PATENT AND TRADEMARK LAW
Emerging Issues on the Internet
By Robert C. Scheinfeld and Parker H. Bagley
The New York Law Journal
November 26, 1997
THE PHENOMENON of the Internet has generated many challenging trademark issues. One of the early questions addressed and now seemingly settled by the courts is who (as between a first trademark user and a first domain name user) should own the rights to a particular domain name. Most, if not all, are now familiar with this controversy, initiated by businesses discovering that others had usurped their trademark as a domain name, and were holding them for ransom. Famous trademarks could garner the savvy "cybersquatters" thousands of dollars.1
This extortion led to numerous lawsuits, premised on claims of trademark dilution and infringement. The courts have been in almost universal agreement that a domain name identical to another's federally registered trademark is infringing, and have enjoined such misappropriations.2 These decisions made it clear that the most important thing that could be done to protect one's trademark on the Internet is to have a federal trademark or service mark registration. These decisions also led the United States Patent and Trademark Office (PTO) to recognize that domain names could be federally registered as service marks.3
The allocation of domain names presents yet another controversy. Who should have the right to run this multimillion-dollar enterprise?4 The U.S. registering body for domain names ending in the top-level domain ".com" has been the Herndon, Va., private company of Network Solutions Inc. Recently, the chief organizers of Internet operations have decided to expand to add a number of other private domain name-registering bodies, and up to seven additional top-level domain names.5 NSI's domain name dispute resolution policy has also been recently updated.6
Allocating Domain Names
Domain names continue to be allocated on a first-come, first-served basis, but an important loophole has been closed. Both in the prior and present versions of NSI's dispute-resolution policy, if the challenger presented evidence of a federal trademark registration identical to the domain name, the original domain name registrant could lose the domain name registration. The loophole was that the first domain name registrant, once challenged, had time within which to obtain trademark registration. Tunisia became the country of choice because it offered a virtually "overnight" registration procedure. The current NSI policy closed this loophole by requiring the first domain name registrant to predate the challengers.
While considering the interplay of trademark law and domain name registration, it is important to bear in mind that trademark rights are not global, but regional, while the Internet is everywhere. Civil law countries, unlike the United States, do not require use of a mark in commerce as a prerequisite to registration. Many countries do not recognize the concept of common law rights. Moreover, there are many companies that share the same word or words as a trademark or service mark. For example, there are 85 active U.S. registrations that include the term "Century," and more than 300 containing "Eagle."7 Should each of these 300 owners of Eagle marks wish to use Eagle as a domain name, or part of the domain name, there is no logical and just way to decide who should prevail.
"Linking" has been called the concept on which the Internet is based. "The power of the Web stems from the ability of a link to point to any document, regardless of its status or physical location."8 Linking was purposefully built into the software language spoken throughout the Internet, called hypertext markup language (HTML). It allows jumping from one page to another page of a particular Web site, or from any page of any Web site to any other page of any other Web site, and makes "surfing the 'Net" possible.
To understand how linking can be considered an act of unfair competition in some instances, one needs to realize the commercial realities of the Internet. Typically, one Web site contracts with another Web site to display an advertisement. Much like a billboard, the cost of renting this space directly relates to the "landlord's" Internet location. Web sites that are often viewed, or "hit," are considered prime real estate, and are relatively costly sites at which to advertise. Most Web sites offer a number of pages of data. The "home page" is where most viewers enter a site, and it is on this page that the majority of advertising occurs.
The complaints about linking are not linking per se, but rather about links that circumvent another's home page or the advertising portion, thereof. Such links result in the advertisements of the sponsor Web site not being viewed, and the value inherent to that Web site to advertisers is therefore reduced. In essence, the Web site that has established the link benefits because viewers find its value enhanced (i.e., additional relevant access), but this is done at the expense of the linked site.
It is only this year that linking has been charged as an unfair trade practice. In April, Ticketmaster Corp. v. Microsoft Corp.9 was filed. In its complaint, Ticketmaster alleged that Microsoft's link to its site violates federal anti-dilution and trademark infringement law. At the core of the complaint is Microsoft's link to Ticketmaster's live-event and ticketing data bases. Microsoft's on-line guide to Seattle, called "Seattle Sidewalk," provides information about current events in the area. Viewers benefit from the listings Ticketmaster has compiled, but bypass Ticketmaster's home page, and hence its advertisers, by linking directly to the event listings.
