could not claim trade dress rights in the look and feel of its website because the website “is functional on its face and […] the “look and feel” of Plaintiff’s website has not acquired secondary meaning.”[42] While the court ruled that the Plaintiff had pled the elements of non-functionality and secondary meaning well enough to state a claim, it reserved “a more exacting factual analysis of these elements […] for a later stage of the case.”[43] The case settled, however, before the court could carry out such an analysis. While courts are slowly gaining more experience with look and feel claims there is still some confusion as to the proper nature of these claims. Website owners most frequently assert one of two kinds of claims: copyright and trade dress. The next sections address the viability of each.

B. The Limits of Copyright Law in Protecting Look and Feel

Because copyright law generally protects both legible text and software code, it is logical to start there as a potential basis for website protection.[44] Copyright law protects many components of a website, including the text an end user can read and the software code that specifies how that code will be displayed. Unlike more traditional art forms, the design features of a web page are controlled through a specific kind of code called a Cascading Style Sheet (CSS), which can be embedded in a website’s HTML. Like other kinds of software, CSS may be subject to copyright.[45][46] In order to bring a copyright claim for infringement of a website’s look and feel, however, the website owner must meet the threshold requirements of any copyright claim.[47] Specifically, the plaintiff must show (1) that it owns a valid copyright, (2) that the copyright has been registered with the Copyright Office, and (3) that the defendant has copied its protected work, which requires both copying-in-fact and substantially similar copying.[48] Applying these requirements in the context of website look and feel is not straightforward. Copying-in-fact can be demonstrated either by direct evidence or by showing access to the work and probative similarities.[49] Given the public nature of websites, it is hard to imagine a defendant successfully challenging the issue of access. While it is easier to establish access to the copied work, it may be more difficult to establish substantial similarity among websites, in part because of their interactivity. “An allegedly infringing work is considered substantially similar to a copyrighted work if ‘the ordinary observer, unless he set out to detect the disparities, would be disposed to overlook them, and regard their aesthetic appeal as the same.'”[50] Some simple changes in a copied website design may lead a court to conclude that no such substantial similarity exists, even where there is direct evidence of intentional copying-in-fact. In the New York case of Crown Awards Inc. v. Trophydepot, for example, the defendant conceded that it had intentionally copied plaintiffs’ website. Because the defendants had changed some colors and added some text, however, the court found that the knockoff website was not “substantially similar” and denied plaintiff’s copyright claim on that basis.[51] Facebook, however, pursued claims against StudiVZ for look and feel infringement even though the knockoff site used red as its primary color instead of Facebook’s hallmark blue.[52]
1. The Copyright Office May Not Register “Look and Feel” Elements
The registration of a valid copyright is required to assert copyright claims.[53] While the Copyright Office will accept registration of the specific source code or text to be used in a website, it sends conflicting signals as to whether it will accept registration of the website’s look and feel overall. As at least one court has recognized, source code itself is not a proxy for the look and feel of the website. In Athleta v. Pitbull, the defendants argued that the court should compare the source code of their website with that of the plaintiffs’ website in order to evaluate their similarity.[54] The court rejected that argument, noting that any similarity between the source codes of the website was irrelevant to the main harm claimed, which was a likelihood of confusion by consumers who might mistake the defendants’ website for that of the plaintiffs. “The court does not see how the source code of the parties’ respective websites could create a likelihood of confusion as to the source of defendants’ goods or to defendants’ affiliation with plaintiff; it is the overall look and feel of defendants’ website vis-a-vis plaintiff’s that causes confusion.”[55] Stepping back from the source code, then, the questions remains as to whether the Copyright Office will register websites overall. Its own publications suggest that it will not register websites, website designs or website layouts. In Circular 66, the Copyright Office offers guidance for registering online works. It advises authors completing an application for registration to do the following:
Use terms that clearly refer to copyrightable authorship. Examples are “text,” “compilation,” “music,” “artwork,” […] Do not use terms that refer to elements that are not protected by copyright or may be ambiguous, for example, “website,” “interface,” “format,” “layout,” “design,” “look of website,” “lettering,” or “concept.”[56]
The Copyright Office’s inclusion of the terms “website” and “look of website” among the “elements that are not protected by copyright or may be ambiguous” dissuades companies wishing to copyright their websites as a whole. This language is somewhat vague, since it is not clear whether “website” and “look of website” fall into the category of ambiguous elements or unprotectable elements. Regardless, this language suggests that neither may be registered. The Copyright Office explains that while the code underlying a website may be registered for copyright protection, the resulting website content may not be included in that protection unless it can be described according to the Office’s requirements:
For a claim in a computer program that establishes the format of text and graphics on the computer screen when a website is viewed (such as a program written in html), registration will extend to the entire copyrightable content of the computer program code. It will not, however, extend to any website content generated by the program that is not present in the identifying material received and that is not described on the application.[57]
The Copyright Office also suggests that it will not register the spatial arrangement of individually registered elements or the appearance of the website. A prior version of Circular 66 stated that “in general, formatting of web pages is not copyrightable” and that its “longstanding practice … is to deny registration of the arrangement of elements on the basis of physical or directional layout in a given space, whether that space is a sheet of paper or a screen of space meant for information displayed digitally.”[58] The July 2012 revision of Circular 66 deleted that language, but did not replace it with any more specific or encouraging guidance for website creators. One reason for the Copyright Office’s reluctance to register websites is their mutability. The frequently changing nature of web sites may be fundamentally incompatible with the fixation requirement of copyrightable works. Copyright traditionally protects only texts that are fixed and unchanging.[59] Indeed, the Copyright Office requires that a web page’s source code and text be separately registered each time they are updated, revised or otherwise changed, unless they are databases, serials or newsletters. Again, Circular 66 provides the following guidance:
Many works transmitted online, such as websites, are revised or updated frequently. Generally, copyrightable revisions to online works that are published on separate days must each be registered individually, with a separate application and filing fee[.][60]
Characteristically, and increasingly, website texts are not fixed.[61] They must change frequently in order to fulfill one of their purposes: as a means of updated communication between the business and the consumer. Consumers expect websites to change often to reflect, for example, news that affects the company, daily or weekly sales or promotions, and any other element of a marketing campaign. A strict reading of Circular 66 suggests that a website owner would have to register new content with the Copyright Office on a daily basis to ensure protection since each change is likely to result in a new work, even if the overall look and feel of the website remains the same. The administrative burden of filing new paperwork daily with the Copyright Office is probably more than most website owners wish to expend.
2. Judicial Use of Copyright Claims in Look and Feel Cases
Given the Copyright Office’s directives, one might expect that copyright law would not protect the look and feel of websites effectively. While the Copyright Office may register both the individual elements of a website and the overall arrangement of those elements in theory, its advice to authors suggests that it is unlikely to do so in practice. That seemed to be the case after Darden v. Peters.[62] In Darden, the plaintiff had created a website designed to help users locate real estate appraisers nearby. The website consisted largely of maps taken from census sources, which were colored and labeled. When Darden tried to register its website with the Copyright Office, the clerk denied his request. The adaptation of the maps, the clerk reasoned, did not reflect an original work of authorship and therefore was not entitled to copyright registration. Darden sued to overturn the denial. The court sided with the Copyright Office, concluding that the website was primarily a compilation of maps rather than an original work. While certain creative compilations may be entitled to copyright protection, the court ruled, it upheld the denial because it did not find anything especially original about Darden’s compilation.[63] In a later case, Salt Optics v. Jand, the court again refused to allow copyright protection for either the specific elements of a website or its arrangement of them.[64] The case arose from competition between two online eyewear sellers. Plaintiff Salt Optics had created a distinctive, modern-looking website to sell its line of eyewear, featuring sparse white backgrounds and blocky black type. Defendant Jand operates the popular eyewear website Warby Parker. Salt Optics asserted that the Warby Parker website had an “overall look and feel” that was confusingly similar to Salt’s, and claimed both trade dress and copyright infringement.[65] The court granted Jand’s motion to dismiss Salt Optics’ claim for copyright protection in the specific elements of its website. It based its decision on its view that Salt Optics had not claimed anything particularly original about the basic elements or their layout. The court categorized the elements as “(1) eyeglasses bearing names (2) smaller images of products displayed on one side of a web page, (3) an image of a model wearing the product on the other side of the web page, (4) a close-up photograph of the product shown at a distinct angle (5) multiple smaller images of the product displaying color choices, and (6) two different views of a model wearing the product.”[66] The court saw nothing protectable about any of them, and deemed them all incidental to the sale of eyeglasses online. “Each of these elements, considered alone, are, on their face, commonplace expressions naturally associated with the idea of selling a fashion accessory over the Internet,” the court observed.[67] As such, none was eligible for copyright protection individually. In addition, the court saw no reason why Salt Optics’ arrangement of these elements might qualify for copyright protection. Nor, it pointed out, had Salt Optics alleged that its arrangement was creative enough to merit such protection. As the court stated, “[a]lthough certain combinations of unprotectable elements may qualify for copyright protection (e.g. music notes combined together to form a composition), “commonplace” or “typical” combinations do not.”[68] Implicit in that observation, however, was the possibility of copyright protection for a demonstrably original array of elements on a website. Unfortunately for Salt Optics, the court regarded its particular compilation of elements as “standard, stock and common” and therefore unworthy of protection.[69] In at least one case, however, a court denied copyright protection for individual elements of a website but upheld copyright protection for their arrangement. The case, Allen v. Ghoulish Gallery, concerned two competitors who “design, create, market and sell ‘changing portraits’ in the haunt industry.”[70] Both the plaintiff and defendant were artists who “take an antique photo and use lenticular technology to make the portrait ‘change’ into a haunted spirit, vampire or other spooky creature.”[71] The plaintiff alleged, among other things, that the defendant copied the look and feel of its website. Although the court found that there was nothing copyrightable about the individual elements of the plaintiff’s website, it held that he was entitled to copyright protection for “the selection, arrangement and presentation of the contents of his website.”[72] Two other federal courts have suggested that copyright law may protect website design and layout.[73] The Darden and Salt Optics courts’ refusals to allow copyright protection has not stopped other companies from alleging copyright infringement as a means of protecting their websites from illicit copying. Indeed, more recent decisions also suggest that the Copyright Office may be more willing to register website elements than its publications imply. In an unpublished 2011 case from the District of New Jersey, an educational organization sued its former webmaster for converting its website and copying it for another organization. The plaintiff brought a number of state law claims including conversion and trespass to chattel.[74] The court noted that the defendant “registered the website contents […] with the U.S. Copyright Office, effective July 1, 2009” and that the “registration application included text, software, code and graphical elements taken from the [plaintiff’s] website.”[75] The Copyright Act, the court ruled, preempted the state law claims, prompting the removal of the case to federal court.[76] The parties apparently settled before any ruling issued on the copyright claim. While the Copyright Office may be increasingly willing to register certain components of a web page, whether the Copyright Office will register the entire site as a unit is a separate question. Under Darden and Salt Optics, the protection will not extend to the entire web page unless that page is a particularly innovative composition of elements. Whether a website’s design is innovative, however, is a different inquiry from whether it has a distinctive look and feel. Darden led at least one commentator to conclude that “a copyright claim concerning the overall layout and design of a website would be weak.”[77] Even if the Copyright Office were willing to register the overall composition of a website, it is not clear that copyright would effectively protect the look and feel of the site. The user’s overall impression of the website depends on more than the static visual layout that copyright would protect. It depends in part, for example, on the user’s interaction with the menus and other functionality of the site. It transcends the specific colors used in one part of the website or another, although such minor changes might prevent a competing website from being similar enough for infringement. The limits of copyright law to protect look and feel are becoming more obvious. Even if a website owner copyrights every page of its website, it will need to bring trade dress claims for the look and feel of that site because the overall impression of a website is more than the sum of the appearance of its pages. In Express Lien v. National Association of Credit Management Inc., plaintiff Zlien did exactly that.[78] It alleged that defendant NACM violated copyright law by copying Zlien’s compilation of materials related to mechanics’ liens and bond claim laws, as it appeared on Zlien’s website.[79] It also alleged that NACM violated trade dress law by copying its “stylistic choices along with the content of the website, including the color, font, and hyperlinks, in a manner that is confusingly similar to Zlien’s website.”[80] Zlien’s decision to assert separate copyright claims to protect its pages and compilation and trade dress claims to protect the look and feel of its website reflects an evolving awareness of each doctrine’s proper use in website copying cases. In sum, copyright is a less effective basis for website look and feel protection. The Copyright Office’s guidance on what may be registered for copyright protection is somewhat at odds with at least some case law. The level of detail at which these cases are reported makes it difficult to parse exactly what makes an element of a website, or the composition of a website, sufficiently original to merit copyright protection. Even if a plaintiff could demonstrate originality, the formal requirements of copyright registration, including the obligation to register revised text on a daily basis, diminish the appeal of using copyright to protect websites’ look and feel. The uncertainty of copyright as protection for the look and feel of websites may explain why Facebook did not include a copyright claim with the eight other causes of action in its complaint against StudiVZ.[81]

