Lost in Translation: Repairing Rosetta Stone v. Google’s Indecipherable Functionality Holding

Washington and Lee Law Review, Aug 2018

A. J. Frey

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://scholarlycommons.law.wlu.edu/cgi/viewcontent.cgi?article=3333&context=wlulr

Lost in Translation: Repairing Rosetta Stone v. Google’s Indecipherable Functionality Holding

Lost in Translation: Repairing Rosetta Stone v. Google's Indecipherable Functionality Holding A. J. Frey 0 1 0 Thi s Note is brought to you for free and open access by the Law School Journals at Washington & Lee University School of Law Scholarly Commons. It has been accepted for inclusion in Washington and Lee Law Review by an authorized administrator of Washington & Lee University School of Law Scholarly Commons. For more information , please contact , USA 1 A. J. Frey, Lost in Translation: Repairing Rosetta Stone v. Google's Indecipherable Functionality Holding 68 Wash. & Lee L. Rev. 1505 (2011), http://scholarlycommons.law.wlu.edu/wlulr/vol68/iss3/23 - Lost in Translation: Repairing Rosetta Stone v. Google’s Indecipherable Functionality Holding A.J. Frey∗ I. Introduction ................................................................................1506 ∗ Candidate for J.D., Washington and Lee University School of Law, May 2012. I would like to thank Professor Sarah Wiant, without whom this Note would not have been published; my family for being my tireless cheerleaders; and my wife Chloe for her unyielding love and support. I. Introduction In the last year, two new developments—one legal and one technological—have converged, placing the legal system at a critical juncture as it strives towards a workable set of rules for understanding and regulating Internet search engine technology. The August 2010 ruling by the District Court for the Eastern District of Virginia in Rosetta Stone v. Google1 and the October 2010 release of Google Inc.’s "Google Goggles"2 technology for the Apple iPhone3 (which greatly expanded consumer access to Google Goggles)4 together represent an impending paradigm shift in both the technical means of Internet searching and the legal response to evolving technology. The District Court’s ruling in Rosetta Stone, while correctly decided and forward-thinking, contains fatal flaws in its analysis of trademark law’s "functionality doctrine"; meanwhile, the increasing ubiquity of Google’s Google Goggles technology is challenging the boundaries of 2. See infra Part II.C (explaining in more detail the Google Goggles technology, which is a software program for smartphones that allows a user to take pictures of virtually anything, using his or her mobile device’s camera, and use those digital images as search inputs, in lieu of keywords, to run Google searches). 3. See Frank Reed, Google Goggles for iPhone Released, MARKETING PILGRIM (Oct. 6, 2010), http://www.marketingpilgrim.com/2010/10/google-goggles-for-iphone-released. html (last visited Oct. 10, 2011) (announcing the release of Google Goggles for the Apple iPhone) (on file with the Washington and Lee Law Review). 4. See MG Siegler, Search by Sight with Google Goggles, TECHCRUNCH (Dec. 7, 2009), http://techcrunch.com/2009/12/07/google-goggles (last visited Oct. 10, 2011) (announcing the availability of Google Goggles for Android smart phones) (on file with the Washington and Lee Law Review). The release of Google Goggles for the iPhone followed the December 2009 release of a beta version of Google Goggles for smart phones running the Google Android operating system. The release of Google Goggles for the iPhone is of particular significance because it marks the point at which Google Goggles became freely available to the majority of the smartphone-using public. See Paul Miller, Canalys: Android Takes Q2 Smartphone Market Share Lead in US with 886 Percent Year-Over-Year Growth, ENGADGET (Aug. 2, 2010, 10:44 AM), http://www.engadget.com/2010/08/02/canalysandroid-takes-q2-smartphone-market-share-lead-in-us-wit (last visited Oct. 10, 2011) (reporting the combined U.S. market share of Apple mobile devices and Android mobile devices as 55.7% of the market) (on file with the Washington and Lee Law Review). what it means to interface with search engines and is a portent of greater advances to come. The convergence of these two events—the Rosetta Stone decision and the arrival of Google Goggles—provides a dramatic lesson in why courts must decide trademark disputes correctly if trademark law is to remain an effective body of law to police online commerce and why the intellectual property doctrine of functionality has no place in trademark infringement claims on the Internet. The argument presented in this Note is threefold. First, Rosetta Stone was rightly decided for the wrong reasons. The district court, struggling under the constraints of traditional trademark law doctrine to reach a result that would allow Google to use the Rosetta Stone marks without imposing legal liability, misinterpreted and misapplied the functionality doctrine to provide Google with a superfluous legal justification for its actions.5 Second, although the court in Rosetta Stone was well-meaning, its use of the functionality doctrine to validate Google’s sale of trademarks to thirdparty online advertisers will have unintended and damaging consequences for search engines, mark-holders, and the legal system as a whole. The emerging technology embodied in Google Goggles provides a perfect predictive lens through which to imagine these consequences. In short, new search technologies militate against using the functionality doctrine in trademark infringement claims against search engines and similar online entities. The district court’s decision to dismiss Rosetta Stone’s trademark infringement claims had sufficient footing on separate legal grounds without needing to resort to use of the functionality doctrine: The doctrine has no place in Internet-based trademark infringement claims. Third, the fact that Rosetta Stone’s claim could have been dismissed on other grounds alone does not negate the fact that courts struggle mightily with trademark infringement claims against search engines and are in need of a new legal doctrine in this context to allow fair and economically efficient use of trademarks by online entities. The Rosetta Stone court’s sweeping expansion of the functionality doctrine was a direct result of its inability to utilize the better-suited "nominative fair use" doctrine adopted by the Ninth Circuit, but thus far rejected by the Fourth Circuit.6 On appeal, the Fourth 5. See Amy E. Bivins, Google Prevails in AdWords Infringement Case Under Keyword ‘Functionality’ Doctrine, BNA: PATENT, TRADEMARK & COPYRIGHT LAW DAILY (Aug. 13, 2010), available at http://news.bna.com.ezproxy.wlu.edu/ptln/PTLNWB/split_ display.adp?fedfid=17619264&vname=ptcjnotallissues&fcn=16&wsn=497363500&fn=176 19264&split=0 ("[T]he functionality doctrine has not been raised frequently in keyword advertising litigation, and was a somewhat novel component of the court’s decision . . . ."). 6. See Brief of International Trademark Association as Amicus Curiae at 23, Rosetta Stone Ltd. v. Google, Inc., No. 10-2007 (4th Cir. Nov. 1, 2010) [hereinafter INTA Brief] Circuit should vacate the court’s decision and adopt the doctrine of nominative use. Part II of this Note explains the revolution in Internet searching precipitated by Google Goggles, its relation to Google’s revenue-generating AdWords technology,7 and the legal response that must necessarily be forthcoming in order to make efficient use of these technologies. Part III introduces the latest shot across Google’s bow in the battle over Google’s AdWords technology and its use of trademarks in that technology, Rosetta Stone v. Google. Part III’s discussion focuses primarily on the district court’s flawed application of trademark law’s functionality doctrine: It demonstrates that the functionality doctrine has no place in the determination of whether Internet search companies’ use of trademarks to generate advertising revenue is legal or fair to mark-holders. Part IV connects the court’s erroneous functionality analysis in Rosetta Stone to the new search technology encapsulated by Google Goggles and explains why the court’s reasoning is wholly inadequate to deal with this emerging technology and should be overturned on appeal. Part V of this Note proposes a way forward for the Fourth Circuit, and for courts generally, as they struggle to regulate Google and other Internet search companies’ use of trademarks. It concludes, first, that both total protection for mark-holders based on traditional trademark law and total indemnification from liability for search engines based on the functionality doctrine are inefficient and undesirable outcomes; and second, that the Rosetta Stone decision should be vacated and that the Fourth Circuit should adopt the nominative use doctrine. (urging the adoption of the nominative fair use doctrine by the Fourth Circuit). The International Trademark Association, in its amicus brief to the Fourth Circuit in Rosetta Stone v. Google argues that the "nominative fair use" doctrine gives the defendant the "right to use another’s trademark . . . as long as: (1) the product or service in question is one not readily identifiable without the use of the trademark, (2) the use of the mark . . . is reasonably necessary to identify the product or service, and (3) the user does not do anything that would, in conjunction with the mark, suggest sponsorship or endorsement by the trademark holder." Id. 7. See infra Part II.B (explaining in further detail the operation of AdWords, which is an online auction program that allows businesses to bid on and buy search terms, which Google users input when searching, in order to have the winning bidder’s advertisements displayed on the search results pages for certain terms). II. Google AdWords, Google Goggles, and the Search Revolution A. Google’s Position in the Market In the realm of Internet search technology, one name stands above all others in the popular consciousness: Google. Google, Inc. is a hugely influential company, the operations of which affect every consumer of Internet services on a daily basis and provide a plethora of services that are indispensible to the function of the Internet as we know it.8 Google’s breadand-butter service is, and always has been, Internet searching.9 Constantly pioneering new means of Internet searching, Google is both the dominant actor and the leading "norm creator"10 in the search technology sector. Using aggregating, indexing, and ranking technology superior to that of its competitors,11 Google has grown from a novel start-up to a household name in Internet services in just over a decade.12 Along the way, as Google has defended its various business practices in court, the legal system has struggled to keep up with Google’s ever-evolving, increasingly ubiquitous technology.13 Perhaps more than any other of its business practices, Google’s advertising programs—specifically its AdWords 8. See Inside the Googleplex, ECONOMIST, Aug. 30, 2007, at 56–58 (noting that many Internet-users "keep their photos, blogs, videos, calendars, e-mail, news feeds, maps, contacts, social networks, documents, spreadsheets, presentations, and credit-card information—in short, much of their lives—on Google’s computers"); Who’s Afraid of Google?, ECONOMIST, Aug. 30, 2007, at 9 (observing that, while other search engines do exist, "Google, through the sheer speed with which it accumulates the treasure of information, will be the one to test the limits of what society can tolerate"). 9. See KEN AULETTA, GOOGLED: THE END OF THE WORLD AS WE KNOW IT 7 (2009) (estimating the number of daily searches performed on Google each day to be over 3 billion). 10. Stacy L. Dogan & Mark A. Lemley, Trademarks and Consumer Search Costs on the Internet, 41 HOUS. L. REV. 777, 784 (2004) (coining the term to describe courts that deal with consumer confusion cases involving Internet search technology, noting that "the rules they develop will determine practices on the Web and whether the Internet realizes its potential as a vast clearinghouse of information and content"). 11. See AULETTA, supra note 9, at 6 (describing Google’s "secret sauce," a complex set of algorithms that "not only rank those [Internet] links that generate the most traffic, and are therefore presumed to be more reliable," but also "assign a higher qualitative ranking to more reliable sources—like for instance, a New York Times story"). 12. See Googleplex, supra note 8, at 56–58 (noting that Google’s "share price has quintupled since 2004 [the year of its IPO], making Google worth $160 billion" in 2007) . 13. See Greg Lastowka, Google’s Law, 73 BROOK. L. REV. 1327, 1328–29 (2008) ("While the law should be cautious in how it regulates new technologies such as Google, as Justice Cardozo once noted, major technological changes often call for the transformation of law."). program—have generated the most complex and novel legal challenges for the courts.14 Driven by trademark-holders who accuse Google of direct and contributory trademark infringement through its AdWords program, these legal actions have led to a patchwork of inconsistent judicial opinions interpreting federal trademark law and stretching the limits of traditional trademark doctrines.15 The underlying tension in these rulings is an economic balancing act requiring courts to weigh the commercial benefits of strong trademark protection for mark-holders on the one hand, against the social and economic value of consumer access to freely and quickly accessible online information provided by search engine companies on the other.16 Google’s track record in these trademark disputes has been mixed.17 As search engine technology has matured, the courts seem to have reached a 14. See, e.g., Rosetta Stone Ltd. v. Google Inc., 730 F. Supp. 2d 531, 545 (E.D. Va. 2010) (involving an action against Google for the sale of trademarked terms to third-party advertisers through Google’s AdWords advertising system); Rescuecom Corp. v. Google Inc., 562 F.3d 123, 124 (2nd Cir. 2009) (same); Google Inc. v. Am. Blind & Wallpaper Factory, Inc., No. C 03-5340 JF, 2007 U.S. Dist. LEXIS 32450, at *3–5 (N.D. Cal. Apr. 18, 2007) (same); Gov’t Employees Ins. Co. ("GEICO") v. Google, Inc., 330 F. Supp. 2d 700, 702 (E.D. Va. 2004) (same). 