Microsoft's answer to the complaint has been that "any business participating in the Internet invites other participants to use the business's Internet addresses and uniform resource locators (URLs) to contact it."10 Absent a settlement, the court will decide whether this "implied license to link" argument will prevail.
How the court rules could have far-reaching consequences, since it is common to link to other's pages that
contain related information. On the one hand, prohibiting such links, or limiting them to linking only to home pages, could significantly slow the ability for Web "surfers" to locate wanted information. The Internet is now bogged down by traffic, and forcing viewers to sift through unwanted pages is likely to ruffle more than a few feathers.
"Framing" is a special form of linking. The viewer is linked to the information of another site, but the advertisements present on the parent site continue to surround, or "frame" the linked page. As such, advertisers have access to the visitor not only through the home page of the parent Web site, but on every page or site the visitor subsequently views.
One can frame a complete Web site. In so doing, the framer provides no content itself, but rather uses the content of another's Web site and sells the advertising space on the surrounding borders. With framing, there is no way to tell if the content is appropriately attributed to the framing site or to someone else. Framing can be thought of as removing the label from another's goods, and replacing it with the framer's label. The consumer has no way of knowing that it is someone else who is responsible for the quality of the product.
There have been, so far, few legal challenges with regard to framing, but those that have been brought have received intense legal analysis. Commentators tend simultaneously to address, and sometimes confuse, the framing and linking issues. Some defendants have turned to the "implied license to link" defense to gather sympathy and justify framing.
Seemingly the first framing lawsuit, Washington Post Co. v. TotalNews Inc., was filed in New York federal court in February. A settlement was reached before the defendant's answer. The defendant news service, TotalNews, was alleged to have violated both copyright and trademark law.11 The plaintiffs, including The Washington Post, CNN, Time, Reuters, Dow Jones and The Los Angeles Times asserted that TotalNews' practice of framing their Web sites misappropriated their copyrighted material and infringed their trademarks.
Apparently, TotalNews provided no content or value of its own. The Web site merely linked to numerous other well-known news sites. At all times, the visitor saw only the Internet address of TotalNews, no matter which Web site appeared. Advertisements framing the linked sites were sold by TotalNews, and were under TotalNews' complete control.
In the settlement, TotalNews agreed to refrain permanently from either directly or indirectly causing any of the plaintiffs' Web sites to appear on a user's computer screen with any material associated with the defendant or any third party. Since reaching the settlement, TotalNews has negotiated a link license with some of the plaintiffs.
Because a settlement was reached, interesting trademark questions were not reached. It is likely, however, that the court would have agreed that violations of both copyright and trademark law had occurred.
A meta-tag is a software parameter of HTML, in which Web pages are written. Through use of meta-tags, a Web site creator can describe what is available at that particular site. This description is relied on by Internet search engines, such as Yahoo or Infoseek, to match the Web site to the search query. Absent a meta-tag, coders for a particular search engine must either read, and manually enter the Web site content (rarely done now), or use an automated (or robot) system to scan a certain amount of text (usually the first 200 characters). The site is characterized by the words or phrases present in this scanned test.
A meta-tag is not necessary for normal operation of a Web site, but it is an accessory more Web page creators are including in design. Although it can be used to list the author, or make a claim of copyright, it is most often used to specify "key words" matched during the search. While the descriptive meta-tags will appear in the generated search engine report, key word meta-tags will not.
As mentioned above, Internet advertising is geared toward the number of "hits" or times a certain Web site is accessed. Sites visited frequently can simply command higher advertising fees. Because a site will be visited more frequently the more often it is uncovered during a computer search, there exists a strong commercial incentive for a site to use meta-tags and develop extensive key word lists. Moreover, there are no limits to the number of key words a meta-tag might contain. Theoretically, a site developer could list every word in the English language and thereby increases the site's hit rate.
Not surprisingly, there are those who have used meta-tags to compete unfairly. Meta-tags have been used improperly to list the names of specific well-known businesses, including direct competitors of those actually offering the sites, to increase the chances that their site will be viewed. Searchers are eventually directed to a site that has nothing to do with the intended subject of the search. The unscrupulous site operator then capitalizes on the opportunity by offering the visitor competing goods or services.
For instance, a contemporary search under the terms "Disney" or "Mickey Mouse" will turn up travel agencies offering package deals to the Disney resorts. While Disney has apparently not yet mounted a legal challenge, some other wronged businesses have been more aggressive.