C. The Limits of Trade Dress Law to Protect Website Look and Feel

Given the limits of copyright, some plaintiffs turn to trade dress law instead to protect the look and feel of their websites. Trade dress, one of the less common forms of intellectual property, protects the overall image of a product, store, or potentially a website, which connotes a particular source, from illicit copying. Like trademark law, trade dress protection is primarily defined by the Lanham Act.[82] Perhaps the most famous example of such protected packaging is the iconic Coca-Cola bottle. Another well known example of trade dress protection is the size, shape and graphic design of National Geographic magazine. Its signature yellow border identifies the magazine with its source, regardless of whether one can read the “National Geographic” title on the cover. In August 2012, a jury in the Northern District of California found that Samsung had diluted Apple’s trade dress in the iPhone 3G because several of Samsung’s phones, including the Galaxy, had confusingly similar designs.[83] Trade dress protection is also available for retail stores, hotels and restaurants. This protection ensures that customers can reliably identify a particular business by the overall look and feel of the commercial interior. The characteristic white walls, floors, and fixtures of an Apple Store, together with other key design elements, are protectable as trade dress. So too is the combination of taupe walls, green accents and pale wood fixtures that tell a customer she is in Starbucks. In other words, it protects the affiliation in the consumer’s mind between a specific interior design scheme and a commercial source. Websites are in many ways the storefront equivalents of today’s commercial environment.[84] Trade dress therefore provides the most logical basis of intellectual property protection for the look and feel of a website. Yet the particular requirements of trade dress and the case law applying this type of protection to the look and feel of websites make its likelihood of success unclear. Indeed, even in July 2013, the District Court for the Eastern District of New York wrote that “the application of trade dress law to websites is a somewhat ‘novel’ concept” and an issue of “first impression in this Circuit.”[85]
1. Articulating the Elements of the Alleged Trade Dress
In order to bring a trade dress infringement claim for website copying, the plaintiff must first articulate the elements of the website that make up the trade dress at issue. Specificity and finality are both critical to the success of a trade dress infringement claim, especially for website copying. This point was underscored in Sleep Science Partners v. Lieberman, a 2010 decision from the Northern District of California.[86] Plaintiff Sleep Science made a mandibular repositioning (e.g., snore-reducing) device and sold it on its website. Sleep Science showed its product and website to Avery Lieberman. He showed them to two Vermont residents, who allegedly copied the website and launched a similar one under the corporate name Sleeping Well. Sleep Science alleged, among other things, that the Sleeping Well website “has the same format, design and feel as [its] website.”[87] It alleged trade dress infringement, and described its trade dress as the “unique look and feel of SSP’s website, including its user interface, telephone ordering system and television commercial[.]”[88] Sleep Science itemized various features of its website in its complaint, including the ability to view its commercial, user testimonials, a screening questionnaire and “the PureSleep method.”[89] Sleeping Well argued that this listing was too vague to give it sufficient notice of the trade dress claim.[90] The court agreed with Sleeping Well. “Although it has catalogued several components of its website, Plaintiff has not clearly articulated which of them constitute its purported trade dress,” the court observed.[91] The open-ended nature of the list also worried the court: “Notably, Plaintiff employs language suggesting that these components are only some among many . . . . Without an adequate definition of the elements comprising the website’s ‘look and feel,’ Sleeping Well is not given adequate notice.”[92] This decision underscored the importance of explaining both (1) the elements that give rise to a website’s look and feel and (2) the interaction among those elements clearly and specifically to the trier of fact.
2. Proving Distinctive Trade Dress
Articulating the elements of a claimed trade dress in the look and feel of a website is necessary but not sufficient. A plaintiff must also show that the alleged trade dress is distinctive and non-functional, and that there is a likelihood of consumer confusion between its alleged trade dress and that of the defendant.[93] Of these requirements, distinctiveness is the most elusive in the context of websites. Although Section 43(a) of the Lanham Act itself does not require distinctiveness, distinctiveness is required for registration under Section 2. As the Supreme Court has noted, “Nothing in § 43(a) explicitly requires a producer to show that its trade dress is distinctive, but courts have universally imposed that requirement, since without distinctiveness the trade dress would not ‘cause confusion . . . as to the origin, sponsorship, or approval of [the] goods,’ as the section requires.”[94] Two legal standards of distinctiveness in trade dress law have evolved in case law. This branching evolved in Wal-Mart Stores, Inc. v. Samara Brothers, Inc.[95] In Wal-Mart, the Supreme Court ruled that the distinctiveness of product design is measured differently from that of product packaging.[96] The difference between product packaging and product design is somewhat elusive. Product packaging refers to the appearance of the package a product comes in, as well as to interior design schemes found in stores, restaurants and hotels.[97] In order to be eligible for trade dress protection, the product packaging must meet the “inherently distinctive” standard, set in the Court’s 1992 ruling in Two Pesos, Inc. v. Taco Cabana Inc.[98] “Inherently distinctive” trade dress suggests a particular source to consumers without need for repeated association.[99] Product design, in contrast, refers to the way a product looks and feels. Product design can only be distinctive enough to warrant trade dress protection if it has acquired “secondary meaning.”[100] Secondary meaning refers to a level of distinctiveness acquired by longstanding use in the market, as opposed to the inherent design of the site.[101] In part because it must evolve over time, secondary meaning is intrinsically more difficult to demonstrate than inherent distinctiveness. Given the case law, it is unclear whether websites must be inherently distinctive or have acquired secondary meaning. One argument posits that because websites are more akin to storefronts than to tangible products, the inherent distinctiveness standard of Two Pesos is more appropriate. On the other hand, that standard may be hard to apply to website design in practice. In a commercial world where many different websites may have similar general layouts, as dictated by principles of efficient user experience, what should qualify as “inherently distinctive”? The few published decisions applying trade dress law to website copying suggest that the owners must demonstrate secondary meaning to state a trade dress claim, although no published decision has discussed whether this standard logically applies to websites in particular.[102]
3. Secondary Meaning in the Website Context
Courts consider several factors in order to determine whether secondary meaning exists, including:
i. Whether consumers in the relevant market associate the trade dress with the maker; ii. The degree and manner of advertising under the claimed trade dress; iii. The length and manner of use of the claimed trade dress; iv. Whether the use of the claimed trade dress has been exclusive; v. Evidence of sales, advertising, and promotional activities; vi. Unsolicited media coverage of the product; and vii. Attempts to plagiarize the trade dress.[103]
In Wal-Mart, Justice Scalia noted that it was entirely reasonable to require proof of secondary meaning in order to secure trade dress protection. Questioning the value of an “inherently distinctive” standard for product design, he observed that:
[T]he game of allowing suit based upon alleged inherent distinctiveness seems to us not worth the candle. That is especially so since the producer can ordinarily obtain protection for a design that is inherently source identifying (if any such exists), but that does not yet have secondary meaning, by securing a design patent or a copyright for the design . . . . [104]
In other words, the Court reasoned that any inherently distinctive design could be protected as well by copyright or by a design patent. The implication seems to be that manufacturers could seek relief from infringement of their allegedly proprietary product designs by resorting to copyright or patent law instead of or in addition to trade dress law. Unfortunately for website owners, as we have seen, copyright does not protect web sites effectively. It is also unlikely that a website would qualify for a design patent because it is not an “ornamental design for an article of manufacture” as 35 U.S.C. § 171 requires.[105] It is tempting to question the relevance of secondary meaning in today’s commercial context. One factor traditionally used to demonstrate secondary meaning is length of time the product design has been used in the market. Length of time in the modern marketplace, however, is not an efficient proxy for consumer identification. On the Internet, companies that have only been in business for a short time, and thus have only had operational web sites for a relatively short time, can develop stronger consumer associations than more established companies with less striking or resonant websites.
4. The Difficulty of Proving Non-Functionality
Protectable trade dress cannot be functional. The Ninth Circuit Court described the functionality test as follows:
“A product feature is functional if it is essential to the product’s use or if it affects the cost and quality of the product . . . . “Functional features of a product are features which ‘constitute the actual benefit that the consumer wishes to purchase, as distinguished from an assurance that a particular entity made, sponsored or endorsed a product.’”[106]