15. See Dogan & Lemley, supra note 10, at 802 ("[A] handful of courts have widened the net of trademark infringement to encompass search engines . . . and others who help competitors reach their audiences through nontraditional ‘uses’ of established marks."). 16. See id. at 786–92 (discussing the general economic rationale behind the law of trademarks, including the reduction of "search costs" and the protection of "goodwill"). 17. See cases cited supra note 14 (involving trademark infringement or dilution actions against Google for the sale of trademarked terms to third-party advertisers through Google’s AdWords advertising system). Rosetta Stone’s trademark infringement claims against Google were dismissed, and Rosetta Stone has appealed to the Fourth Circuit. See Don Jeffrey, Rosetta Stone Appeals Loss to Google in Trademark Case over Web Searches, BLOOMBERG (Sept. 1, 2010, 4:34 PM), http://www.bloomberg.com/news/2010-0901/rosetta-stone-appeals-loss-to-google-in-trademark-case-over-web-searches.html (last visited Oct. 10, 2011) ("Rosetta Stone Inc. appealed a judge’s ruling that Google Inc. didn’t infringe its trademarks by selling the language-software maker’s marks to rivals for use as search keywords.") (on file with the Washington and Lee Law Review). Rescuecom’s infringement claim against Google was dismissed by the trial court; however, the Second Circuit reversed the trial court, finding that Google’s AdWords practices constituted "use in commerce." Rescuecom Corp. v. Google Inc., 562 F.3d 123, 124 (2009). Google settled its litigation with American Blinds in what was widely seen as a victory for Google. See Linda Rosencrance, American Blinds Drops Trademark Infringement Suit Against Google, COMPUTERWORLD (Sept. 4, 2007, 12:00 PM), http://www.computerworld.com/s/article/ 9034322/American_Blinds_drops_trademark_infringement_suit_against_Google_?intsrc=ne ws_ts_head (last visited Oct. 10, 2011) (noting that American Blinds dropped the suit without Google having "to pay any money to settle the dispute") (on file with the Washington and Lee Law Review). The court in GEICO denied Google’s motion to dismiss the claims against it, but ultimately found for Google "on the issue of infringement, finding loose consensus that Google’s sale of trademarked terms to third parties does fall within the Lanham Act’s definition of "use[] in commerce,"18 potentially subjecting Google to liability for trademark infringement claims.19 Additionally, courts have expanded the traditional reach of the trademark infringement cause of action by developing the doctrine of "initial interest confusion,"20 which can function as an almost separate cause of action in Internet cases21 and has the potential to further mire Google and other Internet search providers in trademark infringement litigation.22 The legal theories underlying trademark infringement actions against Google are discussed in more detail in Parts III and V. B. Google AdWords Foremost among the legal challenges to Google’s search business have been actions based in trademark law that attack Google’s primary source of that GEICO failed to demonstrate a likelihood of consumer confusion." Lastowka, supra note 13, at 1387. 18. 15 U.S.C. § 1125(a)(1) (2000). The Lanham Act (15 U.S.C. §§ 1051–1127) sets out the federal cause of action for trademark infringement. One element of an infringement claim is proof that a defendant "uses" a mark "in commerce," "on or in connection with any goods or services." Id. "Use in commerce" is defined as "the bona fide use of a mark in the ordinary course of trade, and not merely to reserve a right in a mark." 15 U.S.C. § 1127 (2000). 19. See, e.g., GEICO, 330 F. Supp. 2d at 704 (denying Google’s motion to dismiss; finding that plaintiff GEICO adequately alleged trademark use). Along with the trademark infringement cause of action, courts have also used the "trademark dilution" cause of action as a means of protecting trademark owners, but the focus of this Note will be solely on trademark infringement. For a thorough explanation of the developing doctrine of trademark dilution, including the 2006 passage of the Trademark Dilution Revision Act, see Lastowka, supra note 13, at 1367–70; see also Trademark Dilution Revision Act, 15 U.S.C. § 1125(c)(2) (2000) (establishing two types of federal dilution harm, "dilution by blurring" and "dilution by tarnishment"); Lastowka, supra note 13, at 1367 ("Dilution’s controversial innovation is that it protects marks without the need for plaintiffs to demonstrate consumer confusion."). 20. See 4 MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION § 23:6 (4th ed.) ("Infringement can be based upon confusion that creates initial customer interest, even though no actual sale is finally completed as a result of the confusion."). 21. See Eric Goldman, Deregulating Relevancy in Internet Trademark Law, 54 EMORY L.J. 507, 564 (2005) ("Courts have also been unclear whether IIC [initial interest confusion] supports a determination of a ‘likelihood of consumer confusion,’ or if IIC creates a different measurement for infringement, such as a ‘likelihood of initial interest confusion.’"). 22. See id. at 559 ("With its doctrinal flexibility, [initial interest confusion] has become the tool of choice for plaintiffs to shut down junior users who have not actually engaged in misappropriative uses."). revenue, online advertising.23 Google’s substantial advertising revenue24 is generated primarily through its AdWords technology.25 AdWords is an online auction system that allows third-party advertisers to purchase from Google, as "triggers," individual words or phrases that Internet users might input when using Google to search.26 AdWords constitutes just one of the many mechanisms Google uses to generate advertising revenue through commercial use of its search engine, although it is by far the most prominent.27 The AdWords system works by displaying sponsored search results separately from "organic" search results on the search results webpage.28 When a Google user searches for a trigger term, the organic search results generated by Google’s algorithms are accompanied by "sponsored links" (advertisements) to the right and top of the results page.29 These sponsored 23. See cases cited supra note 14 (citing cases involving trademark infringement or dilution actions against Google for the sale of trademarked terms to third-party advertisers through Google’s AdWords advertising system). 24. See AULETTA, supra note 9, at xi ("Google’s advertising revenues—more than twenty billion dollars a year—account for 40 percent of all the advertising dollars spent online."); Google Inc., Annual Report (Form 10K) at 41 (Feb. 12, 2010), available at http://www.sec.gov/Archives/edgar/data/1288776/000119312510030774/d10k.htm (showing Google’s revenue from advertising to be 97% of its total revenue in 2009) . 25. See Duncan Ross, Beginner’s Guide to Google AdWords, SYDNEY MORNING HERALD, Apr. 28, 2009, available at http://www.smh.com.au/news/technology/biztech/ beginners-guide-to-google-adwords/20090615-c9n1.html (providing, in layman’s terms, a brief overview of Google’s AdWords technology aimed at small business-owners who might be interested in utilizing AdWords to advertise their businesses). 26. See Google, How Search Ads Works, YOUTUBE (Jan. 18, 2010), http://www.youtube.com/watch?v=ka4tCkYXHiE (last visited Oct. 10, 2011) (explaining the function of AdWords). For a more detailed explanation of the way in which AdWords technology operates and the arguments for and against Google’s use of AdWords, see Ashley Tan, Note, Google AdWords: Trademark Infringer or Trade Liberalizer?, 16 MICH. TELECOMM. & TECH. L. REV. 473, 474–80 (2010). 27. See Lastowka, supra note 13, at 1348 (discussing a second prominent advertising method for Google, its "AdSense" technology). AdSense is a "program whereby website owners are paid by Google to provide advertising space on their websites where AdWords advertisements are displayed." Id. at 1349. In this regard, AdSense is effectively an outgrowth of Google’s AdWords program. 28. See How Search Ads Work, supra note 26 (noting that Google’s organic search results are always displayed on the left-hand side of the results page, while advertisements are displayed, if at all, on the top and right-hand side of the page). 29. See id. (explaining Google’s commitment to transparency and ease of use on its search results page and noting that it is in the company’s self-interest to allow users to easily distinguish between organic and sponsored results). Google is extremely diligent and inflexible about the on-screen graphical boundaries between organic and sponsored search results: The company maintains as part of its stated mission the goal of making clear to users which results are organic and which are sponsored. But cf. Lastowka, supra note 13, at links are displayed because advertisers have purchased specific search terms from Google, through a competitive bidding system, which are then tethered to the advertisers’ sponsored results.30 Trigger words can be terms in common usage, like "fly fishing," or they might be trademarked terms, like "Orvis."31 Thus, a competitor company to Orvis may purchase the term "Orvis" as a Google AdWord so that when a user searches for "Orvis," the competitor’s advertisement will appear on the right-hand side of the search results page, allowing users to click on it and view the competitor’s website.32 The ability of third-party advertisers to buy terms with trademark meanings did not always exist. In 2004, Google made a conscious decision to stop restricting the sale of trademarked terms to only the holders of those marks.33 Before its initial public offering in 2004, Google announced that it would accept "bids for terms that corresponded with the names of brands" from any bidder, not just the holder of the brand’s trademark.34 Shortly after Google announced its decision, the first of many trademark infringement and dilution suits against the company was brought by the insurance carrier GEICO for selling its brand name, "Geico," as an AdWords trigger term.35 Since 2004, Google has faced several more legal challenges to its sale of terms with trademark meanings to advertisers other than the mark-holders, the most recent of which is Rosetta Stone v. Google, discussed at length in Part III.36 1345 ("The left/right distinction is very important to Google, but studies have shown it is not important to the average user."). 30. See Lastowka, supra note 13, at 1339–40 (detailing the operation and origin of Google’s bidding system and noting that it was a business model originally developed by one of Google’s competitors, GoTo.com, which Google later acquired). 31. See id. at 1343–44 (using another example, the search term "Nike," to illustrate the distinction between sponsored and organic results and providing screenshots of sample Google results pages to clarify the explanation). 32. See id. at 1344 ("In the right column, various AdWords advertisements for the search term ‘Nike’ are listed. These include advertisers that sell Nike footwear as well as other brands of sneakers."). 33. See Stefanie Olsen, Google Plans Trademark Gambit, CNET NEWS (Apr. 13, 2004), http://news.cnet.com/2100-1038_3-5190324.html (last visited Oct. 10, 2011) ("Google plans to stop limiting sales of trademarks in its popular keyword advertising program, a high-stakes gamble that could boost revenue but also create new legal problems for the company.") (on file with the Washington and Lee Law Review). 34. Lastowka, supra note 13, at 1360. 35. See GEICO, 330 F. Supp. 2d at 701 (involving an action by an insurance company against Google for the sale of a trademarked term, in this case "Geico," to third-party advertisers through Google’s AdWords advertising program). 36. See Stephanie Yu Lim, Comment, Can Google Be Liable For Trademark Infringement? A Look at the "Trademark Use" Requirement as Applied to Google AdWords, C. Google Goggles With AdWords as its core revenue-generating engine, Google has continued to expand the channels by which consumers of Internet search services can access its search engine and thereby be exposed to advertising content.37 Not traditionally a company to rest on its laurels, Google has recently thrown the world of Internet search technology into flux once again with a potentially paradigm-shifting expansion of its search technology.38 Historically, the dominant, indeed the sole, way to input information into an Internet search engine has been through either text or voice input (which is then converted to text).39 Until quite recently, in order to search for something using a conventional search engine, an Internet user would have needed to input the desired search term in textual form—for instance the word "Pepsi"—into the search engine’s text bar, and then the search engine would return a list of relevant results.40 Google’s latest search technology, Google Goggles, has expanded the way Internet users can search by eliminating the need for textual input.41 Now, using the Google Goggles software, those with smartphones and other mobile devices can search based directly on images—pictures taken with a device’s camera—with the images themselves functioning as the inputs into the search engine.42 The technology 14 UCLA ENT. L. REV. 265, 270–81 (2007) (providing a comprehensive survey of previous legal challenges to Google’s practice of selling trademarked terms as AdWords triggers). 37. See Harry McCracken, The Undiscovered Google: 7 Services You Need to Try, FOX NEWS (Mar. 9, 2010), http://www.foxnews.com/scitech/2010/03/09/undiscoveredgoogle-services-need-try (last visited Oct. 10, 2011) (listing some of the lesser known and emerging services provided by Google, including Google Fast Flip, Google Product Search, Google Squared, and Google Voice) (on file with the Washington and Lee Law Review). 38. See Claudine Beaumont, Google Launches Google Goggles Visual Search, TELEGRAPH (Dec. 7, 2009, 6:30 PM), http://www.telegraph.co.uk/technology/google /6752725/Google-launches-Google-Goggles-visual-search.html (last visited Oct. 10, 2011) (announcing the worldwide launch of Google’s "Google Goggles" visual search technology) (on file with the Washington and Lee Law Review). 