One meta-tags suit, Insituform Technologies Inc. v. Nat'l Envirotech Group, brought in Louisiana federal court, settled in August.12 Soon after being challenged, the defendant agreed to delete all reference to the plaintiff at its Web site.
It is likely the plaintiff in Insituform would have prevailed. First, both companies were direct competitors in the business of pipeline-reconstruction. Second, the defendant not only included the plaintiff's name in its meta-tag, but also allegedly decorated its Web site with slogans and illustrations copied directly from the plaintiff.13 Thus, in addition to a "hidden" meta-tag, there was visible evidence of infringement.
An important ruling regarding the use of meta-tag recently came down from a California federal district court in Playboy Enterprises Inc. v. Calvin Designer Label.14 As in Insituform, the alleged misconduct was not limited to an offending key word in the defendant's meta-tag. By the preliminary injunction, the alleged infringer, an adult-oriented service, was ordered to remove all references to "Playboy" and "Playmate" from its Web site, including use of those famous marks as meta-tags.15
There are, however, legitimate uses of meta-tags. They represent an important means by which today's automated search engines find and classify Web sites. But listing a competitor or even a non-competitor as a key word in the meta-tag for a Web site can expose a company to an infringement suit.
(1) See Sougata Mukherjee, "Internet Domain Name Sellers Want Big Bucks for Key Names," Orlando Bus. J. (Fla.), March 14, 1997, available in 1997 WL 7548349 (listing negotiations of $12,000 for cybercafe.com, $350,000 for abc.net and $200,000 for jimihendrix.com).
(2) E.g., Hasbro Inc. v. Internet Entertainment Group Ltd., 40 USPQ2d 1479 (W.D. Wash. 1996) (enjoining an adult-oriented Web site from using candyland.com); Intermatic Inc. v. Toeppen, 947 F.Supp. 1227 (N.D. Ill.1996) (enjoining defendant's use of intermatic.com); Panavision Int'l LP v. Toeppen, 945 F.Supp. 1296 (C.D. Ca.1996) (enjoining defendant's use of "panavision.com").
(3) See, e.g., Alan J. Hartnick, "Copyright & Trademark on the Internet -- And Where to Sue," 217 New York Law Journal 34 (1997) (noting the USPTO's position that "the service of providing the information offered at [a] Web site is an information service that should be classified in the class of the information [being offered]"). See also the USPTO Web site at http://www.uspto.gov/web/uspto/info/domain
(4) See, e.g., Alan Cohen, "The Web Meets the Law," AmLaw Tech, Spring 1997; Sally M. Abel & Marilyn Tiki Dare, "Trademark Issues in Cyberspace: The Brave New Frontier" (visited April 4, 1997) http://www.fenwick.com/pub/sma-trade2
(5) Bill Pietrucha, "Internet Group Adds Seven Top Level Domain Names," Newsbytes News Network, Feb. 5, 1997, available in 1997 WL 7971793.
(6) See Network Solutions' Domain Name Dispute Resolution Policy (Rev. 02) (visited March 29, 1997) http://rs.internic.net/domain-info/
(7) Anne Hiaring, "Trademarks and the Internet," 454 PLI/PAT. 447, 451-52 (1996).
(8) ACLU v. Reno, 929 F. Supp. 824, 837 (E.D. Pa.) (prob. juris. noted 117 S.Ct. 554 ).
(9) CV 97-3055 RAP (C.D.Cal. filed April 28, 1997).
(10) Martin J. Elgison & James M. Jordan III, "Trademark Cases Arise From Meta-Tags, Frames," National Law Journal, Oct. 20, 1997 at C6.
(11) Meeka Jun, "Been 'Framed'?: Imposters Beware," NYLJ, June 20, 1997 at 5.
(12) Ann Davis, "Web Weaves a Tangled Trademark Issue," The Wall Street Journal, Sept. 15, 1997, at B10.
(14) No. C-97-3204 CAL (N.D. Cal. Sept. 8, 1997).
Robert C. Scheinfeld and Parker H. Bagley are partners in the intellectual property group and the New York office of Baker & Botts LLP.
Home|Contents|Marketplace|News|Employment|Practice Areas|Resources|Law Tech.|Law Firms
Copyright 1997, The New York Law Publishing Company. All Rights Reserved.
Access to Law Journal EXTRA! is governed by its Rules of Use. Send comments to email@example.com