This raises a difficult set of issues for website protection. Every company wants its website to be functional, in that it encourages consumer use, but the overlap between website functionality and the “benefit the consumer wishes to purchase” can vary greatly. While some websites merely describe the product the customer wishes to purchase, other websites, such as search engines, review sites or booking sites, are the product the customer wishes to purchase. Accordingly, it is impossible to impose a single trade dress standard for all website look and feel cases with regard to their functionality. Courts have approached the non-functionality requirement in website cases in different ways. At least one court has denied trade dress protection for a website based on its functionality. In an unpublished 2008 opinion, the District Court of New Jersey analyzed whether visuals illustrating various features of the bond market on the plaintiff’s website were protectable under copyright or trade dress.[107] Because the visuals were inextricably linked to their function, the district court held that they could not meet the non-functionality requirement of trade dress. It found, however, that these functional features were entitled to copyright protection because of their creativity and “unique expression.” The decision did not extend copyright protection, however, to the look and feel of the website as a whole. Few cases discuss the functionality as a potential bar to trade dress protection of websites. In SG Services, Inc. v. God’s Girls, plaintiff (“SG”) owned an adult services web site and claimed that the defendant (“GG”) had copied various elements of SG’s web site.[108] Among other claims, SG asserted that GG had infringed its trade dress by using (1) the color pink and (2) certain stock phrases, including “they’re the girl next door” and “so you wanna be a suicide girl?” on its website. The SG Services court agreed with the plaintiff that the color and stock phrases were “merely adornment” and therefore eligible for trade dress protection.[109] Non-functionality is somewhat easier to assess in the context of a website promoting a physical product. In Card Tech International v. Provenzano, for example, the District Court for the Central District of California concluded that both the website for a cleaning card, which is used to clean the slots of credit card machines and other card readers, and the packaging of the card were nonfunctional and qualified as trade dress.[110] “There is nothing about the layout or overall appearance of the trade dress, both packaging and website, that enables the package or website, respectively, to function. […] The content of the website can be arranged differently; the package can have a different appearance. Neither the appearance of the packaging nor the website provides a benefit apart from identifying the source of the product.”[111] The court concluded on this basis that protecting the website as trade dress would not “impair competition in the industry.”[112] That said, courts have allowed trade dress claims to go forward where the copied website provided a service as well. In Conference Archives v. Sound Images, an unpublished case from the Western District of Pennsylvania, a competitor admitted copying a product that “displays recorded video within a web browser.”[113] The court ruled that the presence of some functional features in the website does not exclude the possibility of trade dress protection. The determination of functionality, the court held, should properly focus on the “look and feel” of the website overall than on the function of its individual elements. “Look and feel” could serve “several possible functions,” only some of which might bar trade dress protection: First, it can provide ‘branding, helping to identify a set of products from one company, [and s]econd, it [can] increase[] ease of use, since users will become familiar with how one product functions (looks, reads, etc.) and can translate their experience to other products with the same look and feel. If a ‘look and feel’ becomes functional, it can no longer avail itself of trade dress protection. Thus the look and feel must be distinctive, but nonfunctional. But, the mere presence of functional elements does not by necessity preclude trade dress protection. Rather, a web site may be protectable ‘as trade dress if the site as a whole identifies its owner as the creator or product source.’[114] The court’s observation that the look and feel of a website must be “distinctive, but non-functional” serves as a guide for other courts assessing the potential for trade dress protection.
5. Likelihood of Confusion in Look and Feel Cases
The last requirement for trade dress infringement is proof of likelihood of consumer confusion. While the specific factors vary among the circuit courts, the Ninth Circuit’s eight-factor test to assess this likelihood, including the following, is illustrative:
1. The similarity of the two trade dresses; 2. The relatedness of the two companies’ products or services; 3. The advertising or marketing channels each party uses; 4. The strength or distinctiveness of the plaintiff’s trade dress; 5. The defendant’s intent in selecting the mark, including evidence of intent to infringe; 6. Evidence of actual confusion; 7. The likelihood of expansion in product lines leading to more direct competition down the line; and 8. The degree of care that consumers are likely to use.[115]
Likelihood of confusion plays an important role in website look and feel cases. In Faegre & Benson LLP v. Purdy,[116] the plaintiff law firm sued a group that had copied its website, alleging trade dress infringement. Although the defendant had labeled every page of its website as a “parody,” the plaintiff alleged that the defendant’s website “feature[d] the same color scheme, layout, buttons, fonts and graphics” as its own.[117] The court concluded that the “overall dissimilarity” between the two sites, however, created a “low likelihood of confusion,” and the parody disclaimer further reduced that likelihood.[118] Consumer confusion on fake websites can cause real harm to the public as well as the website owners. As noted above, the Department of Homeland Security has launched an investigation into scores of knockoff websites because of the harm that results to defrauded consumers. One would expect state Attorneys General to investigate these knockoff websites as a matter of consumer protection. As more government agencies step up their investigations, it may become easier to develop proof of actual consumer confusion. Although trade dress is imperfectly suited to protect the look and feel of websites, it is the most sensible basis for an improved test for website look and feel infringement. The complexities inherent in proving secondary meaning, establishing non-functionality, and validating evidence of actual consumer confusion required by traditional trade dress law in the context of website claims suggest that a new solution is needed to address this unique form of infringement. The multi-factor test described in Part III will clarify many of these issues.

D. The Threat of Copyright Preemption in Look and Feel Cases

One problem resulting from the lack of clarity surrounding look and feel cases is the ongoing question of copyright preemption. Because neither copyright nor trade dress provides a sure basis for protecting the look and feel of websites, plaintiffs have asserted both claims at once. When they do so with insufficient clarity, they face the threat of copyright preemption. “Claims for copyright infringement under the Copyright Act and claims for Trade Dress Infringement under the Lanham Act are mutually exclusive,” one court explained. “Plaintiff[s] cannot receive a remedy for both.”[119] Whether or not a plaintiff claims copyright infringement, its claims may be preempted if the defendant can prove that the gist of its claims is within the scope of copyright law. Courts have “long limited application of the Lanham Act so as not to encroach on copyright interests.”[120] The preemption defense played a key role in Blue Nile v.[121] Blue Nile developed a unique website that allows consumers to search for a specific diamond from its inventory, and then select a setting for the diamond, resulting in a customized piece of diamond jewelry. The diamond search feature of Blue Nile’s website allowed consumers to perform a search using a number of preferences, including carat size, cut, and clarity, which are presented in a distinct layout. Blue Nile alleged that the defendant,, copied its proprietary search tool and incorporated that tool it in its own website. It also alleged that effectively infringed the “look and feel” of its website. While these were two distinct allegations, they are in some ways hard to separate in fact. The alleged copying of various elements was integral to copying the overall look and feel of the website. Blue Nile asserted both copyright and trade dress claims against, in addition to a number of state law claims. moved to dismiss the trade dress claims on the basis that relief for the gist of those claims was available under the Copyright Act. If there is an adequate relief for a claim under the Copyright Act, defendants argued, relief under the Lanham Act should be preempted. The court did not agree. It acknowledged that “[c]ourts limit application of the Lanham Act in areas traditionally occupied by copyright or where the copyright laws ‘provide an adequate remedy.’”[122] It also observed that “[p]arallel claims under the Copyright Act and Lanham Act, however, are not per se impermissible.”[123] That was not the end of the preemption analysis. “Although the Ninth Circuit has not addressed the issue of whether a website’s “look and feel” is protected under § 102, other circuits have held that a work may be unprotected by copyright under § 102(b) and yet be within copyright’s subject matter for preemption purposes.”[124] In effect, the court warned, certain elements might fall into a legal no-man’s land, where they could not be protected by either copyright or any other intellectual property law. A critical issue for the court, then, was whether allegations of copying the design and presentation of specific search features on a website should be analyzed under trade dress or copyright law. While the look and feel of a website might be protected by trade dress, the court reasoned, more evidence was needed to isolate exactly what constituted the look and feel of the website at issue.[125] The defendant also asserted copyright preemption as a basis for its motion to dismiss in Salt Optics v. Jand, but the court denied its motion. Relying in part on Darden v. Peters, and apparently in part on intuition, the court held that look and feel of a website as a whole cannot be copyrighted: “Although the Ninth Circuit has yet to rule on this issue, existing precedent and common sense indicates that, absent special circumstances, a website’s overall “look and feel” is not entitled to protection under the Copyright Act.”[126] Because it falls outside of the scope of copyright, the court ruled, there can be no preemption of that type of claim.[127] It was possible, the court observed, to obtain copyright and trade dress protection at the same time without running afoul of preemption standards. “Indeed, relevant to the present case, a website may contain original works that infringe another’s copyright and simultaneously present an overall “look and feel” that violates another’s trade dress.”[128] Adopting a multi-factor test for look and feel infringement rooted in trade dress law will mitigate the risk that a website owner’s look and feel claims will be preempted by copyright law. It will do so by dispelling the confusion that currently surrounds the proper judicial treatment of website infringement and clarifying the proper scope of any copyright claims the website owner may also assert. Until courts adopt some form of the multi-factor test, plaintiffs in look and feel infringement cases can best avoid dismissal based on copyright preemption by pleading their trade dress and copyright claims in the alternative. In Sleep Science Partners, for example, the “either/or” nature of the plaintiffs’ claims compounded the problem of determining preemption. “Plaintiff may have plead its trade dress claim in the alternative, accounting for a possibility that its website may not be copyrightable,” noted the court, but it did not. The court found it hard to tell where the trade dress claim ended and the copyright claim began. “[If] it intends to maintain a Lanham Act claim based on its website’s “look and feel,” in addition to articulating clearly the website features that comprise its alleged trade dress, Plaintiff must plead a “look and feel” that does not fall under the purview of the Copyright Act.”[129]