39. See Goldman, supra note 21, at 511 (explaining that, conventionally, "[s]earch engines [would] allow a searcher to enter a keyword into a search box and make a query against a database of content"). works by breaking down images into "object-based signatures."43 Google’s search engine "then compares those signatures against every item it can find in its image database" and returns relevant webpage results to the user, just as if the search had been done using conventional text input.44 In other words, instead of typing the word "Pepsi" into Google’s search bar, an individual with a smartphone can simply take a picture of the Pepsi logo on a can of Pepsi, and Google will return search results based on that logo.45 While still in its technological infancy, Google Goggles and its successors have the potential to expand drastically the available potential inputs for Internet search engines. No longer are users confined to textual keywords for their searches; now, images, and in time sounds, music,46 video content, and even tastes and smells will be available channels for generating search engine results.47 Of course, with this sweeping expansion of search 43. JR Raphael, A Hands-On Tours: Google Goggles Visual Search, PC WORLD (Dec. 7, 2009), http://www.pcworld.com/article/183933/a_handson_tour_google_goggles_ visual_search.html (last visited Oct. 10, 2011) (on file with the Washington and Lee Law Review). 44. Id. 45. See id. (noting that "Google Goggles doesn’t work well with food, cars, plants, or animals," but that "[d]evelopers say the app will soon be able to recognize plants by their leaves, even suggest chess moves by ‘seeing’ an image of your current board"). The Google Goggles software currently works best with business cards, logos, book covers, wine labels, pieces of art, and the like. Id. The ultimate goal for the program is for Goggles to have the capability to recognize any image with the aid of Google’s enormous database of digital images. Id. 46. See Carrie-Ann Skinner, Shazam Launches BlackBerry Music Recognition App, PC ADVISOR (Apr. 9, 2009), http://www.pcadvisor.co.uk/news/index.cfm?NewsID=113645 (last visited Oct. 10, 2011) (announcing the launch of music recognition software for the BlackBerry mobile device) (on file with the Washington and Lee Law Review). Software that allows users to identify songs by holding their computers or mobile devices up to a source of music have existed for a number of years, and while Google has not been at the forefront of this particular technology, there is every reason to believe that a marriage of song-recognition software and search engine technology is in the near future. See Fabrizio Pilato, AT&T Wireless Launches First Music Recognition Software in the U.S., MOBILE MAG. (Apr. 15, 2004, 1:01 AM), http://www.mobilemag.com/2004/04/15/att-wirelesslaunches-first-music-recognition-service-in-the-us (last visited Oct. 10, 2011) (announcing one of the first music recognition software offerings by a U.S. cellular phone company) (on file with the Washington and Lee Law Review). 47. See, e.g., Nick Bilton, Bits: Video Search Company Gets Investor Attention, N.Y. TIMES (Dec. 21, 2010, 11:00 AM), http://bits.blogs.nytimes.com/2010/12/21/video-searchcompany-gets-investor-attention (last visited Oct. 10, 2011) ("3VR Security, based in San Francisco, Calif., says it has solved that problem with technology that automatically catalogs video footage so that it can be searched at a later date, just like looking up a query on Google.") (on file with the Washington and Lee Law Review); Stephen Williams, Gadgetwise: Discovered While Nosing Around CES . . . ., N.Y. TIMES (Jan. 14, 2011, 7:14 AM), http://gadgetwise.blogs.nytimes. com/2011 /01/14/discovered-while-nosing-around-c-emethods comes a tempest of legal challenges for courts to contend with. What rights, if any, will trademark and copyright holders have to keep Google from using their corporate logos or brand images as visual triggers for a new, evolved AdWords (or perhaps "AdPics") program? Should trademark law restrict Google or others from allowing their consumers to use sights, sounds, and words as inputs into its search engine? Does public policy weigh in favor of giving search engines unfettered use of protected content in the name of generating relevant search results? Most importantly, is the body of trademark law equipped to decide these questions? Although the rapid evolution of search engine technology leaves many legal problems to be resolved, they are largely beyond the scope of this Note.48 The latest legal challenge to Google’s AdWords program does, however, offer a vision of what future legal responses to this changing technology may look like if courts insist on misapplying and contorting existing intellectual property law doctrines to fit the novel facts presented by Internet cases of this type. After presenting the facts and legal analysis of Rosetta Stone v. Google in Part III of this Note, Part IV’s discussion centers around the dangers to both the body of trademark law and the right of free access to relevant online information caused by further flawed legal reasoning of the type present in Rosetta Stone. Part V proposes a solution that keeps Google’s challenged practices within traditional trademark law analysis but ensures continued broad access to the wealth of information Google’s technology provides. III. Rosetta Stone v. Google A. Facts of the Case: Google’s AdWords; Rosetta Stone’s Marks Rosetta Stone is a Virginia corporation that creates "technology-based language learning products and services"49 and is internationally known as s/?scp=1&sq=CES%20smell&st=cse (last visited Oct. 10, 2011) (discussing an emerging technology that coordinates the release of specific smells into a consumer’s home based on digital signals embedded in television and video programming) (on file with the Washington and Lee Law Review). 48. See Goldman, supra note 21, at 552 (noting that "[f]ederal trademark law applies to Internet conduct in three principal ways: trademark infringement, trademark dilution, and anti-cybersquatting"). As Professor Goldman points out, "all three laws have important implications for Internet search." Id. However, the focus of this Note is chiefly on trademark infringement. 49. Rosetta Stone Ltd. v. Google Inc., 730 F. Supp. 2d 531, 535 (E.D. Va. 2010). one of the premier makers of language learning software in the world.50 After learning that Google had begun selling keywords that Rosetta Stone had registered as trademarks51 to third-party advertisers through the AdWords program, the company brought claims of trademark infringement and dilution against Google in the District Court for the Eastern District of Virginia.52 The claims were made on the grounds that Google was "actively assisting . . . advertisers to mislead consumers and misappropriate Rosetta Stone’s trademarks by using the trademarks (1) as keyword triggers for paid advertisements and (2) within the title and text of paid advertisements on Google’s website."53 Rosetta Stone alleged that, by giving third-party advertisers the right to use the Rosetta Stone Marks or words, phrases, or terms similar to those Marks as keyword triggers that cause Sponsored Links to be displayed, Google [was] helping these advertisers misdirect web users to websites of companies that (i) compete with Rosetta Stone, (ii) sell language education programs from Rosetta Stone’s competitors, (iii) sell counterfeit Rosetta Stone products, or (iv) are entirely unrelated to language education.54 Both parties moved for summary judgment, and the court ruled in Google’s favor on all relevant trademark issues, finding that Google was "not directly, vicariously, or contributorily liable for its sale of trademarks as advertising keywords."55 Rosetta Stone has appealed the ruling to the 50. See Alex Madrigal, My First Night with Rosetta Stone, THE ATLANTIC (Jan. 12, 2011, 11:01 AM), http://www.theatlantic. com/technology/archive/2011 /01/my-first-nightwith-rosetta-stone/69384 (last visited Oct. 10, 2011) (noting that Rosetta Stone is "by far the most heavily marketed and well-known" language software) (on file with the Washington and Lee Law Review); Ernest Scheyder, Rosetta Stone Suit vs. Google Dismissed, REUTERS (Apr. 29, 2010, 6:09 PM), http://www.reuters.com/article/idUSTRE63S5DD20100429 (last visited Oct. 10, 2010) ("Rosetta sells software to help users learn one of 31 languages, including Farsi, French and Irish.") (on file with the Washington and Lee Law Review). 54. Id. at 539. 55. Bivins, supra note 5. The court held that: (1) Google’s practice of auctioning Fourth Circuit Court of Appeals, where it is currently pending.56 The district court’s decision in favor of Google has been controversial and has drawn both criticism and praise from interested parties,57 many of whom have joined the action as amici, filing numerous briefs with the appeals court.58 The critiques of the district court’s decision are too numerous to fit within the scope of this Note,59 but the one which has drawn perhaps the most critical attention is the district court’s ruling that Google’s use of the Rosetta Stone marks was protected by trademark law’s functionality doctrine.60 B. The Court’s Decision Before delving into the district court’s functionality analysis, it is important to frame it within the court’s broader opinion. The district marks as keyword triggers for links sponsored by third-party advertisers did not constitute direct trademark infringement; (2) functionality doctrine protected Google’s use of the marks; (3) Google was not liable for contributory trademark infringement; (4) Google was not liable for vicarious trademark infringement; and (5) Google could not be held liable for trademark dilution. Rosetta Stone, 730 F. Supp. 2d at 534. The court granted Google’s motion for summary judgment on the trademark infringement claim because "no reasonable trier of fact could find that Google’s practice of auctioning Rosetta Stone’s trademarks as keyword triggers to third-party advertisers creates a likelihood of confusion as to the source or origin of Rosetta Stone’s products." Id. at 540. See generally Rosetta Stone Ltd. v. Google Inc., No. 10-2007 (4th Cir. filed Sept. 57. Sheri Qualters, Showdown over Web Searches, NAT’L L.J. (Dec. 20, 2010), http://www.law.com/jsp/nlj/PubArticleNLJ.jsp?id=1202476363617 (last visited Oct. 10, 2011) (noting that various interested parties have filed briefs challenging the district court’s analysis of "likelihood of confusion," "trademark dilution," "functionality doctrine," and "contributory infringement," among other issues) (on file with the Washington and Lee Law Review). 58. See id. (reporting that "[m]ore than three dozen technology and consumer products companies and trade groups" have joined as amici, including "Coach Inc., Ford Motor Co. and Tiffany & Co and technology companies such as TiVo Inc. and Viacom Inc."). 59. See id. (enumerating the errors of the district court’s ruling). Criticisms of the district court’s ruling include claims that it decided the "functionality doctrine" issue wrongly, that it failed to find Google liable for "contributory trademark infringement," that it failed to apply the doctrine of "nominative fair use," and that it misapplied the nine-factor test for analyzing "consumers’ likelihood of confusion" by "ignoring six factors" and "erroneously" analyzing the other three. Id. 60. See Bivins, supra note 5 (noting that the International Trademark Association called the district court’s analysis a "‘radical interpretation of the functionality doctrine’ to . . . focus on whether Google’s use of the Rosetta Stone mark served a functional role in Google’s business"). court’s holding—that Google’s use of the Rosetta Stone marks is protected by virtue of the functionality doctrine—came in the final section of its direct trademark infringement analysis, after its conclusion that Google’s use of the mark was noninfringing on other grounds.61 Mapping the traditional cause of action for direct trademark infringement,62 the court found the parties’ sole dispute to be over the existence of "consumer confusion" related to Google’s use of the marks.63 Furthermore, of the traditional "nine factors" used to judge consumer confusion, "only three of the nine confusion factors [were] in dispute."64 Examining the issues of "(1) defendant’s intent; (2) actual [consumer] confusion; and (3) the consuming public’s sophistication,"65 one by one, the court found that "Google’s use of the Rosetta Stone marks [did] not amount to direct trademark infringement."66 In particular, on the issue of Google’s intent, the court found that "Google [was] not attempting to pass off its goods or services as Rosetta Stone’s."67 Rebuffing one of Rosetta Stone’s main contentions, the court 61. See Rosetta Stone, 730 F. Supp. 2d at 545 ("Notwithstanding a favorable finding for Google under the relevant infringement elements, the functionality doctrine protects Google’s use of the Rosetta Stone Marks as keyword triggers."). 62. See 15 U.S.C. § 1125 (2000) (setting forth the federal cause of action for trademark infringement). Under the Lanham Act, a cause of action for trademark infringement requires a plaintiff to prove five elements: "that (1) it possesses a mark; (2) defendant used the mark; (3) defendant’s use of the mark occurred in commerce; (4) defendant used the mark in connection with the sale, offering for sale, distribution, or advertising of goods or service; and (5) defendant used the mark in a manner likely to confuse consumers as to the source or origin of goods or service." Rosetta Stone, 730 F. Supp. 2d at 540 (citing People for the Ethical Treatment of Animals v. Doughney, 263 F.3d 359, 364 (4th Cir. 2001)). 63. See Rosetta Stone, 730 F. Supp. 2d at 540. As the court pointed out, the traditional measure of "consumer confusion" within a trademark infringement claim consists of a "nine factor" test. The factors can include: (1) strength or distinctiveness of the mark; (2) similarity of the two marks to consumers; (3) similarity of the goods and services the marks identify; (4) similarity between the facilities used by the mark-holders; (5) similarity of advertising used by the mark-holders; (6) defendant’s intent; (7) actual confusion; (8) quality of defendant’s product; (9) sophistication of the consuming public. See id. Professor Goldman notes that "[e]ach federal circuit has developed its own set of factors to consider" for determining the likelihood of consumer confusion. Goldman, supra note 21, at 552–53. Because the number of factors varies from circuit to circuit, the consumer confusion test is commonly referred to as "the multifactor likelihood of consumer confusion test" or "MFLOCC test." Id. For a summary of the common factors included in the MFLOCC test from circuit to circuit, see 4 MCCARTHY, supra note 20, § 24:29. 