E. New Technology Requires a New Framework

Rather than offering a coherent precedent for litigants to follow, look and feel infringement cases offer inconsistent and incomplete reasoning on the best approach to website protection. In general, they compound rather than resolve the questions surrounding the legal protection of websites’ look and feel. Why has this issue been so difficult to resolve? One possible explanation is that websites represent a relatively new hybrid of product, text, service and marketing functions. They integrate decorative elements and functional ones in a way that is fundamentally novel and that has never required protection before. The appearance of a product used to be relatively straightforward and limited to two or three dimensions. Whether it was a can of soda or a motorcycle, each item looked more or less the same to everyone. Websites, however, are mutable. They often have multiple levels, and their specific appearance depends on the user’s interaction. Complicating this further is the fact that websites may look different depending on where and how they are viewed. Two consumers may see even the same page of a website differently if, for example, one of them is viewing the website on an iPad and the other is viewing it on an iPhone. The shape of the screen and its orientation (vertical or horizontal) will vary across devices. On tablets and phones, the view may change depending on how the device is held. The characteristic look and feel of a site may not be defined by a static screen, but by something that can be generalized across different devices.

III. Proposing a Trade Dress-Based Solution for Look and Feel Infringement

If the two most obvious doctrinal homes for a solution are copyright and trade dress law, as discussed above, the nature of the doctrines makes it clear that the latter is more appropriate. One productive way to think about these two areas is the difference between protecting expression, through copyright, and protecting impression, through trade dress. While websites express many things, the gravamen of this problem is protecting the user’s first impression of the source of the site. That is what the Lanham Act was written to protect. Trade dress, with its emphasis on impression rather than expression, source identification and consumer confusion, is inherently better suited to protect website look and feel. Grounding a new solution in trade dress law might happen in either of two ways. One option would be to amend the Lanham Act to address the infringement of website trade dress specifically. Laws do evolve to address changes in technology, although they generally move much more slowly than technology itself. In 1999, for example, Congress passed the Anti-Cybersquatting Consumer Protection Act, amending the Lanham Act to prohibit the registration and use of domain names based on another person’s trademark.[130] In 2003, Congress amended the Criminal Code to make computer hacking illegal.[131] While there may be a legitimate need for a Lanham Act amendment to protect website look and feel, the urgency of the problem demands another potential solution. An alternative idea is to encourage courts to adopt an infringement test that is well suited to the unique nature of websites and yet grounded in the traditional concerns of trade dress law. This test might resemble, in its basic form, the multi-factor test for likelihood of confusion that courts have developed based on the Polaroid Corp v. Polarad Electronics case.[132] One form of such a test follows.

A. A Multi-Factor Test for Website Look and Feel Infringement

A test to determine whether the look and feel of a website has been infringed should take a number of factors into account. The range of factors should be calibrated to ensure that a court does not restrict fair competition in the process of protecting and encouraging innovation in website development. An outline of the factors looks like this:
1. Overall Similarity

a. Fonts and Formatting

b. Color Scheme

c. Sounds, Animations, Visual Effects

d. Symbols, Logos and Marks

e. Layout and Arrangement

f. User Experience Design

2. Proximity of Products or Services in the Relevant Market(s) 3. Intentional Copying 4. Likelihood of Consumer Confusion
In taking all of these sub-factors into account, no single element should be dispositive. It is possible to have confusingly similar websites even when, for example, the colors of the two websites are somewhat different. The determination of overall similarity should be holistic. In addition, the trier of fact should be guided by expert testimony from user experience designers, as described in more detail below.
1. Overall Similarity
The first factor is the similarity of the overall impression between the two websites, when comparing views on similar devices. The look and feel of a website is the result of multiple elements working together in a complex way. The imperative, stated in cases like Sleep Science Partners, for plaintiffs alleging trade dress infringement to explain which elements of the website work together to create that proprietary look and feel reflects that fact.[133] In Salt Optics v. Jand, the court required the website owner to provide both a detailed list of the elements giving rise to its website’s distinctive look and feel and an explanation of how they work together.[134] The court granted Jand’s motion to dismiss the trade dress claims, with leave to amend, on the grounds that Salt Optics did not give sufficient notice as to what constituted the look and feel of the website. Although the plaintiff had listed some elements of its alleged trade dress, the court observed that “absent from the FAC is any attempt to synthesize these elements in order to describe the way that the listed elements, in conjunction, combine to create the website’s protectable ‘look and feel.’ The lack of any such synthesis, in conjunction with Plaintiff’s expressly stated intention to incorporate other elements of the website into the trade dress claim at a later stage, gives the court pause.”[135] When Salt filed an amended complaint, Jand moved to dismiss again.[136] This time, however, the court was satisfied with Salt’s amended claims.[137] It noted with approval that Salt had submitted a “fixed and finite” list of website elements, helpfully juxtaposing images from Salt’s website with images from the competing website so that the court could appreciate the overall similarities between the two. Because it is difficult to assess overall similarity without more specific points of comparison, the test should include several sub-factors to help the court assess similarity.
i. Fonts and Formatting
One such sub-factor concerns the fonts used on the senior website and the junior website. The court should compare the specific fonts used in each site, taking into account any customization of the fonts and any associated text formatting. Many corporations customize fonts for their marketing materials, [138] including website texts, often at great expense[139]. Examining the specific fonts used on each of two competitive websites can increase the granularity of the comparison process.
ii. Color Scheme
Another sub-factor that helps a trier of fact assess overall similarity is the website’s color scheme. As the court noted in Conference Archives, website colors can be ascribed precisely because they are determined in accordance with a specific hexadecimal notation system. That is, each color on a website has a six-digit code. Accordingly, when a court compares the color scheme of two websites, it can make a more specific comparison than “light blue vs. medium blue.” The court’s comparison should include not only the exact colors of each site but the predominance of each color on the websites. In Conference Archives, the court parsed the technology behind the look and feel of a website in more detail than any court had done before. The court noted that there are “three technical elements that determine how a web site appears: colors, orientation and code elements.”[140] It explained that the colors used on websites can be described more exactly than traditional paint colors because of their digital nature. Website colors are coded according to a hexadecimal notation system that converts every color into a six number code known as a hex triplet. As a result, the court noted, it was possible to reach a greater degree of specificity when comparing the colors of two websites. This greater specificity, the court noted, allowed a greater degree of certainty about potential copying: “While some colors are more common than others, if two products utilize the same exact hex triplet, there is a likelihood that the color was copied.”[141] An effective comparison of website colors will include not only their technical differences but also the resulting difference in user experience.[142] Website colors are chosen carefully, not only to reflect firm branding but also to improve user experience and maximize accessibility (for example, to color-blind users). Effective use of color is integral to good website design.[143] The impact of different colors on website users has been well documented, but the consequences of certain color changes may not be immediately obvious to a trier of fact.
iii. Sound Effects and Animation
A third sub-factor is the similarity of sounds, animations, and other effects associated with the two websites at issue. As an increasing number of websites incorporate unique sound effects, fixed animations, and embedded videos, these should be taken into account when comparing the overall look and feel of a website. Although none of these would be apparent from a singular screen shot of a website (or even a series of screen shots), they all contribute to the overall user experience and the likelihood of consumer confusion.
iv. Symbols, Logos and Other Marks
In addition to the sounds and animations, the comparison must also take into account the similarity of any symbols, logos, marks, motifs, or other non-textual visual elements on the site. The court need not resolve the question of trademark infringement to find that similarities between marks contribute to the overall similarity of two websites’ look and feel. In Athleta v. Pitbull, for example, the court compared the logos of the two sites.[144] In mid-2012, Pitbull had affixed an “athletic” logo to its goods that closely resembled that of
Original Athletica Logo
By the time of the lawsuit, Pitbull had changed its logo to look somewhat more distinctive.
Later Athletica Logo
Even with the change in Athletica’s logo, the court found that the similarity between the two marks contributed to consumers’ likelihood of confusion. Importantly, the court ruled that the marks should be viewed in the larger context of the websites overall. “[T]he marks themselves are only part of the inquiry. These marks coexist in a marketplace that has changed as a result of defendants’ gradual encroachment on plaintiff’s marks. […] [T]his encroachment has taken many forms, including defendants’ … generally adopting a similar look and feel to plaintiff’s website for their website.”[145] The court’s reasoning has merit. Whether a competitor’s mark infringes a trademark is a separate inquiry from whether the use of that competitor’s mark contributes to the overall similarity of the look and feel of the websites at issue. A competitor’s use of marks should play a role in the look and feel assessment, but as with every other factor, should not be dispositive in itself.
v. Layout and Arrangement
Moreover, the court should consider the layout and arrangement of all content on the website, including its menus, navigational elements and visual elements, even though many of those elements may themselves be functional. The fact that graphic layouts can be described by reference to the number of pixels between elements might allow the court to infer copying by comparing pixel distance instead of merely eyeballing similarities between layouts.[146] Importantly, the court’s comparison should ensure that the websites are viewed on comparable devices. Many websites are designed to adapt their appearance to the device that is being used to view them, like technological chameleons. This aspect of website development is called responsive design. Responsive design is an increasingly common element of website design. Its purpose is to ensure that users have a consistent, intuitive experience of the website, no matter what device it is viewed on. Responsive design also helps website owners address differences in detail across viewing platforms. Because laptop screens allow users to view a website in relatively greater detail than a smart phone, for example, a responsively designed website will display different but equivalent displays on each device. Although the specific display will differ from device to device, the overall look and feel of the website will be the same.
vi. User Experience Designer Testimony
The few issued look and feel decisions suggest that courts are not uniformly well equipped to assess the effect of these factors on the user’s experience of a website. Indeed, it can be difficult for any lay person to appreciate the ways in which graphic and interactive factors affect the overall user experience of a website. The interactivity of a website adds another dimension to what has traditionally been a fairly straightforward comparison of appearance in trade dress cases. In Sleep Science Partners v. Lieberman, the court suggested that the “look and feel” of a website is something more than, or different from, a rote list of the website’s constituent elements. This raises the possibility that courts would benefit from reliable evidence as to what, exactly, gives rise to the specific look and feel of a website. Plaintiffs may benefit from introducing, for example, expert testimony from user experience designers. These designers can testify about the impact of connections among specific website elements on the look and feel of a website from the user’s point of view.[147] They can also help the court assess usability as a potential element of the website’s overall look and feel. Such testimony would be more helpful to the trier of fact than the court’s own subjective impressions.[148] Just as courts consider expert testimony in assessing economic damages, it may be helpful for a trier of fact to receive testimony from an expert in website design. Such expert testimony should include (1) an exhaustive list of all website elements that give rise to the unique look and feel of the website and (2) a description of how each one of these elements affect user experience and contribute to the users’ ability to connect the website with its source. The list should be finite and final. It is important to avoid disclaimers that imply the possibility of later amendment, such as “by way of illustration, not limitation.” As Salt Optics v. Jand demonstrates, such open-ended claims provide an easy mark for the defendants’ motion to dismiss. A lack of finality in the list of alleged trade dress elements may well be read as a lack of sufficient notice for the defendant. While the defendant will submit its own supporting expert testimony, as in Markman hearings, the detailed descriptions provided by both sides will help the trier of fact reach a more informed decision.
i. Proximity of Products or Services in the Relevant Market(s)
In addition to overall visual similarity, several other indicia of likely copying should factor into the court’s analysis. One additional factor should be the proximity of the products and/or services offered on the senior and junior websites. As with trademarks, the likelihood of consumer confusion is heightened when the two competing websites are in the same field. Because similar products can be marketed in the same way, the extent of similarity among other websites in the same industry or field should be taken into account as well.
ii. Intentional Copying and Other Fraudulent Measures
Any evidence of intentional copying, including replication of HTML and/or CSS code, should play into the court’s analysis as well.[149].[150] Unlike traditional print media, websites can be replicated by copying code. As discussed above, there is a thriving website copying industry, and providers of such services are ubiquitous. While intent has been criticized as a factor in some kinds of IP infringement cases,[151] courts should consider it here because of the deterrent impact it would likely have on future infringement. Reviewing evidence of intentional copying would allow the court to consider the extent to which the website copying was strategic, as opposed to accidental. Fake websites can use a variety of strategies to lure consumers into providing their credit card information and other sensitive data. One common strategy is “typo-squatting,” in which the infringer registers the copycat website at a domain name that is a close misspelling of the original site, hoping that consumers who mistype the domain name will not notice the difference. [152] Another is the use of meta-tags, in programs such as Google AdWords, which drive traffic on search engines to a knockoff website instead of the original.[153]
iii. Likelihood of Confusion
As with more traditional Lanham Act claims, evidence of likely consumer confusion should play a central role in the court’s analysis. The court should consider evidence of actual consumer confusion, as indicated by customer surveys and other data, in addition to likely confusion. As with all such data, the weight accorded to these reports should depend in part on the methodology used. While evidence of actual confusion may be the best indicator of likely confusion, it may also be the hardest to prove. User comments tending to suggest that the users are confused about which website they have visited may be tempting to include in a complaint. Because so many people use fictitious or partial names when commenting on websites, however, it may be nearly impossible to verify their identity, let alone get them to sign sworn testimony. In order to prove likelihood of confusion, as required for trade dress infringement claims, plaintiffs should submit side-by-side comparisons of their original websites and the alleged infringing site. A comprehensive comparison of the two websites will help the court understand every point of similarity. Including technical details for each page of the website, including the hexadecimal notation of the colors used and the pixel distance between elements, will also support a finding of likelihood of confusion.[154] A side-by-side comparison, however, illustrates the effects of these technical details on the consumer and, in turn, the likelihood of confusion.
iv. Additional Considerations
One issue that should not play a role in the court’s decision is the proximity of competitor’s site in search engine results. Search results differ based on a variety of factors, including the skills of a coder in search engine optimization and the geographic location where the search is being performed. The fact that two websites come up next to or near each other in a search engine result should not be considered evidence of “look and feel” infringement or its absence.

B. Designing an Effective Damage Scheme

Designing a damages scheme that provides a sufficient deterrent to website copying is a particularly important issue for further study. Because it is so easy to copy another website, as we’ve seen, courts must ensure that their damages awards are severe enough to deter such copying. But what is the most effective and efficient form of penalty for copying the look and feel of a website? While a full answer to this question deserves greater study, some starting points are clear. Courts may adapt remedies for website infringement from notable trade dress cases. One noteworthy example is the trial court’s penalty in Taco Cabana v. Two Pesos. The Two Pesos restaurant chain, found guilty of willful trade dress infringement, was required to post a sign on the interior and exterior of each of its locations for one year, notifying customers that it had been found guilty of copying Taco Cabana’s concept.[155] Adapting this “sign of shame” remedy to websites, a court could require an infringing website to post a banner advertisement explaining the verdict and directing users to the original website. At least one court has issued a similar directive.[156] Some observers have characterized changing a website’s colors as a relatively simple process, and one that is less time consuming than redesigning a product or a store interior. On that basis, it is tempting to conclude that requiring a look and feel infringer to redesign a website, may not be as effective a deterrent as it might be in the retail context:
Losing a trade dress claim involving a tangible product like a children’s toy could cost the copycat millions in packaging redesign, withdrawing and destroying infringing packages, revising promotional materials, and retooling factories. By contrast, losing a web site trade dress case may mean little more than asking a junior IT consultant to spend a few hours changing the site’s design and color scheme.[157]
Although a website can be changed quickly, quick changes are not always effective from a marketing or user experience perspective. The perspective quoted above on the time it takes to revamp a website is too dismissive, and reflects a misperception that good website design is arbitrary. In fact, compelling website design is as much science as art. Most companies hire professional website designers who understand that the apparently simple act of “changing the site’s … color scheme” is not so simple. As noted above, for example, the choice of color can have a significant psychological effect on a user’s experience of the site, so a site-wide color change cannot be undertaken lightly without changing the potential effect on consumers.[158] Even so, changing a website’s colors may not be enough to distinguish its look and feel from a competitor if other graphic elements are identical. StudiVZ, the knockoff website that incited Facebook to sue, featured red as its predominant color instead of Facebook’s blue – yet Facebook still concluded that the similarities between the sites were troubling enough to merit a set of lawsuits. The scope of a proper remedy requires attention to the realities of website design. Some courts have issued injunctions that go too far in response to look and feel claims, without regard to these concerns. The Athleta v. Pitbull case, for example, suggests the dangers of an insufficiently specific injunction in remedy for look and feel infringement. In that case, Athleta successfully argued that it was suffering irreparable harm as a result of Pitbull’s look and feel infringement. The court issued a preliminary injunction forbidding Pitbull from, among other things, using “any graphic, textual or other design elements on website that are similar to plaintiff’s website.”[159] The terms of that injunction are troublingly vague. What constitutes “graphic, textual or other design elements,” for example? Athleta’s website, like most, contains an elliptical search button, links to customer support information, and other graphic and textual elements that may not contribute to its distinctive look and feel but are essential to a legitimate consumer-facing website. Prohibiting another website from using any of those elements prevents that website from competing fairly in the online marketplace. Another example of courts’ overreaching in the definition of a protectable look and feel comes from the Conference Archives decision.[160] Unlike earlier decisions concerning the look and feel of a website, the court suggested that “look” and Feel’ be considered separately. In distinguishing a website’s “look” from its “feel,” the court noted that websites are unlike posters, in that they involve user interaction. It went on to describe a website as having a series of functional “layers.” One layer, it noted, was the visual design or appearance of the site on the screen. A second layer is the user’s interaction with the website, or the interface design. The court characterized the visual design as the “look” and the interface design as the “feel” of a website.[161] Functioning in harmony, the court noted, these elements can give rise to a protectable website “experience:” “Combined, the ‘look’ and ‘feel’ coalesce to form a protectable virtual experience that provides the user with ‘cognitive absorption’ […]. This interface promotes the efficient, predictable, and reliable use of a web site. The hallmark of a protectable “look and feel” trade dress is a graphical user interface that promotes the intuitive use of the web site.[162] This last statement reflects a troubling view of protectability. Extending protection to a graphical user interface because it “promotes the intuitive use” of a website is not in the best interests of consumers. In general, all web sites should be intuitive and easy to use. The professional fields of user experience design and computer-human interaction are dedicated in part to maximizing the intuitive use of websites. To be sure, two well designed websites that are not confusingly similar can both be easy to use. There should be no protection for this essentially functional aspect of a website. Ease of use is one aspect of website design that should be ubiquitous rather than proprietary.