64. Rosetta Stone, 730 F. Supp. 2d at 540–41. 65. Id. at 541. 66. Id. at 545. 67. Id. at 541. stated that "evidence of [Google’s] financial gain alone is insufficient evidence of intent" to "trade on the Rosetta Stone Marks."68 Addressing the issue of actual consumer confusion, the court made two findings that derailed Rosetta Stone’s argument: First, the court discounted Rosetta Stone’s reliance on prior case law,69 which it found distinguishable because "Rosetta Stone and Google are not direct competitors in the languagelearning software market."70 Second, the court concluded that survey evidence submitted by Rosetta Stone purporting to demonstrate actual consumer confusion was "de minimis."71 Finally, on the issue of consumer sophistication, the court found the relevant segment of consumers to be "not the public at-large, but only potential buyers of [Rosetta Stone’s] products," whose sophistication the court concluded to be higher than average based on the nature and price of Rosetta Stone’s product.72 The court’s finding that the "expertise and sophistication" of Rosetta Stone’s potential customers tended "to demonstrate that they are able to distinguish between the Sponsored Links and organic results displayed on Google’s results page" was the final nail in the coffin of Rosetta Stone’s consumer confusion case.73 Given the thorough and decidedly unequivocal treatment the court gave to Rosetta Stone’s direct trademark infringement case in the first half of its opinion, it is all the more surprising that Judge Lee felt the need to venture into the dim and shadowy realm of the functionality doctrine—an area usually reserved for trade dress analysis—to press his point. And yet, this is exactly what he did. After forcefully concluding that Rosetta Stone had no claim against Google for infringement based on consumer confusion, the court further indemnified Google against Rosetta Stone’s attacks by holding that Google’s use of the Rosetta Stone marks was functional, and thus noninfringing.74 68. Id. The court made explicit an argument that Google has long advanced: "In fact, it is in Google’s own business interest, as a search engine, not to confuse its users by preventing counterfeiters from taking advantage of its service. Google’s success depends on its users finding relevant responses to their inquiries." Id. at 542. 69. See Sara Lee Corp. v. Kayser-Roth Corp., 81 F.3d 455, 459 (4th Cir. 1996) (resolving a protracted trademark contest between the holders of the marks "L’eggs" and "Leg Looks"). 70. Rosetta Stone, 730 F. Supp. 2d at 543. 71. Id. 72. Id. at 544. 73. Id. at 545. 74. See id. ("Notwithstanding a favorable finding for Google under the relevant infringement elements, the functionality doctrine protects Google’s use of the Rosetta Stone C. Examining the Application of the Functionality Doctrine to Word Marks In supporting its application of the functionality doctrine to word marks,75 the court in Rosetta Stone relied primarily on two cases for precedent.76 The first was Sega Enterprises v. Accolade,77 in which "the Ninth Circuit held the use of an initialization sequence that caused plaintiff’s trademark to be displayed on the screen was . . . a functional Marks as keyword triggers."). 75. See Stoller v. Sutech U.S.A., Inc., Opp’n No. 91117894, 2005 TTAB LEXIS 464, at *7 (Oct. 26, 2005) (referring to trademarks that are words or phrases "shown in standard character form"). The term "word mark" is used merely to identify a trademark that is solely textual. For instance, the word "Pepsi" is a word mark: It is a word that has a trademark meaning. The iconic blue and red Pepsi logo found on every can of Pepsi brand soda is a trademark but not a word mark. To confuse the issue further, a term such as "Rosetta Stone" is a word mark because it has a trademark meaning. However, unlike "Pepsi," the term "Rosetta Stone" has nontrademark meaning as well—for instance, it may also refer to the physical, historical artifact. See 1 MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION § 3:2 (4th ed.) ("[T]he meaning of a word or symbol is not necessarily fixed for all time as it is first used, or as it is defined in the dictionary, but may grow and develop new meaning and nuances according to its use."). 76. See Rosetta Stone Ltd. v. Google Inc., 730 F. Supp. 2d 531, 546 (E.D. Va. 2010) ("The keywords, therefore, have an essential indexing function because they enable Google to readily identify in its databases relevant information in response to a web user’s query."); see also Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1531 (9th Cir. 1992) (finding that use of the plaintiff’s trademark initialization sequence to achieve compatibility was functional because interoperability could not be achieved without the trademark sequence); Compaq Computer Corp. v. Procom Tech., Inc., 908 F. Supp. 1409, 1423 (S.D. Tex. 1995) (finding that the word "Compaq" inserted in computer code for purposes of compatibility was functional). 77. See Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1531 (9th Cir. 1992) (holding that defendant’s use of plaintiff’s trademarked "initialization sequence" in its video game cartridges was a functional use). In Sega, the court dealt with the issue of whether Sega, a videogame cartridge and console manufacturer, could "enjoin competing cartridge manufacturers from gaining access to its [gaming consoles]" through the use of an "initialization code" that was necessary for all cartridges to function in Sega consoles, but that also caused Sega’s trademark to briefly appear onscreen when a game started. Id. at 1514. The court held that "when there is no other method of access to the [console] that is known or readily available to rival cartridge manufacturers, the use of the initialization code by a rival does not violate the [Lanham] Act even though that use triggers a misleading trademark display." Id. The U.S. District Court for the Northern District of California had granted Sega a preliminary injunction against Accolade to compel it to stop selling games that used Sega’s initialization code. Id. The Court of Appeals reversed. Id. In overturning the district court, the Ninth Circuit noted that the burden of "proving nonfunctionality [was] on Sega," and that Sega failed to "produce sufficient evidence regarding the existence of a feasible alternative" to Accolade’s use of its initialization code. Id. at 1532. As a result, the court concluded that the district court erred in granting Sega "preliminary injunctive relief under the Lanham Act," but the court left the door open to the possibility that Sega might be able to "meet its burden of proof at [a future] trial." Id. display of the trademark because using the initialization sequence was the only feasible means for providing compatibility with plaintiff’s product."78 The second case was Compaq Computer Corp. v. Procom Technology,79 in which the District Court for the Southern District of Texas "merely held that a specific use of a word mark was functional because it was the only commercially viable way for the defendant to make its product compatible with the plaintiff’s computer program."80 In making its comparison to Compaq and Sega, the Rosetta Stone court made two assertions. First, it stated that Google’s use of trademarked terms as AdWords triggers was "no different than the use of a Google search query to trigger organic search results relevant to the user’s search."81 In doing so, the court declined to treat Google’s use of trademarks as triggers for organic search results and its use of trademarks as AdWords triggers as separate functions.82 The second step in the court’s logic was to conclude that "keywords"—both in the context of organic and sponsored results— 78. Autodesk, Inc. v. Dassault Systèmes Solidworks Corp., 685 F. Supp. 2d 1001, 1009 (N.D. Cal. 2009). 79. See Compaq Computer Corp. v. Procom Tech., Inc., 908 F. Supp. 1409, 1423 (S.D. Tex. 1995) (holding that Procom Technology’s use of manufacturer Compaq’s trademarked "vendor identification" to enable Procom’s hardware products to interface with Compaq’s did not violate the Lanham Act because Procom’s use of the trademark was "purely functional"). The Compaq court considered whether Procom’s inclusion of competitor Compaq’s "vendor ID"—a code which "allows computers to identify themselves electronically to each other"—in its server hard drives’ software constituted trademark infringement under the Lanham Act on the grounds that Compaq’s vendor ID contained within it the word "Compaq." Id. at 1416. Procom used the Compaq vendor ID in its hard drives so that they would be compatible with Compaq servers and would be recognized by Compaq’s server maintenance software, which informed customers when server hard drives were worn out and in need of replacement. Id. at 1415–16. The court noted that Compaq’s placement of its trademark, "Compaq," in its vendor ID was a voluntary decision and that Compaq should have recognized the potential for this action to "exclud[e] competitors from the market" or else force them to defend their use of the vendor ID in court. Id. at 1423. The court found that, while Procom technically "used" the Compaq trademark in commerce, the use was a "purely functional" one. Id. Importantly, the court found that the Lanham Act does not protect functional product features and that "[t]his is true even when the functional feature is a trademark." Id. The court noted that the use of Compaq’s trademark in the vendor ID provided "compatibility" for Procom’s products, and that Compaq had not "established that there is any other commercially feasible method of achieving this desired" compatibility. Id. Ultimately, the court held that Procom was not liable for "trademark infringement or unfair competition," but found it liable for contributory copyright infringement based on other features of its products not discussed here. Id. at 1423–26. 80. 81. Autodesk, 685 F. Supp. 2d at 1009 (emphasis in original). Rosetta Stone, 730 F. Supp. 2d at 546. 82. See id. (stating that "[i]n both cases, a search term like ‘Rosetta Stone’ will return a string of Sponsored Links and organic links on Google’s search results page"). serve an "essential indexing function," within the framework of Google’s search technology.83 Implicit in the court’s analysis was the assumption that the sale of keywords to third parties through AdWords was the only commercially feasible way84 for Google to display "paid advertisements on its website."85 However, while relying on Sega and Compaq for support, the court did not explicitly claim that Google’s use of the protected marks was the only feasible way for Google to generate advertising revenue; instead, the court found that, were Google forbidden from using trademarks as triggers for AdWords, it might be forced to invent a more costly, less effective way of generating advertising revenue.86 This small difference between the Rosetta Stone court’s findings and the holdings of Sega and Compaq belies a fundamental misunderstanding of these two cases and establishes the foundation for the court’s erroneous conclusion that Google’s use of the Rosetta Stone marks was protected functional use. The great weight of trademark law’s doctrinal authority points to the idea that, in almost all instances, word marks—that is, words or phrases, "shown in standard character form,"87 that have been registered as trademarks—cannot be deemed "functional."88 Sega and Compaq, the only two cases the Rosetta Stone court cited for precedent in its functionality holding, comprise substantially the entire canon of case law supporting the idea that a word mark can be functional.89 Compaq and Sega constitute a 83. See id. ("The keywords, therefore, have an essential indexing function because they enable Google to readily identify in its databases relevant information in response to a web user’s query."). 84. See Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1531 (9th Cir. 1992) (determining that in order to find a product feature functional a court should consider the several factors, including the "availability of alternative designs; and whether a particular design results from a comparatively simple or cheap method of manufacture," meaning a "commercially feasible alternative" exists (citations omitted)). Rosetta Stone, 730 F. Supp. 2d at 546. 86. See id. ("Google would be required to create an alternative system for displaying paid advertisements on its website—a system which is potentially more costly and less effective in generating relevant advertisements."). 87. Stoller v. Sutech U.S.A., Inc., Opp’n No. 91117894, 2005 TTAB LEXIS 464, at *7 (T.T.A.B. Oct. 26, 2005). 88. See, e.g., Playboy Enters., Inc. v. Netscape Commc’ns Corp., 354 F.3d 1020, 1030–31 (9th Cir. 2004) ("Nothing about the marks used to identify PEI’s products is a functional part of the design of those products."); Autodesk, Inc. v. Dassault Systèmes Solidworks Corp., 685 F. Supp. 2d 1001, 1008 (N.D. Cal. 2009) ("[I]t is not apparent how a word mark could be essential to the use or purpose of an article or affect its cost or quality."). 89. See Rosetta Stone Ltd. v. Google Inc., 730 F. Supp. 2d 531, 546 (E.D. Va. 2010) (citing Compaq and Sega as precedent for rule that functionality doctrine can apply to word alternatives analysis for aesthetic design features. It clearly leaves legal space for nonfunctional aesthetic design features.146 Scholarly opinion and court rulings have come to conflicting conclusions as to the true effect of the new aesthetic versus non-aesthetic feature distinction in functionality cases, even in the conventional area for the functionality defense—trade dress law.147 Some scholars have interpreted Traffix as a minor tweak to an already misunderstood body of jurisprudence, while others view Traffix as having obliterated what was coalescing into a clearly understood, albeit still developing doctrine.148 Without offering an opinion on the ultimate impact of the Traffix ruling on functionality analysis in trade dress cases, one can safely speculate that removal of the doctrine from the trade dress context into the relatively uncharted realm of Internet-based trademark infringement claims will raise many more questions than it resolves.149 The fact is that the functionality doctrine is in a state of turmoil, even within its historical "home base" of trade dress law.150 This uncertainty in the doctrine militates against appropriating it now, as the Rosetta Stone court did, to resolve the unsettled issues surrounding the rights of trademark-holders in infringement actions against search engine companies 146. Id. at 328–29. 