IV. Additional Issues

There is a rich stream of issues related to the look and feel of websites that is beyond the scope of this article. Additional scholarship on these issues will improve our understanding of the best approach to protecting websites from illicit copying. One critical issue is the extent to which global intellectual property laws can protect U.S. companies against website look and feel copying abroad. As the Facebook vs. StudiVZ case demonstrates, the websites of U.S. companies may be copied overseas without an obvious, effective remedy. Even if U.S. law ultimately provides a satisfactory defense against website look and feel copying, the borderless nature of the Internet and global commerce in general demands a global response to the problem. Whether the problem can be remedied adequately through the Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPS) binding on all WTO members, or requires protection through a different treaty or another form of intervention, bears further study. Given the borderless nature of the Internet, one critical area for further study might be a comparison of the legal protection available for the look and feel of websites in other countries and in the European Union. Facebook’s unsuccessful attempt to sue the creators of StudiVZ in Germany, their host country, invites questions as to how any United States corporation might protect the look and feel of its websites abroad. Another question is the extent to which the World Intellectual Property Organization is a proper venue for resolving look and feel claims. In at least one case, the WIPO agreed to transfer the domain name of a knock-off website to another party in response to a look and feel infringement claim.[163] The complainant owned LeSportsac, a trademark used on women’s handbags and luggage in use since 1974. In August 2012, the respondent, an Arizona resident, registered the name “” for a website carrying counterfeit LeSportsac goods. In evaluating LeSportsac’s claims, the arbitrator noted that “Respondent’s site shares many of the distinct elements of the ‘look and feel’ of Complainant’s website, including the prominent use of the LESPORTSAC mark at the top left and a vertical column on the left side of the home page in which LeSportsac products are divided by category.”[164] Finding that the respondent used the website in bad faith, the arbitrator ordered the transfer of the domain name to LeSportsac.[165] An additional area of research might be to compare the effectiveness of private law remedies, including breach of contract claims, with some intellectual property theories described in this article as a basis for protecting the look and feel of websites. The provision that “no part of this website may be reproduced or transmitted” is often included in a website’s “Terms and Conditions.”[166] Competitors who access websites for the purpose of copying them arguably subscribe to those terms. Website owners therefore may protect against illicit copying by asserting breach of contract claims against such competitors. A particularly interesting topic for smaller companies is the extent to which they may be liable for website copying based on their adaptation of a standard website template. Companies such as Hubspot and WordPress offer “one stop shopping” convenience for companies in need of a new website. While the range of basic website models they offer may be impressive, it is limited. It would seem to be only a matter of time before the owner of one website finds a competing website developed from the same professional source. While adopting a specific test can clarify the legal issues surrounding website look and feel infringement, it is an imperfect solution. The issue of copyright preemption, for example, is still troubling. The Copyright Act intended to provide an exclusive remedy for anything subject to copyright protection, and it may be difficult in practice to draw the line between copying several individually copyrightable web pages and copying the look and feel of a website. A more detailed trade dress law, no matter how carefully crafted, cannot encroach on that territory. If courts were to grant copyright protection for creative and original compilations on websites, it would be more difficult for website owners to argue that their trade dress claims were not preempted by the Copyright Act. As always, in matters of intellectual property protection, the legal system must strike a delicate balance. If the courts were to grant website designers too much protection, that practice could result in a monopoly of highly intuitive, useful website designs. On the other hand, too little protection could stifle innovative website design, increase the chances of consumers being defrauded by counterfeit websites, and impede the normal growth of e-commerce.