147. See, e.g., Walker & Zanger, Inc. v. Paragon Indus., 465 F. Supp. 2d 956, 965 (N.D. Cal. 2006) opinion corrected and superseded by 549 F. Supp. 2d 1168 (N.D. Cal. 2007) (discussing the Ninth Circuit’s functionality jurisprudence); Eco Mfg. LLC v. Honeywell Int’l, Inc., 357 F.3d 649, 654–55 (7th Cir. 2003) ("Traffix rejected an equation of functionality with necessity; it is enough that the design be useful."); Eppendorf-NethelerHinz GmbH v. Ritter GmbH, 289 F.3d 351, 356 (5th Cir. 2002) ("Traffix supersedes the definition of functionality previously adopted by this court."); see also WILLIAM LANDES & RICHARD POSNER, THE ECONOMIC STRUCTURE OF INTELLECTUAL PROPERTY LAW 198 (2003) (discussing aesthetic functionality); 1 MCCARTHY, supra note 75, §§ 7:81, 7:82 (describing the aesthetic functionality distinction as "unwarranted and illogical"); STEPHEN F. MOHR & GLENN MITCHELL, FUNCTIONALITY OF TRADE DRESS: A REVIEW AND ANALYSIS OF U.S. CASE LAW 151 (3d ed. 1997) (noting that, before Traffix, most courts analyzed both aesthetic and utilitarian functionality in the same way). 148. See Thurmon, supra note 136, at 246 (noting that some courts "have read Traffix as radically changing the law" by "equat[ing] functionality with utility, a standard very different from the competition-based standard used prior to Traffix"). 149. See Qualters, supra note 57 (quoting intellectual property law practitioners and parties to Rosetta Stone’s appeal who expressed concern that the district court’s legal "errors threaten to affect how [Internet trademark infringement] cases are resolved" and noted that courts need to "address what the [legal] standards should be" in these types of cases). 150. See INTA Brief, supra note 6, at 9 (emphasizing that "the functionality doctrine is used to determine whether the plaintiff’s trade dress serves a useful or functional role" (emphasis added)); supra note 139 and accompanying text (discussing the uncertainty surrounding the functionality doctrine after the Traffix decision). and other online entities. With the functionality doctrine ill-suited to Internet-based infringement claims and ill-defined, even in its proper context, the courts must turn to a new doctrine to balance properly the rights of trademark-holders against the those of Internet search companies and consumers. V. A New Way Forward: Recasting Rosetta Stone v. Google A. Identifying an Alternative to Functionality This Note’s central premise is that Rosetta Stone was rightly decided using the wrong legal justification. As discussed in Part III, the court’s functionality analysis was superfluous when considered as part of its complete holding.151 Moreover, the analysis was flawed in its application of legal doctrine and carried with it destabilizing consequences for markholders and search engine operators alike.152 To explain fully why, in light of the preceding critique of the court’s method, its overall result was correct, one must first establish what the Rosetta Stone court was attempting to achieve with its ruling. The court’s ultimate disposition of the relevant infringement issues shows that Judge Lee was attempting to give Google, and other search engine operators in its position, a limited right to freely use word marks, like the marks owned by Rosetta Stone, without fear of legal liability.153 The court appeared to make an economic judgment that it is more efficient to let Google use such marks with relative freedom—because of the benefit to searchers from free, reliable organic search results—than it is to let mark-holders extend their trademark rights so far as to disallow search engine operators to use trademarks in a so-called "functional manner."154 Unfortunately, as discussed in Part IV, the court’s ruling missed its mark 151. Supra Part III.B. 152. Supra Part IV.A. 153. See INTA Brief, supra note 6, at 14 ("[T]he District Court appears to have been trying to recognize that some use of Rosetta Stone’s trademarks, as trademarks, might be appropriate both for Google and its advertisers . . . so that they could communicate they are resellers of Rosetta Stone’s products, or are making fair, comparative advertising claims . . . ."). 154. See Dogan & Lemley, supra note 10, at 811 ("Like functionality or genericness, the key is to balance the search costs that will be imposed if we forbid relevant advertising against the search costs that might result from permitting that advertising and potentially confusing consumers."). and went too far.155 By placing Google’s use of the Rosetta Stone marks under the umbrella of the functionality doctrine, the court in effect indemnified every search engine company from trademark infringement liability, provided the company can show that it uses the mark in a manner "essential" to its search process.156 A reworking of the district court’s conclusion is necessary to ensure that, in future trademark infringement actions against search engines, the rights of mark-holders, search companies, and consumers are equitably balanced.157 In order to bring about the intended, but unrealized, effect of Judge Lee’s ruling, the crucial first step is to accurately characterize the nature of Google’s specific "use" of trademarks in the situation presented by Rosetta Stone.158 Acknowledging a growing trend in the federal courts, it is assumed for the purposes of this argument that a search engine’s use of trademarks in search-driven keyword advertising constitutes "use in commerce" under the Lanham Act.159 Assuming the general use of trademarks in search engine advertising is "use in commerce," the issue becomes how to specifically characterize trademark use by search engines.160 The INTA Brief submitted to the Fourth Circuit in the Rosetta Stone appeal provides a starting point for arguing that Google and other search engines’ use of trademarks is best characterized as "nominative fair use."161 155. Supra Part IV.A. 156. See supra notes 129–35 and accompanying text (speculating that future search engine technology will render the Rosetta Stone court’s functionality analysis ineffective to protect mark-holders from search engines’ infringing uses of their trademarks). 157. See Dogan & Lemley, supra note 10, at 779–82 (noting that, especially in the Internet context, recent court decisions have favored the interests of mark-holders too heavily over those of consumers, "search engines, directories, and other parties that use marks as classification tools"). 158. See Rosetta Stone Ltd. v. Google Inc., 730 F. Supp. 2d 531, 539 (E.D. Va. 2010) ("Rosetta Stone . . . alleg[es] that Google, through its AdWords Program, is helping third parties to mislead consumers and misappropriate the Rosetta Stone Marks by using them as keyword triggers for Sponsored Links and using them within the text or title of the Sponsored Links."). 159. See Lastowka, supra note 13, at 1395 (noting that courts outside the Second Circuit have found arguments to the contrary "unpersuasive"). Contra Goldman, supra note 21, at 593–94 (proposing that "[s]earch providers could be immunized from [infringement] liability using a rigorous definition of trademark ‘use’ under the Lanham Act" and arguing that "[s]earch providers do not ‘use’ a trademark regardless of the editorial role" they play). 160. See Goldman, supra note 21, at 554–58 (explaining that most trademark uses by search engines are "referential" or "associative"). 161. INTA Brief, supra note 6, at 6. B. Nominative Fair Use: A Better Path to the Right Result The doctrine of nominative fair use, originally developed by the Ninth Circuit, protects an otherwise infringing party’s right to use another’s trademark, such as in comparative advertising, as long as: (1) the product or service in question is one not readily identifiable without the use of the trademark, (2) the use of the mark or marks is only so much as is reasonably necessary to identify the product or service, and (3) the user does not do anything that would, in conjunction with the mark, suggest sponsorship or endorsement by the trademark holder.162 As the Ninth Circuit explained in New Kids on the Block v. News America Publishing,163 "[a] trademark is a limited property right in a particular word, phrase, or symbol."164 Trademark-holders do not hold property rights in their marks "in gross"165—the right in a trademark is, by definition, limited to "[p]reventing producers from free-riding on their rival’s marks."166 The doctrine of nominative use comes into play "where [a] mark is used only ‘to describe the goods or services of [a] party, or their geographic origin.’"167 The doctrine’s purpose is to prevent trademarks from functioning as "exclusive rights" to a word or forbidding "a trademark registrant to appropriate a descriptive term for his exclusive use and so prevent others from accurately describing a characteristic of their goods."168 The nominative fair use doctrine was applied in New Kids on the Block to protect defendant newspaper companies’ use of the trademarked name "New Kids on the Block" to describe plaintiff mark-holders, a popular band, in fan surveys published by the newspaper companies.169 Fans were asked to vote for the band member they found "the sexiest" by calling the newspapers’ "900 area code number[s]" to cast their votes.170 The newspapers, which made a profit on each call to the 900 numbers, then printed the results of the telephone survey, which required repeatedly using the band’s trademarked name.171 The Ninth Circuit, affirming the district court’s ruling on the trademark infringement issues,172 found no evidence that the defendants did anything to suggest endorsement of the surveys by New Kids on the Block.173 Neither did the defendants, in the circuit court’s view, make any more reference to the plaintiffs’ trademarks than was "necessary to identify [plaintiffs] as the subject of the polls."174 The thrust of the Ninth Circuit’s argument with respect to the nominative fair use defense was that "a problem closely related to genericity and descriptiveness is presented when many goods and services are effectively identifiable only by their trademarks."175 The court expressed concern that, were a "trademark holder . . . allowed exclusive rights in such use, the language would be depleted in much the same way as if generic words were protectable."176 The Ninth Circuit’s solution to the problem was to endorse the doctrine of nominative fair use, ensuring that descriptive uses of trademarks would be granted broad protection in the future.177 Since the Ninth Circuit’s New Kids on the Block decision, the nominative fair use doctrine has been reaffirmed and refined in that circuit,178 explicitly adopted in the Third Circuit,179 adopted in part in the 170. Id. 171. See id. 172. See id. at 309 ("Summary judgment was proper as to the first seven causes of action because they all hinge on a theory of implied endorsement; there was none here as the uses in question were purely nominative."). 173. See id. at 308 ("While plaintiffs’ trademark certainly deserves protection against copycats and those who falsely claim that the New Kids have endorsed or sponsored them, such protection does not extend to rendering newspaper articles, conversations, polls and comparative advertising impossible."). 174. Id. 175. Id. at 306. 176. Id. 177. See id. at 308 (holding that descriptive use of a trademarked name is "nominative fair use" and thus noninfringing). 178. See Playboy Enters., Inc. v. Netscape Commc’ns Corp., 354 F.3d 1020, 1029 (9th Cir. 2004) (finding the New Kids test for nominative fair use to be controlling, but noting that "[w]e apply a slightly different test for confusion in the nominative use, as opposed to the fair use, context"). 179. See Century 21 Real Estate Corp. v. Lendingtree, Inc., 425 F.3d 211, 222 (3d Cir. 2005) ("Today we adopt a two-step approach in nominative fair use cases."). Fifth and Sixth Circuits,180 and adopted in principle, though not in name, in the Second Circuit.181 The holding of the Second Circuit in Tiffany v. eBay182 is of particular relevance for the issue of search engine advertising because it represents the most recent instance of an appellate court’s applying the nominative fair use doctrine (in principle at least) to an Internet-based defendant.183 The Second Circuit’s holding also represents an attractive option for the Fourth Circuit as it considers the Rosetta Stone appeal—the possibility of hewing to the spirit of the nominative use doctrine without explicitly following the Ninth Circuit’s holding in New Kids on the Block.184 In Tiffany, the Second Circuit considered whether defendant Internet auction company’s use of plaintiff jewelry manufacturer’s trademark in "advertisements on its website and in connection with sponsored links purchased from search engines" violated trademark laws.185 Plaintiff Tiffany and Co., a world-renowned jeweler, was concerned that eBay was encouraging the sale of counterfeit Tiffany products on its website through the advertisement of its members’ ongoing auctions of purportedly genuine Tiffany-brand merchandise.186 Tiffany sued eBay alleging "trademark infringement, trademark dilution, and false advertising," and the district court found "in favor of eBay on all claims."187 Restating the basic 180. See PACCAR Inc. v. TeleScan Technologies, L.L.C., 319 F.3d 243, 256 (6th Cir. 2003) (declining to follow the Ninth Circuit’s analysis directly and applying the conventional "likelihood of confusion" test instead); Pebble Beach Co. v. Tour 18 I Ltd., 155 F.3d 526, 546 n.13 (5th Cir. 1998) (adopting the Ninth Circuit’s nominative fair use test in part). 181. See Tiffany (NJ), Inc. v. eBay, Inc., 600 F.3d 93, 102–03 (2d Cir. 2010) ("We have recognized that a defendant may lawfully use a plaintiff’s trademark where doing so is necessary to describe plaintiff’s product and does not imply a false affiliation or endorsement by the plaintiff of the defendant."). 182. See id. (holding, based on nominative fair use-like analysis, that defendant’s use of plaintiff’s trademark in online advertising was a noninfringing use). 183. See id. at 96–97 ("eBay is the proprietor of www.ebay.com, an Internet-based marketplace that allows those who register with it to purchase goods from and sell goods to one another."). 184. See id. at 102 (noting that the Second Circuit has "referred to the [nominative fair use] doctrine, albeit without adopting or rejecting it" and concluding that the court "need not address the viability of the doctrine to resolve Tiffany’s claim" in favor of eBay). 186. See Tiffany, 600 F.3d at 95 ("Tiffany has instituted this action against eBay, asserting various causes of action . . . arising from eBay’s advertising and listing practices."). 187. Id. at 95–101. principles of the nominative fair use defense, the Second Circuit held that eBay’s use of the Tiffany marks was noninfringing use "so long as there is no likelihood of confusion about the source of [the] defendant’s product or the mark-holder’s sponsorship or affiliation."188 Without addressing "the viability of the doctrine to resolve Tiffany’s claim," the court concluded: (1) that "a defendant may lawfully use a plaintiff’s trademark where doing so is necessary to describe the plaintiff’s product and does not imply a false affiliation of endorsement by the plaintiff of the defendant"; (2) that "eBay used the [plaintiff’s] mark to describe accurately the genuine Tiffany goods offered for sale on its website"; and (3) that "none of eBay’s uses of the mark suggested that Tiffany affiliated itself with eBay or endorsed the sale of its products through eBay’s website."189 In affirming the lower court’s trademark infringement ruling in favor of eBay, the Second Circuit discussed at length the nature of the "use" eBay was making of Tiffany’s mark. eBay used the Tiffany mark in advertisements on its own site and on search engine sites, like Google and Yahoo!, to advertise the fact that eBay featured a number of merchants who traded in Tiffany merchandise.190 Tiffany’s main complaint about eBay’s use of the mark came down to the fact that eBay "knowingly" advertised members’ sales of counterfeit Tiffany products.191 Tiffany insisted that eBay was aware that a "significant portion" of the allegedly "Tiffany" sterling silver jewelry listed on its site was counterfeit and that eBay profited from every transaction completed through its website, including sales of counterfeit Tiffany goods.192 Therefore, Tiffany argued, eBay was using the Tiffany mark to mislead consumers into thinking the items for sale on eBay’s site were genuine, to encourage sales, and to generate more revenue for itself.193 188. Id. at 102 (citing Merck Co. v. Mediplan Health Consulting, Inc., 425 F. Supp. 2d 402, 413 (S.D.N.Y. 2006)). The court noted that "[t]he Third Circuit treats the doctrine as an affirmative defense, while the Ninth Circuit views the doctrine as a modification to the likelihood-of-confusion analysis of the plaintiff’s underlying infringement claim." Id. at 102 n.7 (citations omitted). 189. Id. at 102–04. 190. Id. at 100–01. 191. Id. at 98. 192. Id. 193. See id. ("Reducing or eliminating the sale of all second-hand Tiffany goods, including genuine Tiffany pieces, through eBay’s website would benefit Tiffany in at least one sense: It would diminish the competition in the market for genuine Tiffany merchandise."). The Second Circuit concluded that Tiffany’s trademark infringement claim against eBay suffered from a critical factual weakness: eBay had taken numerous precautions and countermeasures to discourage the sale of counterfeit Tiffany products on its website and actively policed its sellers, going so far as to ban specific sellers from the site if they were deemed to be trafficking in counterfeit merchandise.194 Moreover, neither the fact that eBay generated revenue from "all sales of goods on its site"195 nor the fact that eBay may have had a "generalized knowledge"196 that there were counterfeit Tiffany goods being sold on its site was evidence of eBay’s wrongdoing because eBay "took affirmative steps to identify and remove illegitimate Tiffany goods" from its site.197 In short, the court concluded that, on the issue of trademark infringement, Tiffany was going after the wrong party.198 eBay, though making use of the Tiffany mark to identify sellers of Tiffany goods on its site, was using the mark merely to refer to Tiffany goods by name. This allowed its users to locate quickly and efficiently the goods on eBay’s website.199 While counterfeiters could be held liable to Tiffany for trademark infringement, eBay could not because its use was nominative and fair, and thus protected.200 Together, New Kids on the Block, Tiffany, and other recently decided nominative fair use cases201 stand for the principle that the "use of a trademark term to refer to a product originating from the trademark owner 194. See id. ("Although eBay was generating revenue from all sales of goods on its site . . . the district court found eBay to have ‘an interest in eliminating counterfeit Tiffany merchandise from eBay . . . to preserve the reputation of its website as a safe place to do business.’"). 195. Id. at 98. 196. Id. at 107. 197. Id. at 103. 198. See id. ("We conclude that eBay’s use of Tiffany’s mark in the described manner did not constitute direct trademark infringement."). 199. See id. ("eBay used the mark to describe accurately the genuine Tiffany goods offered for sale on its website."). 200. See id. at 102 (affirming the district court’s holding that defendant’s use of plaintiff’s mark "was protected by the doctrine of ‘nominative fair use’"). 201. See 1 IAN C. BALLON, E-COMMERCE & INTERNET LAW § 6.14[3] (2010) ("While not all circuits have specifically considered nominative fair use, it is fair to assume that even in the absence of controlling case law all circuits would recognize the concept of nontrademark use."); see also Univ. Commc’ns Sys., Inc. v. Lycos, Inc., 478 F.3d 413, 424 (1st Cir. 2007) (recognizing the concept of non-trademark use); Century 21 Real Estate Corp. v. Lendingtree, Inc., 425 F.3d 211, 222 (3d Cir. 2005) (same); PACCAR Inc. v. TeleScan Tech., L.L.C., 319 F.3d 243, 256 (6th Cir. 2003) (same); Pebble Beach Co. v. Tour 18 I Ltd., 155 F.3d 526, 547 (5th Cir. 1998) (same). is not actionable."202 And Tiffany specifically concludes that trademark law, through the nominative use doctrine, "embraces the right of online services to engage in keyword advertising."203 In light of this burgeoning recognition of an online entity’s right to use trademarks in a nominative way, the facts of Rosetta Stone require fresh reconsideration to determine whether Google’s use of the Rosetta Stone marks is a nominative fair use. C. Google’s Use Is Nominative In order to demonstrate that Google’s use—and by extension all search engines’ use—of Rosetta Stone’s marks is nominative, Google’s actions must be weighed against the standard test for nominative fair use, which, for purposes of this argument, will be the standard articulated in New Kids on the Block and presented for the Fourth Circuit’s consideration in the INTA amicus brief.204 In order for Google’s use of the Rosetta Stone marks to be nominative fair use: (1) [T]he product or service in question [may] not [be] readily identifiable without the use of the trademark, (2) the use of the mark or marks [may be] only so much as is reasonably necessary to identify the product or service, and (3) the user [may] not do anything that would, in conjunction with the mark, suggest sponsorship or endorsement by the trademark holder.205 1. Identifiability Google’s "use" of a search term (or other input), trademark or otherwise, begins when a user enters it into Google’s search engine.206 At this point, Google’s search algorithms compare the search input against databases of possible Internet results, using complex mathematical formulas and data generated by billions of searches to return organic search results to 202. eBay Brief, supra note 185, at 8 (citations omitted). 203. Id. at 4. 204. See INTA Brief, supra note 6, at 15 (outlining the nominative fair use test as articulated by the Ninth Circuit in New Kids on the Block). 205. Id. 206. See THOMAS A. POWELL, WEB DESIGN: THE COMPLETE REFERENCE 276–77 (2d ed. 2002), available at http://www.webdesignref.com/chapters/09/ch9-01.htm#01 ("Depending on the search facility being used, the query formed by the user may vary greatly. A simple query might include only keywords, like ‘Robot Butler.’ More complex queries might include Boolean queries like ‘Robot AND Butler.’"). the user that have the highest likelihood of being relevant to his or her search query.207 In addition to organic search results, Google returns advertising in the form of sponsored links that correspond to the user’s search input and are thus deemed by Google more likely to be related to the user’s search needs or interests.208 In its search process, Google uses inputs from the "real world" to generate search results on the Internet: It references real-world names, brands, photos, logos—many of them trademarked or copyrighted—because its users input these marks into its search engine.209 Just like eBay, or the newspaper companies in New Kids on the Block, Google needs to be able to describe and compare various trademark and non-trademark inputs, given to it by users, in order to conduct its business.210 Google must be able to respond to users’ search inputs, many of which will carry trademark meaning.211 Focusing purely on its organic search results, Google’s use of marks is clearly nominative because Google operates as an online index of "real world" concepts.212 The search engine is using trademarked terms and comparing those terms against a database of possible relevant results.213 The products and services that users name with their inputs—be they Orvis, Pepsi, John Deere, or Tiffany—cannot be readily described without use of the marks.214 207. See id. at 277 ("[T]he search engine builds as big a haystack as possible, then tries to organize the haystack somehow, and finally lets the user try to find the proverbial needle in the resulting haystack of information by entering a query on a search page."). 208. See How Search Ads Work, supra note 26 (explaining the process by which "sponsored links" are generated and displayed on Google’s search results page). 209. See eBay Brief, supra note 185, at 2 (noting that search engines "give users the ability to locate and purchase an almost unlimited array of products and services and to access information on any conceivable topic by simply typing a search term—a keyword or keywords—and pushing a button"). 210. See Rosetta Stone Ltd. v. Google Inc., 730 F. Supp. 2d 531, 546 (E.D. Va. 2010) ("If Google is deprived of [the] use of the Rosetta Stone Marks, consumers would lose the ability to rapidly locate potentially relevant websites that promote genuine Rosetta Stone products at competitive prices."). 211. See eBay Brief, supra note 185, at 5 (explaining that "consumers searching for a specific company, good, service, or other piece of information without knowing the exact website address for that information will likely use a search engine to identify and locate it"). 212. See Lastowka, supra note 13, at 1400 ("Google’s index is dynamically created by its algorithms in response to user queries."). 213. See Rosetta Stone, 730 F. Supp. 2d at 546 (noting that Google’s organic use of keywords, including trademarks, "enable[s] Google to readily identify in its database relevant information in response to a web user’s query"). 214. See New Kids on the Block v. News Am. Publ’g, Inc., 971 F.2d 302, 306 (9th Cir. 1992) ("[O]ne might refer to ‘the two-time world champions’ or ‘the professional basketball Indeed, the Rosetta Stone court’s justification for its functionality holding expressly condones Google’s use of the Rosetta Stone marks on the grounds that Rosetta Stone and its products cannot be readily identified without the use of its marks. The court calls Google’s use of "trademarked keywords, including the Rosetta Stone Marks, to identify relevant Sponsored Links . . . no different than the use of a Google search query to trigger organic search results relevant to the user’s search."215 "[A]dvertisers," the court continues, "rely on keywords to place their products and services before interested consumers," and because consumers search using names in common usage, such as a brand’s trademarked name, Google must use certain trademarks as keywords in order for its search engine to operate correctly.216 Google’s sponsored results, driven by its AdWords program, are merely an alternative method for returning relevant search results to searchers.217 Whereas organic results are driven by algorithms, sponsored results are driven by the market.218 The highest bidders’ advertisements are the ones that will appear on any given search results page.219 In either case, however, a trademark used by a searcher to generate results is used to describe the thing itself.220 Its use to generate market-driven search results is no more or less a commercial use of the mark on Google’s part than it is when used to generate organic search results.221 Trademark law does not team from Chicago,’ but it’s far simpler (and more likely to be understood) to refer to the Chicago Bulls."). 215. Rosetta Stone, 730 F. Supp. 2d at 546. 216. Id. Contrary to the court’s conclusion, this state of affairs does not render Google’s use "functional." Id. 217. See Lastowka, supra note 13, at 1401–02 (using the analogy of white and yellow pages in a physical phonebook to describe Google’s left and right-hand search results columns). 218. See Tan, supra note 26, at 476–77 ("In effect, an account holder ‘bidding’ on a keyword is competing for it with all other account holders who wish to use that keyword— an account holder’s financial power thus plays a significant role in determining the degree of exposure his ad will receive."). 219. See Lastowka, supra note 13, at 1339 (noting that Google modeled its AdWords program on the technology developed by GoTo.com (which Google eventually acquired), which "ranked [search results] according to the amount of money each purchaser paid" for advertising). 220. See id. at 1362 (explaining that, outside of a word’s trademark use, the trademarkholder does not "own any interest in the word . . . in the abstract"). 