There is a critical gap in the legal protection available for one of today’s most valuable business assets: the consumer-facing website. Although every company strives for a unique and memorable website, there is a growing threat that what makes a website memorable – its “look and feel” – may be copied wholesale by competitors without clear legal recourse for the website’s owners. Website look and feel protection falls into a chasm between copyright on one hand and trade dress on the other, neither of which adequately addresses this modern problem. The lack of effective protection for the look and feel of websites presents a serious legal and economic threat. The cost to businesses, in the forms of revenue, trust and goodwill, of such unlicensed copying is vast. The cost to consumers, who provide confidential data and payments based on misimpression of commercial source, has sparked investigations by the Department of Justice. This article has explained the nature of the problem, analyzed recent case law, explored the limits of copyright and trade dress as bases for protecting website look and feel, and proposed a multi-factor judicial test for assessing look and feel infringement. The underlying thematic differences between copyright and trade dress law – in essence, the difference between protecting commercial expression and protecting consumer impression – suggest that trade dress is the more appropriate basis for look and feel protection. Adopting a trade-dress based multi-factor test will increase the stability of intellectual property protection for the look and feel of websites and provide greater security for businesses and consumers alike.
* Assistant Professor of Business Law at Bentley University in Waltham, Massachusetts. The author gratefully acknowledges the assistance generously given by participants in the 2013 American Business Law Journal Invited Scholars Colloquium, who discussed a draft of this paper. The author can be reached at
[1] See, e.g., H.R. Rep. No. 94-1476, at 54 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5667 (“The term ‘literary works’ . . . includes . . . computer programs to the extent that they incorporate authorship in the programmer’s expression of original ideas, as distinguished from the ideas themselves.”); accord Atari Games Corp. v. Nintendo of Am., Inc., 975 F.2d 832, 838 (Fed. Cir. 1992).
[2] See infra note 23.
[3] See, e.g., Allen v. Ghoulish Gallery, No. 06CV371 NLS, 2007 WL 4207923 (S.D. Cal. Nov. 20, 2007).
[4] See, e.g., Card Tech Int’l, LLLP v. Provenzano, No. CV 11-2434 DSF PLAX, 2012 WL 2135357 (C.D. Cal. June 7, 2012).
[5] See, e.g., Lisa M. Byerly, Look and Feel Protection of Web Site User Interfaces: Copyright or Trade Dress? Santa Clara Computer & High Tech. L.J. 221, 246 (1997); Xuan-Thao N. Nguyen, Should It Be a Free For all? The Challenge of Extending Trade Dress Protection to the Look and Feel of Web Sites in the Evolving Internet, 49 Am. U. L. Rev. 1233 (2000); Gary Franklin & Kevin Henry, Protecting Your Company’s Website: The Application of Intellectual Property to the Digital Marketplace, 37 Vt. B. J. 28 (2012).
[6] The term “look and feel” has evolved in the case law. See, e.g., Blue Nile, Inc. v., Inc., 478 F. Supp. 2d 1240, 1241-42 (W.D. Wash. 2007); Conference Archives, Inc. v. Sound Images, Inc., No. CivCiv. 3:2006-76, 2010 WL 1626072, at *11 (W.D. Pa. Mar. 31, 2010); Sleep Sci. Partners v. Lieberman, No. 09-04200 CW, 2010 WL 1881770, at *8 (N.D. Cal. May 10, 2010); Athleta, Inc. v. Pitbull Clothing Co., Inc., No. CV 12-10499-CAS FMOX, 2013 WL 142877 at *7 (C.D. Cal. Jan. 7,, 2013).
[7] Conference Archives, Inc., 2010 WL 1626072, at *1515.
[8] The issue of look and feel infringement is different from that of cybersquatting. Two websites legitimately may have similar domain names even if they have similar or identical business models. Unlike trademarks, domain name registration does not require a unique field of use. Look and feel infringers need not have similar domain names in order to confuse customers into false associations. Similarly, cybersquatters need not have visually similar websites to the websites from which they divert traffic.
[9] Purchasing Google Adwords or using other tagging strategies can maximize the success of this mimicking strategy. Scholars have debated whether doing so should support additional claims of trademark or copyright infringement. See, e.g., Rachel Friedman, No Confusion Here: Proposing a New Paradigm for the Litigation of Keyword Advertising Trademark Infringement Cases, 12 Vand. J. Ent. & Tech. L. 355 (2010), and Stephanie Lim, Can Google be Liable for Trademark Infringement? A Look at the Trademark Use Requirement as Applied to Google AdWords, 14 UCLA Ent. L. Rev. 265 (2007).
[10] See Complaint at 75-119, Facebook, Inc. v. Studivz Ltd., No. C08 03468., 2008 WL 2914576, at *1 (N.D. Cal. July 18, 2008).
[11] See Caroline Winter, How Three Germans are Cloning the Web, Businessweek (Feb. 29, 2012),),
[12] Facebook settled its Northern District of California case with StudiVZ in September 2009, three months after a German court ruled against Facebook in its German litigation, finding inter alia that StudiVZ’s actions did not constitute copyright infringement under German law. See Kevin D. Hughes & David E. Rosen, Screen Grabbing: The Marketplace, Rather than the Courthouse, May Determine the Ultimate Winner in Web Site Infringement Battles, 33 Los Angeles Lawyer 4, 40 (2010),
[13] Athleta, Inc. v. Pitbull Clothing Co., Inc., No. CV 12-10499-CAS FMOX, 2013 WL 142877 at *3 (C.D. Cal. Jan. 7, 2013).
[14] Id. at *7.
[15] Id. at *13.
[16] See, e.g., Phentic Tutorials, How to Copy Any Website, YouTube (July 1, 2010),; Ilan Patao, Tutorial – How to Copy / Duplicate Web Sites, YouTube (May 19, 2008),
[17] Winter, supra note 12. Rocket also invested in copycat versions of Facebook and YouTube, which sold for $112 million and $36 million respectively. When Rocket cloned Groupon, the resulting site became Europe’s most popular deal-of-the-day site within five months of its launch. Id. Its success indicates that there is a lot of money to be made from knockoff websites, if the law permits. But see Katie Linsell, Facebook Wins Competition of Clones as Europeans Seek Global Network: Tech, Businessweek (Nov. 7, 2011),
[18] Franklin & Henry, supra note 6.
[19] Cf. Peep Laja, First Impressions Matter: The Importance of Great Visual Design, Conversion XL (Nov. 15, 2012),
[20] Pew Research Internet Project, Cell Internet Use 2013,
[21] Id.
[22] See Global Entertainment and Media Outlook: 2013-2017: Internet Access, PricewaterhouseCoopers, (last accessed Feb. 23, 2014) (showing penetration of mobile-Internet services will reach 54% by year-end 2017 compared with 51% for fixed broadband); see also Chaitrali Amrukar et al., Measuring SSL Indicators on Mobile Browsers: Extended Life, or End of the Road?, In Proceedings of the 15th international conference on Information Security (ISC’12) 86, 91 (Dieter Gollmann & Felix C. Freiling eds. 2012) (describing mobile browser’s limitation on visibility of domain name when rendering screens).
[23] See Erika Morphy, What’s Ahead for the Net: Mary Meeker Explains It All, ECommerce Times (May 30, 2013), (predicting explosive growth of mobile technology and wearable computing); Dave Uhler, Catching the Mobile Commerce Wave, ECommerce Times (Apr. 7, 2012), (noting prediction that mobile commerce will grow at a compound annual rate of 39 percent through 2016 and advising that “[a]t this moment, every enterprise should be devising new strategies for reaching customers through mobile applications in addition to the web.”).
[24] Benny Evangelista, Feds Crackdown on 70 Fake Websites that “Copy Cat” Real Sites, SF Gate (July 12, 2012),
[25] See Athleta, Inc. v. Pitbull Clothing Co., Inc., No. CV 12-10499-CAS FMOX, 2013 WL 142877 at *3 (C.D. Cal. Jan. 7, 2013).
[26] See id. at *3.
[27] See id. at *10.
[28] Id.
[29] Blue Nile, Inc. v., Inc., 478 F. Supp. 2d 1240, 1247 (W.D. Wash. 2007).
[30] Id.
[31] Id. at 1245.
[32] Id. at 1248.
[33] Blue Nile did succeed in asserting a copyright claim four years later, however, when another competitor copied images from the website and used them on its own website without permission. See Blue Nile, Inc. v. Ideal Diamond Solutions, Inc., No. C10-380Z, 2011 WL 3360664, at *3 (W.D. Wash. Aug. 3, 2011).
[34] See Conference Archives, Inc. v. Sound Images, Inc., No. Civ. 3:2006-76, 2010 WL 1626072, at *2 (W.D. Pa. Mar. 31, 2010).
[35] Id.
[36] Id. at *1.
[37] Id. at *9.
[38] Complaint at 3, Conferences Archives, Inc. v. Sound Images, Inc., Civil No. 3:2007-76, 2010 WL 1626072 (W.D. Pa. Mar. 21, 2006), 2006 WL 6495532.
[39] Conference Archives, Inc., 2010 WL 1626072, at*10.
[40] See id. at *10-11.
[41] See also Int’l Union of Operating Engineers, Local 150, AFL-CIO v. Team 150 Party, Inc., No. 07C3972, 2008 WL 4211561 (N.D. Ill. Sept. 5, 2008).
[42] Salt Optics, Inc. v. Jand, Inc., No. SACV 10-0828 DOC, 2010 WL 4961702 at *11 (C.D. Cal. Nov. 19, 2010).
[43] Id.
[44] See, e.g., Franklin & Henry, supra note 6.
[45] See, e.g., 1st Lake Mobile Website, Copyright TXu001885424 (2013-07-16).
[46] An author can register computer code with the Copyright Office under class TX.
[47] As an initial matter, only the author of a copyrightable work may sue for copyright infringement, unless the work is made “for hire.” Most corporations hire a website developer to create an online presence and the developer is the default author for copyright purposes. In order to sue for copyright infringement, the contract with the developer must include a specific work for hire provision. Without such a provision, there may be no basis for claiming rights to the website in the first place. See Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730, 738 (1989) (stating that the work does not satisfy the copyright requirements for a work for hire because there was no written agreement). These provisions are relatively common, but not universal.
[48] See Kelly v. Arriba Soft Corp., 336 F.3d 811, 817 (9th Cir. 2003) (Stating that “the plaintiff must show ownership of the copyright and copying by the defendant”); see also LGS Architects, Inc. v. Concordia Homes of Nevada, 434 F.3d 1150, 1156 (9th Cir. 2006); 17 U.S.C. § 411 (2008) (“no civil action for infringement of the copyright in any United States work shall be instituted until preregistration or registration of the copyright claim has been made in accordance with this title”).
[49] See, e.g., La Resolana Architects, PA v. Reno, Inc., 555 F.3d 1171, 1178 (10th Cir. 2009).
[50] Boisson v. Banian, Ltd., 273 F.3d 262, 272 (2d Cir. 2001) (quoting Folio Impressions, Inc. v. Byer Cal., 937 F.2d 759, 763 (2d Cir. 1991)).
[51] See Crown Awards, Inc. v. Trophydepot, No. 2:03-CV-02448-DRH, 2003 WL 22208409, at *13 (E.D.N.Y. Sept. 3, 2003).
[52] See Complaint, Facebook, Inc. v. Studivz Ltd, supra note 11, at 7.
[53] See 17 U.S.C.§ 411(a).
[54] See Athleta, Inc. v. Pitbull Clothing Co., Inc., No. CV 12-10499-CAS FMOX, 2013 WL 142877, at n.6 (C.D. Cal. Jan. 7, 2013).
[55] Id.
[56] United States Copyright Office, Copyright Registration for Online Works (2002), available at
[57] Id. at 1.
[58] Darden v. Peters, 402 F. Supp. 2d 638, 643 (E.D.N.C. 2005) (quoting United States Copyright Office, Copyright Registration for Online Works 3 (2009)).