221. See id. at 1345–46. Google’s left-hand column is, in fact, subject to market forces in ways that can make it similar to the right hand column. Businesses seeking consumer traffic give mark-holders the right to restrict the use of their marks to describe their products, and nothing about the descriptive use of such names in the search engine context changes that result.222 2. Reasonable Necessity The nominative use doctrine’s requirement that search engines use only so much of the mark as is reasonably necessary will put some limitation on search engine practices as search technology develops.223 This differentiates the nominative use defense from the functionality doctrine as applied by the Rosetta Stone court.224 Instead of giving search engine companies carte blanche to use any form of trademark they can turn into a search input, the nominative use doctrine puts the burden on search companies to demonstrate that there is not a more limited way to use the trademark to generate search results.225 In the Rosetta Stone case, the use Google was making out of Rosetta Stone marks was as limited as possible—sales of the marks as textual keywords—but future search engines will use marks in a more expansive way.226 For instance, a future court might determine that allowing searchers to use images of a unique product design or video of a compelling realize that both columns are simply lists of links. Being first in the left-hand column may provide more traffic to a site than paying for an AdWords advertisement. Id.; see also David Segal, The Dirty Little Secrets of Search, N.Y. TIMES, Feb. 13, 2011, at BU1 (reporting on the hugely profitable "search engine optimization" or "SEO" industry, which advises companies on ways to increase their websites’ rank in Google’s organic search results). 222. See New Kids on the Block v. News Am. Publ’g, Inc., 971 F.2d 302, 308 (9th Cir. 1992) ("[W]here the defendant uses a trademark to describe the plaintiff’s product, rather than its own, we hold that a commercial user is entitled to a nominative fair use defense . . . ."). 223. See id. (allowing a defendant to claim the nominative fair use defense when "only so much of the mark or marks may be used as is reasonably necessary to identify the product or service"). 224. See supra notes 124–35 and accompanying text (arguing that the Rosetta Stone court’s conception of the functionality doctrine gives search engines an unrestricted right to appropriate trademarks as search inputs). 225. See Tiffany (NJ), Inc. v. eBay, Inc., 600 F.3d 93, 102 (2d Cir. 2010) (treating the nominative fair use doctrine as an affirmative defense to direct trademark infringement, placing the burden on the defendant to show that its use of plaintiff’s mark was protected). 226. See supra Part II.C (predicting the use of images, sounds, video, and other sensory inputs as search triggers). advertisement as search inputs is more use than "reasonably necessary" to identify the trademark. This is not to say that generating organic search results from such inputs would be forbidden, only that there might be limitations imposed on which inputs could be sold as "triggers" for sponsored links.227 In this way, search engines’ search functions can remain effective, mark-holders can retain some control of their trademarks in the search engine context, and consumers can continue to rely on the availability of relevant and efficient search results, regardless of the trademark or non-trademark nature of their queries.228 Given the current limits on search technology, which require the vast majority of users to search using keywords exclusively, Google’s "use" of trademarks remains largely limited to what is reasonably necessary to identify the product or service queried—keywords.229 As discussed above, it is easy to view Google’s use of trademarks to generate organic results as a nominative use,230 and there is no reason to treat sponsored links differently. Google’s incidental generation of advertising revenue by way of the search services it provides is not, under the conventional nominative use doctrine, a factor weighing against its successful deployment of the nominative fair use defense.231 As the New Kids on the Block and Tiffany courts held, a trademark-holder does not have the right to direct the disposable income of all possible consumers to products or services licensed by the mark-holder alone.232 The fact that a nominative user generates a profit from the use of a trademark is not dispositive, or indeed even indicative, evidence of trademark infringement.233 Advertising, 227. See Lastowka, supra note 13, at 1409 ("Trademark law cannot describe an optimal index to the Internet, given the fact that much of the information users seek and the problems they encounter are not matters where trademark law has much application."). 228. See id. (noting that "allowing trademark law to dominate the indexical value of search results poses serious risks: [T]rademark meanings might usurp other understandings of terms"). 229. See New Kids on the Block v. News Am. Publ’g, Inc., 971 F.2d 302, 308 (9th Cir. 1992) (holding that defendants who use mark-holders’ trademarks "only to the extent necessary to identify them," as for instance "the subject of [a] poll[]," and do not use a markholder’s "distinctive logo or anything else that isn’t needed to make the announcements intelligible," meet the limited use requirement). 230. Supra notes 213–16 and accompanying text. 231. See New Kids, 917 F.2d at 309 ("Where, as here, the use does not imply sponsorship or endorsement, the fact that it is carried on for profit and in competition with the trademark holder’s business is beside the point." (citations omitted)). 232. See id. ("[T]rademark laws do not give [plaintiffs] the right to channel their fans’ enthusiasm (and dollars) only into items licensed or authorized by them."). 233. See Tiffany (NJ) Inc. v. eBay, Inc., 576 F. Supp. 2d 463, 496 (S.D.N.Y. 2008) through sponsored links, is both a source of revenue for Google and an alternative index of search results for searchers.234 So long as Google keeps its organic and sponsored search results clearly separated and is held accountable by consumers for its results-generating methods, there is no cognizable legal difference between Google’s use of trademarks in organic and sponsored results.235 3. No Indication of Endorsement or Sponsorship Google and other search engines that use keyword advertising are not engaging in practices that would suggest sponsorship or endorsement by mark-holders. The fact that a portion of all search results are "generated from the search engine’s algorithm" while the rest are "paid advertisements that are identified separately on the search results page[]" does not alter the nominative use analysis discussed above.236 Search advertisements are, by virtue of the AdWords program, placed "in front of consumers who identify themselves as interested in certain products or services offered by the advertisers’ companies."237 Additionally, mark-holders who bring infringement claims against search engines like Google for selling trademarked search terms to advertisers misunderstand both the motivations of the average Internet-searcher and the relationship between search engines and third-party advertisers.238 Individuals search for many reasons: If searchers input a query for "Starbucks," does that mean they are necessarily searching with the intent aff’d in part, rev’d in part Tiffany (NJ) Inc. v. eBay Inc., 600 F.3d 93 (2d Cir. 2010) ("[T]rademark owners cannot prevent others from making a descriptive use of their trademark. While a trademark conveys an exclusive right to the use of a mark in commerce . . . that right generally does not prevent one who trades a branded product from accurately describing it by its brand name . . . ."). 234. See Lastowka, supra note 13, at 1344–45 (explaining the distinction between algorithm-driven, left-hand search results and market-driven, right-hand search results). 235. See id. at 1409 (noting that Google and other search engines must be held accountable for the way in which search results are displayed and identified and recognizing that "[a]llowing Google to completely control the indexical function of its search results might lead to public harm"). 236. eBay Brief, supra note 185, at 5. 237. Id. 238. See Goldman, supra note 21, at 513 ("Search processes are complex and defy simplistic analysis. Some searches are very simple and directed; other searches are openended and meandering. Thus, any attempt to describe a universal search process is inherently misleading."). of purchasing Starbucks products as a direct result of their search? Oftentimes, the answer is no.239 To understand why this is so, it is necessary "to make a distinction between two different kinds of searches— informational and commercial."240 As its name suggests, in an informational search, a searcher is looking for information about a concept and does not necessarily seek to transact any consumer business.241 The strictures of trademark law should have no bearing on these searches because "consumer confusion" is an impossibility: The searcher is not a "consumer."242 This is not to say, however, that an informational search cannot regard a commercial product or involve a trademark. To borrow an example from Professor Goldman’s exhaustively thorough treatment of this phenomenon—which he labels "objective opaqueness"—imagine that a searcher enters the term, "Canon PowerShot S400" (the name of a high-end camera) into a search engine.243 The user, and in turn the search engine, use the trademarks "Canon" and "PowerShot" to effectuate the search; the search seems commercial, but perhaps the searcher is already the proud owner of a PowerShot S400 and is merely searching for instructions on home camera-repair techniques.244 On the other hand, the searcher may desire to buy a PowerShot S400, in which case the search could properly be labeled commercial and the searcher a potential consumer. Once again, though, the full picture is obscured. The searcher may be in the market for a used PowerShot from a third-party camera retailer or perhaps a leather case for his previously purchased PowerShot.245 In neither context does Canon, as the trademarkholder, have a right to restrict the use of its mark.246 With no way of 239. See id. at 521 ("Simply put, one cannot make any legally-supportable inferences about searcher objectives based on the keywords used."). 240. Segal, supra note 221, at BU1 (quoting Mark Stevens, the owner of a "black hat" search optimization company). 241. See Goldman, supra note 21, at 522–24 (listing a variety of possible informational search objectives, including for example "prepurchase information," "postpurchase information," "community information," "employment related information," "dictionary uses," and "typographical errors"). 242. Id. 243. See id. at 521 ("This Article refers to our inability to infer searcher’s objectives from their keywords as ‘objective opaqueness.’"). 244. See id. at 522 ( noting that "[t]he searcher may . . . be looking for postpurchase assistance, including customer support or repair or servicing information"). 245. See id. (detailing the many search objectives a searcher might have when searching for the name of a product aside from directly purchasing the product from the mark-holder). 246. See Lastowka, supra note 13, at 1362 (explaining that, outside of a word’s knowing searchers’ search objectives, the line between a commercial search and an informational one becomes blurred, and the result is a glut of timeconsuming factual survey evidence (of questionable utility) to prove likelihood of confusion247 in trademark infringement claims against search engines.248 The foregoing digression into the nature of searcher behavior shows that trademarks, as keywords, serve a search engine’s essential commercial and non-commercial functions alike. Trademarks, when used as keywords, are not necessarily used in their commercial, trademark context. As discussed above, searcher intent is virtually impossible to discern with precision, and the use of trademarks in AdWords advertising "comprehends a number of lawful, pro-competitive practices, such as the sale or resale of products bearing the trademarks; the sale of components, replacement parts, and compatible products; and the provision of non-competitive information about the goods or services corresponding to the trademarks."249 To deem use of these marks "inherently unlawful solely because third parties may make infringing uses of those trademarks in their advertisements" puts too much power in the hands of mark-holders and hurts consumers of Internet search services.250 Google’s use of the Rosetta Stone marks fits neatly into the nominative fair use exception developed by the Ninth Circuit in New Kids on the Block.251 More broadly though, Google’s use is representative of an increasingly ubiquitous kind of trademark use particular to the Internet.252 trademark use, the trademark-holder does not "own any interest in the word . . . in the abstract"). 247. 1 MCCARTHY, supra note 75, § 2:8 ("[T]he keystone . . . is the avoidance of a likelihood of confusion in the minds of the buying public. Whatever route one travels . . . the signs give direction to the same enquiry—whether defendant’s acts are likely to cause confusion."). 248. See Rosetta Stone Ltd. v. Google Inc., 730 F. Supp. 2d 531, 542–43 (E.D. Va. 2010) (noting that the parties agreed that evidence of "actual confusion" was the "best evidence of likelihood of confusion" and that "Rosetta Stone’s evidence of actual confusion" consisted of "testimonies of five individuals out of more than 100,000 impressions over six years"). 249. eBay Brief, supra note 185, at 6. 250. Id. at 4. 251. See New Kids on the Block v. News Am. Publ’g, Inc., 971 F.2d 302, 309 (9th Cir. 1992) (holding that a use that describes the trademark, is limited in scope, and does not imply sponsorship or endorsement by mark-holder is "nominative fair use" and noting that "the fact that it is carried on for profit and in competition with the trademark holder’s business is beside the point"). 