[59] The problems of requiring fixedness in online text are also apparent in the debate over fluid trademarks, an emerging area of concern. See, e.g., Johanna Pyhälä & Erkki Holmila, Are Fluid Trademarks Your Cup of Tea?, World Trademark Review, (last accessed Feb. 23, 2014).
[60] United States Copyright Office, supra note 57, at 2.
[61] A work need not be entirely fixed to receive copyright protection, however. Video games, for example, satisfy copyright’s fixation requirement by having effects that can be “’reproduced, or otherwise communicated’ for more than a transitory period.” Williams Elecs., Inc. v. Artic Int’l, Inc., 685 F.2d 870, 874 (3d Cir. 1982) (citing Stern Elecs., Inc. v. Kaufman, 669 F.2d 852, 855-56 (2d Cir. 1982)).
[62] Darden v. Peters, 402 F. Supp. 2d 638, 643 (E.D.N.C. 2005).
[63] See id. at 487.
[64] See Salt Optics, Inc. v. Jand, Inc., No. SACV 10-828, 2010 WL 4961702 (C.D. Cal. Mar. 4, 2011).
[65] See id. at *1-2.
[66] Id. at *12-13.
[67] Id. at 13.
[68] Id.
[69] See id.
[70] Allen v. Ghoulish Gallery, No. 06CV371 NLS, 2007 WL 4207923, at *1 (S.D. Cal. Nov. 20, 2007).
[71] Id.
[72] Id. at *4. The Court did not find that there was copyright infringement, however, because there was no substantial similarity between the two websites. Id. at *6.
[73] See Kantemirov v. Goldine, No. C05-01362 HRL, 2005 WL 1593533 (N.D. Cal. June 29, 2005) (“[g]iven the flexible definition of works falling with the scope of the Copyright Act design and layout within the general subject matter of the Copyright Act”); Order,, Inc. v. Wall St. Interactive Media Corp., No. 98 CIV. 6974 (LAP), 1998 WL 34194887 (S.D.N.Y. Nov. 9, 1998) (enjoining alleged infringer from copying design or layout of website).
[74] Infantry Div., World War II Living History Ass’n v. Oprendek, No. Civ.A.11–165, 2011 U.S. Dist. LEXIS 140014 (D. N.J. Aug. 4, 2011).
[75] Id. at *3-4.
[76] See Id. at *21-22.
[77] Franklin & Henry, supra note 6.
[78] Express Lien v. National Association of Credit Management Inc., Civil Action No. 13–3323, 2013 WL 4517944 (E.D. La. August 23, 2013).
[79] Id. at *1.
[80] Id. at *4.
[81] See Complaint, Facebook, Inc. v. Studivz Ltd., supra note 11, at 75-119.
[82] 15 U.S.C. § 1125 (2012).
[83] See Apple Inc. v. Samsung Elecs. Co., 888 F. Supp. 2d 976 (N.D. Cal. 2012).
[84] In fact, the terms “storefront” and “web storefront” are becoming synonymous with e-commerce retail websites.
[85] Parker Waichman LLP v. Gilman Law LLP, No. 12-CV-4784 JS AKT, 2013 WL 3863928, at *4 (E.D.N.Y. July 24, 2013).
[86] See Sleep Sci. Partners v. Lieberman, No. 09-04200 CW, 2010 WL 1881770 (N.D. Cal. May 10, 2010).
[87] Id. at *2.
[88] Id. at *6.
[89] Id. at *9.
[90] See id. at *6.
[91] Id. at *3.
[92] Id. at *8-9.
[93] 15 U.S.C. §§ 1125(a)(3), (c)(2)(B).
[94] Wal-Mart Stores, Inc. v. Samara Bros., 529 U.S. 205, 210 (2000).
[95] See id.
[96] See id. at 215.
[97] See, e.g., id. at 212-13.
[98] 505 U.S. 763, 768 (1992).
[99] Two Pesos, 505 U.S. at 768..
[100] See Wal-Mart Stores, Inc. v. Samara Bros., 529 U.S. 205, 215 (2000).
[101] See id. at 211; see also Qualitex Co. v. Jacobson Prods. Co., 514 U.S. 159, 163 (1995).
[102] See, e.g., SG Servs. Inc. v. God’s Girls Inc., No. CV 06-989 AHM, 2007 WL 2315437, at *9 (C.D. Cal. May 9, 2007); cf. Card Tech Int’l, LLLP v. Provenzano, No. CV 11-2434 DSF PLAX, 2012 WL 2135357, at *22 (C.D. Cal. June 7, 2012) (suggesting that either standard may apply to web sites).
[103] First Brands Corp. v. Fred Meyer, Inc., 809 F.2d 1378, 1383 (9th Cir. 1987).
[104] See Wal-Mart Stores, 529 U.S. at 214.
[105] See, e.g., Ex parte Tayama, 24 U.S.P.Q.2D (BNA) 1614, 1614 (B.P.A.I. 1992) (rejecting design patent for computer icon because it is not a design for an article of manufacture).
[106] Rachel v. Banana Republic Inc., 831 F.2d 1503, 1506 (9th Cir. 1987) (citation omitted).
[107] See Mortgage Mkt. Guide, LLC v. Freedman Report, LLC, No. CIV.A. 06-CV-140-FLW, 2008 WL 2991570, at *1 (D. N.J. July 28, 2008).
[108] See SG Servs. Inc. v. God’s Girls Inc., No. CV 06-989 AHM, 2007 WL 2315437 (C.D. Cal. May 9, 2007).
[109] See id. at *9. Even so, the court ruled in favor of the defendant on the trade dress infringement claim as a whole because the plaintiff failed to prove either distinctiveness or likelihood of confusion. In addition, plaintiff’s sole evidence in support of its infringement claim was a declaration that did not meet the federal evidentiary standards, as it was not signed under penalty of perjury, and was therefore inadmissible. See id. at *34.
[110] See Card Tech Int’l, LLLP v. Provenzano, No. CV 11-2434 DSF PLAX, 2012 WL 2135357 (C.D. Cal. June 7, 2012).
[111] Id. at *6.
[112] Id.
[113] See No. 3:2006-76, 2010 WL 1626072, at *1 (W.D. Pa. Mar. 31, 2010).
[114] Id. at *58-59 (citations omitted).
[115] AMF Inc. v. Sleekcraft Boats, 599 F.2d 341, 349 (9th Cir. 1979).
[116] 367 F. Supp. 2d 1238, 1240 (D. Minn. 2005).
[117] Id. at 1244.
[118] Id. at 1245.
[119] Conference Archives, Inc. v. Sound Images, Inc., No. 3:2006-76, 2010 WL 1626072, at *12 (W.D. Pa. Mar. 31, 2010)
[120] 1 Melvin B. Nimmer & David Nimmer, Nimmer on Copyright, § 1.01[D][2] (2013).
[121] 478 F. Supp. 2d 1240, 1243 (W.D. Wash. 2007).
[122] Id.
[123] Id.
[124] Id. at 1248.
[125] See id at 1245.
[126] Salt Optics, Inc. v. Jand, Inc., No. SACV 10-0828, at 4 (C.D. Cal. Mar. 4, 2011), available at; Salt Optics, supra note 40, at *8.
[127] See id.
[128] Salt Optics, Inc. v. Jand, Inc., No. SACV 10-0828, at 9.
[129] Sleep Sci. Partners v. Lieberman, No. 09-04200 CW, 2010 WL 1881770, at *5 (N.D. Cal. May 10, 2010).
[130] See 15 U.S.C. § 1125(d)(1) (2008).
[131] See 18 U.S.C. § 1030 (2008) (criminalizing fraud and related activity in connection with computers); see also 18 U.S.C. § 2701 (proposed legislation criminalizing unlawful access to stored communications).
[132] See Polaroid Corp. v. Polarad Elecs. Corp., 287 F.2d 492 (2d Cir. 1961); see also Conference Archives, Inc. v. Sound Images, Inc., No. 3:2006-76, 2010 WL 1626072, at *18 (W.D. Pa. Mar. 31, 2010).
[133] See, e.g., Sleep Sci. Partners, 2010 WL 1881770, at *7-9.
[134] See Salt Optics, Inc. v. Jand, Inc., No. SACV 10-0828, 2010 WL 4961702, at *5 (C.D. Cal. Nov. 19, 2010).
[135] Id.
[136] Mot. of Def. Jand, Inc. to Dismiss Pl.’s Second Am. Compl. for Failure to State a Claim, Salt Optics, Inc. v. Jand, Inc., No. SACV 10-0828, 2011 WL 1158953 (C.D. Cal. Jan. 3, 2011).
[138] Times New Roman, for example, originated as a proprietary font designed in 1931 for the British newspaper The Times. See A Brief History of Times New Roman, Practical Typography, (last visited Feb. 23, 2014).
[139] See, e.g., Stefan Thiemert et al., A Digital Watermark for Vector-Based Fonts, Proceedings of the 8th Workshop on Multimedia & Security 120 (2006),
[140] Conference Archives, Inc. v. Sound Images, Inc., No. 3:2006-76, 2010 WL 1626072, at *4 (W.D. Pa. Mar. 31, 2010).
[141] Id. at *5.
[142] Traditional consumer surveys, such as those endorsed in Eveready, would be useful here. See Union Carbide Corp. v. Ever-Ready, Inc., 531 F.2d 366, 385-88 (7th Cir. 1976); see, e.g., 65 J. Thomas McCarthy, Trademarks and Unfair Competition § 32:175 (4th ed. 2007). In designing those surveys, however, comparison on similar devices is critical because of the potential effect of adaptive site design differences on user experience.
[143] See, e.g., Dianne Cyr et al., Colour Appeal in Website Design In and Across Cultures: A Multi-method Evaluation, 68 Int. J. Human-Computer Stud. 1-21 (2010); Christian Vasile, How Colors Help Make Websites Successful, 1st Web Designer, (last visited Feb. 23, 2014); Vishal Verma, The Importance of Color Theory, (Oct. 29, 2007),
[144] Athleta, Inc. v. Pitbull Clothing Co., Inc., No. CV 12-10499-CAS FMOX, 2013 WL 142877, at *2, *7 (C.D. Cal. Jan. 7, 2013).
[145] Id. at *7.
[146] Conference Archives, Inc. v. Sound Images, Inc., No. 3:2006-76, 2010 WL 1626072, at *19 (W.D. Pa. Mar. 31, 2010).
[147] See, e.g., Frank Guo, More Than Usability: The Four Elements of User Experience, Part I, UX Matters (Apr. 24, 2012),
[148] Whether the court ought to protect a website because it is more usable than another, however, is a difficult question. One could argue that usability is akin to functionality, in that enforcing protection of these qualities might have the unwanted effect of stifling competition and/or undercutting the overall goal of consumer protection.
[149] See, e.g., Sally Beauty Co. v. Beautyco, Inc., 304 F.3d 964, 973 (“Proof that a defendant chose a mark with the intent of copying the plaintiff’s mark may, standing alone, justify an inference of likelihood of confusion.”).
[150] See, e.g., Blake Tierney, Missing the Mark: The Misplaced Reliance on Intent in Modern Trademark Law, 19 Intell. Prop. L.J. 229 (2011); Soonbok Lee, Induced Infringement as a Strict Liability Claim: Abolishment of the Specific Intent Requirement, 4 Hastings Sci. & Tech. L.J. 381 (2012).
[151] In many cases, the website owner will bring claims of unfair competition alongside the infringement claims. See, e.g., Allen v. Ghoulish Gallery, No. 06CV371 NLS, 2007 WL 4207923 (S.D. Cal. Nov. 20, 2007); Card Tech Int’l, LLLP v. Provenzano, No. CV 11-2434 DSF PLAX, 2012 WL 2135357 (C.D. Cal. June 7, 2012). Whether the court’s consideration of intent, or scienter, in connection with these unfair competition claims is sufficient to accomplish the deterrent effect suggested here is subject to further debate.
[152] See Hughes & Rosen, supra note 13, at 44.
[153] See, e.g.,, Inc. v., Inc., 679 F. Supp. 2d 312, 315 (E.D.N.Y. 2010) (plaintiff alleged trademark infringement, inter alia, when defendant used plaintiff’s trademarks as meta-tags for its own site).
[154] See Conference Archives, Inc. v. Sound Images, Inc., No. 3:2006-76, 2010 WL 1626072, at *19 (W.D. Pa. Mar. 31, 2010) (describing plaintiff’s exhibit comparing to website in great technical detail and characterizing it as “very instructive to the Court’s analysis.”).
[155] The sign was to read, in one-inch block letters: “Notice: TACO CABANA originated a restaurant concept which Two Pesos was found to have unfairly copied. A court order requires us to display this sign to inform our customers of this fact to eliminate the likelihood of confusion between our restaurant and those of TACO CABANA.” Hughes & Rosen, supra note 13, at 45.
[156] See Garden of Life, Inc. v. Letzer, 318 F. Supp. 2d 946, 968 (C.D. Cal. 2004).
[157] Hughes & Rosen, supra note 13, at 44.
[158] See generally, George A. Agoston, Color Theory and Its Application in Art and Design (Dr. Jay M. Enoch et al., eds., 1979); Wyszecki, Color In Business, Science, and Industry (Stanley S. Ballard ed., 3d ed. 1975); Bleicher, Contemporary Color Theory & Use (2005).
[159] Athleta, Inc. v. Pitbull Clothing Co., Inc., No. CV 12-10499-CAS FMOX, 2013 WL 142877, at *12 (C.D. Cal. Jan. 7, 2013). As of Apr. 27, 2014, was no longer a functioning website.
[160] See Conference Archives, Inc. v. Sound Images, Inc., No. 3:2006-76, 2010 WL 1626072 (W.D. Pa. Mar. 31, 2010).
[161] See id. at *14.
[162] See id. at *15.
[163] Lesportsac, Inc. v. Kathy Clapp, No. D2012-2506, 2013 WL 1146701, at *6 (2013).
[164] Id. at *2.
[165] See id. at *6.
[166] See, e.g., Sleep Sci. Partners v. Lieberman, No. 09-04200 CW, 2010 WL 1881770, at *1 (N.D. Cal. May 10, 2010).