252. Brief of Appellee at 21, Rosetta Stone Ltd. v. Google Inc., No. 10-2007 (4th Cir. Nov. 29, 2010) (arguing that "referential [trademark] uses are inherently protected" in the As technology evolves, Internet entities are claiming commercial roles that were once predominantly the domain of "brick and mortar" entities.253 The Rosetta Stone case demonstrates that courts need to carefully apply a mix of traditional and evolving trademark doctrines to keep the law effective and equitable when it comes to protecting the rights of mark-holders, search engine companies, and consumers.254 The nominative fair use doctrine achieves this balance. D. A Roadmap for the Fourth Circuit Adopting the nominative fair use doctrine will bring the Fourth Circuit to the forefront of Internet intellectual property law and show that the Circuit understands the needs of the diverse stakeholders in the web-based economy. While an explicit adoption of the Ninth Circuit’s New Kids on the Block standard is unnecessary to achieve what this Note contends is the correct result, the Fourth Circuit should, at the least, follow in the footsteps of the Second Circuit and acknowledge the validity of the principles underlying the nominative fair use doctrine.255 Like the Ninth Circuit, the Fourth Circuit has previously recognized as "important" the notion "that trademarks not be transformed from rights against unfair competition to rights to control language."256 If the Fourth Circuit reverses the district court’s trademark infringement holding in favor of Google, it risks "perpetuat[ing] [the] monopolies" that mark-holders continuously seek to gain on the use of their marks.257 Alternatively, if the Fourth Circuit upholds the district court’s ruling in its entirety, including its search engine context and that "there is no need to evaluate the traditional fair use factors or adopt a multi-factored test as in New Kids on the Block"). 253. See 4 MCCARTHY, supra note 20, § 25:70.25 (explaining that "retailer placement of goods and retail promotions like coupons are traditional forms of legitimate advertising and that in cyberspace shopping, on-line intermediaries like GOOGLE and YAHOO have stepped into the role of the brick and mortar retailer"). 254. See Lastowka, supra note 13, at 1390 ("What is needed in this arena is a doctrine that keeps the role of trademark law in search results very limited, but does not abdicate the state’s role entirely."). 255. See Tiffany (NJ), Inc. v. eBay, Inc., 600 F.3d 93, 102 (2d Cir. 2010) (noting that the Second Circuit has "referred to the [nominative fair use] doctrine, albeit without adopting or rejecting it" and concluding that the court "need not address the viability of the doctrine to resolve Tiffany’s claim" in favor of eBay). 256. CPC Int’l, Inc. v. Skippy Inc., 214 F.3d 456, 462 (4th Cir. 2000) (citations omitted). 257. Anti-Monopoly, Inc. v. Gen. Mills Fun Grp., 611 F.2d 296, 300 (9th Cir. 1979). functionality doctrine analysis, it will be legitimizing "a radical reinterpretation of the functionality doctrine" with far-reaching and destabilizing consequences for mark-holders and consumers.258 Put bluntly, such a ruling would give Google and other search engine companies a license to run roughshod over the rights of mark-holders in the Internet context.259 Whether it chooses to explicitly follow New Kids on the Block or not, a clear statement of the Fourth Circuit’s endorsement of nominative fair use principles is vital to resolve current uncertainty surrounding the validity of this doctrine in the Circuit.260 Because "at least two district courts" in the Fourth Circuit have expressed doubts about the current validity of the nominative fair use doctrine in the Circuit, an unequivocal endorsement of its principles is necessary to clear up the present state of ambiguity.261 Having clearly established a nominative use defense to trademark infringement, the Fourth Circuit should, on remand, "direct the District Court to consider whether Google’s keyword practices are protected under the nominative fair use doctrine (rather than the functionality doctrine)."262 A finding on remand that Google’s use of the Rosetta Stone marks is nominative fair use will not leave Rosetta Stone and other mark-holders without legal recourse. A trademark-holder wishing to prevent infringing uses of its mark in the search engine advertising context has an obvious target for its legal action: the infringing third-party advertiser. As the Second Circuit’s Tiffany decision explains, the true wrongdoer in the case of misleading or infringing Internet advertising is the advertiser itself.263 A paradigmatic example of the effectiveness of targeting advertisers rather than search engines is Binder v. Disability Group, Inc.,264 a case recently 258. INTA Brief, supra note 6, at 9. 259. Supra Part IV. 260. See Lorillard Tobacco Co. v. S & M Brands, Inc., 616 F. Supp. 2d 581, 588–89 (E.D. Va. 2009) (suggesting that the nominative fair use doctrine may not be applicable in the Fourth Circuit); Nat’l Fed’n of the Blind v. Loompanics Enters., 936 F. Supp. 1232, 1241 (D. Md. 1996) (same). 261. INTA Brief, supra note 6, at 15; see also cases cited supra note 260. 262. Id. at 14–15. 263. See Tiffany (NJ), Inc. v. eBay, Inc., 600 F.3d 93, 114 (2d Cir. 2010) ("It is true that eBay did not itself sell counterfeit Tiffany goods; only the fraudulent vendors did, and that is in part why we conclude that eBay did not infringe Tiffany’s mark."). 264. See Binder v. Disability Grp., Inc., No. CV 07-2760-GHK, 2011 U.S. Dist. LEXIS 7037 at *4 (C.D. Cal. Jan. 25, 2011) (holding defendants liable for trademark infringement based on their use of plaintiffs’ trademark in "AdWords linked to their websites"). decided in the Central District of California.265 In Binder, plaintiff law firm brought a trademark infringement action against defendants for using Google AdWords to purchase plaintiff’s trademark—"Binder and Binder"—as a trigger term and display misleading and potentially confusing advertisements that appeared as sponsored links when users searched for the term "Binder and Binder."266 Analyzing the defendants’ advertisements under a standard multi-factor likelihood of confusion test,267 the District Court found that the advertisements were infringing and awarded the plaintiff damages and attorney’s fees, based on the exceptional nature of the trademark infringement at issue.268 Binder stands for the notion that leaving search engine companies out of AdWords-based trademark infringement actions does not preclude markholders from vigorously protecting their marks and winning judgments against infringing online advertisers.269 The Fourth Circuit should adopt the nominative fair use doctrine to direct the focus of trademark infringement actions onto the parties responsible for the infringing advertising and allow search engines to continue using trademarks in a descriptive, noninfringing manner. The Fourth Circuit now charts a course between perilous shoals, with the threat of overly robust intellectual property rights for mark-holders on one side and the danger of total indemnification for search engines in trademark infringement actions on the other. Although not a perfect solution, the doctrine of nominative fair use should guide the court’s way.270 By ruling that search engines have a right to use trademarks in both organic and sponsored search results, the court will ensure that markholders are not inadvertently granted a total monopoly on the online use of their trademarks.271 Simultaneously, by virtue of the nominative fair use 265. See id. at *3–4 (noting that plaintiffs brought suit solely against defendant advertisers and not against Google, the search engine operator that sold defendant the AdWords trigger terms). 266. Id. at *5. 267. See supra note 64 and accompanying text (discussing the multi-factor likelihood of confusion test for trademark infringement claims, also known as the "MFLOCC test"). 268. See Binder, 2011 U.S. Dist. LEXIS 7037, at *32 ("An award of reasonable attorney’s fees and costs may be made in ‘exceptional cases’ of trademark infringement."). 269. See id. at *38 ("We further award Plaintiffs enhanced damages in the amount of double the lost profits of a total of $292,235.20."). 270. See INTA Brief, supra note 6, at 15 ("This appeal thus provides [the Fourth Circuit] with an appropriate opportunity to expressly consider the nominative fair use doctrine."). 271. See Dogan & Lemley, supra note 10, at 788 ("Overly restrictive trademark law has doctrine’s requirement that protected trademark use be limited to "reasonable necessity," the court will provide mark-holders a tool with which to keep search engines from ignoring trademark rights.272 The stated goal of this Note’s proposal for the Fourth Circuit is balance: The rights of mark-holders must be balanced against the rights of search engine companies and consumers.273 The nominative fair use doctrine represents the best judicial expression of that balance to date. The Fourth Circuit should join the Second, Third, and Ninth Circuits in leading the legal system’s charge forward into the new digital age. VI. Conclusion The Rosetta Stone case comes at a unique moment for the legal system as it strives to keep pace with the never-ceasing march of Internet technology. The implications of the Fourth Circuit’s ruling will reach far beyond its immediate jurisdiction. As Google and its competitors grow and evolve, the trademark infringement issues facing courts will only become more complex—a fact made starkly clear by the arrival of Google Goggles and its successor technologies. The district court’s Rosetta Stone holding represents a good faith effort to confront the emerging trademark issues that accompany the rapid advance of search technology. While the spirit of the court’s decision is in the right place, its resolution of the trademark infringement issue through reliance on a misreading of the functionality doctrine renders the legal analysis of the court’s ruling an unfit foundation for future precedent in search engine cases. The Fourth Circuit is now poised to repair the district court’s legal errors while preserving the progressive spirit of Judge Lee’s ruling. The appeals court has the tools—in the form of the nominative fair use doctrine—to do this; it merely needs to commit to being a leader in the arena of Internet-based intellectual property law. Regardless of the Fourth Circuit’s decision in this case, it is this Note’s contention that the the potential to stifle competition rather than facilitate it. Particularly when trademark holders have economic power, giving them absolute control over the uses of their marks could erect significant [competitive] barriers . . . ."). 272. See Tiffany (NJ), Inc. v. eBay, Inc., 600 F.3d 93, 102 (2d Cir. 2010) ("[O]nly so much of the mark or marks may be used as is reasonably necessary to identify the product or service . . . ."). 273. Supra notes 254–59 and accompanying text. nominative fair use doctrine represents the legal regime of the future in the area of Internet-based trademark infringement cases. Both Internet search technology, and the legal response to that technology are barreling along at remarkable speed; it is up to the Fourth Circuit to decide whether to keep the pace. II. Google AdWords , Google Goggles, and the Search Revolution ..................................................................................1509 A. Google 's Position in the Market.......................................... 1509 B. Google AdWords.................................................................1511 C. Google Goggles...................................................................1514 III. Rosetta Stone v. Google .............................................................1516 A. Facts of the Case: Google's AdWords; Rosetta Stone's Marks......................................................................1516 B. The Court's Decision .......................................................... 1518 C. Examining the Application of the Functionality Doctrine to Word Marks .....................................................1521 D. Functional Features or Functional Use?..............................1525 IV. Rosetta Stone v. Google's Unintended Consequences ...............1528 A. The "New" Functionality Doctrine's Infinite Scope ........... 1528 B. The Functionality Doctrine's Uncertain Future ..................1531 V. A New Way Forward: Recasting Rosetta Stone v . Google .......1534 A . Identifying an Alternative to Functionality ......................... 1534 B. Nominative Fair Use: A Better Path to the Right Result...................................................................................1536 C. Google's Use Is Nominative ............................................... 1541 1 . Identifiability ................................................................ 1541 2 . Reasonable Necessity ................................................... 1544 3. No Indication of Endorsement or Sponsorship............. 1546 D. A Roadmap for the Fourth Circuit ...................................... 1549 1. See Rosetta Stone Ltd . v. Google Inc., 730 F. Supp . 2d 531 , 534 - 35 (E.D. Va. 2010 ) (holding that Google did not infringe Rosetta Stone's trademarks by using them as keywords in its AdWords online advertising program) . 40 . See Google , How Search Works, YOUTUBE (Mar. 4 , 2010 ), http://www.youtube. com/watch?v=BNHR6IQJGZs&feature=channel (last visited Oct . 10 , 2011 ) (noting that the conventional way to search with Google is to "type in [one's] search . . . and hit return") . 41 . See Google , Google Goggles, YOUTUBE (Dec. 6 , 2009 ), http://www.youtube.com/ watch?v=Hhgfz0zPmH4&feature=player_embedded (last visited Oct . 10 , 2011 ) (giving a brief video overview of Google Goggle's capabilities and uses ). 42 . See id. (demonstrating the uses of a Google Goggles-enabled smartphone , which the images) . 51. See Rosetta Stone , 730 F. Supp . 2d at 535 (establishing that Rosetta Stone holds Learning Success , Rosettastone.com, and Rosetta World"). 52. See id. at 534-35 (summarizing the five main issues before the District Court, Scheyder , supra note 50. Rosetta Stone , 730 F. Supp . 2d at 534 . 162. Id. at 15-16 (citing New Kids on the Block v. News Am . Publ'g, Inc., 971 F.2d 302 , 306 - 08 ( 9th Cir . 1992 )). 163 . See New Kids on the Block v . News Am. Publ'g, Inc., 971 F.2d 302 , 308 (9th Cir. 1992 ) (holding that a newspaper company's use of a popular musical band's trademarked Aktiengesellschaft v . Church , 411 F.2d 350 , 352 ( 9th Cir . 1969 ) (holding, on similar grounds, that Volkswagen could not prevent an automobile repair shop from using its mark) . 164. New Kids, 971 F.2d at 306. 165. Dogan & Lemley, supra note 10, at 788. 166. New Kids, 971 F.2d at 305 . 167. Id. at 306 (citing 15 U.S.C. § 1115 ( b)(4) (emphasis added)) . 168 . Id . (citing Soweco Inc . v. Shell Oil Co., 617 F.2d 1178 , 1185 ( 5th Cir . 1980 )). 169 . Id. at 304- 05 . 185 . Brief of eBay Inc. and Yahoo! Inc. as Amici Curiae Supporting Respondents at 6, Rosetta Stone Ltd . v. Google, Inc., No. 10 -2007 (4th Cir. Nov. 1 , 2010 ), 126 - 1 [hereinafter


This is a preview of a remote PDF: https://scholarlycommons.law.wlu.edu/cgi/viewcontent.cgi?article=3333&context=wlulr

A. J. Frey. Lost in Translation: Repairing Rosetta Stone v. Google’s Indecipherable Functionality Holding, Washington and Lee Law Review, 2018,