DOES DEEP LINKING INFRINGE COPYRIGHT?

September 2000

Guest editor: Steven Melamut, Reference/Electronic Services Librarian, University of North Carolina

Deep linking is a new buzzword among librarians, webmanagers and Internet lawyers. Deep linking refers to links on a website to the interior of another website bypassing the home page of the second one. For example, a deep link is created where website "X" links to a document or webpage belonging to "Z" and the link bypasses the intended route to the site, and it is not made evident that the shortcut has taken place.

Many library webpages contain links to other websites. Librarians often want to deep link into a site rather than having to include instructions on the library’s webpage on how to navigate to the appropriate item. Not only does a deep link get the user of the library’s webpage to the right place on the web but it also saves the user’s time. The deep link may also make it more apparent to the user why the link is included as opposed to a link to the first page. For example, if the library wants to include a link to Professor Lolly Gasaway’s public domain chart, it makes much more sense to link directly to it than to her personal homepage with instructions to scroll down to the courses she teaches, click on “Intellectual Property” then “Other materials” then “Copyright” then finally “Public Domain Chart.”

There are now some cases dealing with deep linking, but to date, all have been cases involving commercial entities, and even those provide little guidance.[1]

Is deep linking really a problem? Both companies and legislators seem to think so. Some states, for example Virginia, have begun to draft legislation in this area. The proposed Virginia Website Protection Act would require the State to establish a website registry and creates new tort claims and remedies for injuries to websites. A site owner could prohibit a user from knowingly and intentionally bypassing the website's homepage and accessing information within the Web site, where the use would cause unfair competition, injure the website owner or his or her property. The Act even includes protection against embarrassment or defamation to the site.

Website owners have valid concerns about deep linking. Problems include avoiding advertising on sites, loss of computer capacity and bandwidth resulting from deep linking and the use of framing to disguise the true source of the material or to block advertising. Framing refers to webpages that are divided into multiple areas each which can contain material from other remote sites. When a frame is book marked, it is the frame site that is saved, not the true home of the material. Concerns regarding links between different sites is not new.[2] There are already existing contracts to create and maintain mutual links, guarantee the prominence of a link, and prohibit the creation of links to competing products. Such contracts have already been the source of litigation.

Many websites derive substantial income from advertising. Bypassing the initial pages containing the advertising decreases its value and may threaten the profitability of the site. In Ticketmaster v. Tickets.com[3] the defendant provides a ticket buying service on the Internet. Some tickets are exclusive to Ticketmaster, so Tickets.com created links directly to the sales pages on Ticketmaster. Although Ticketmaster ultimately sold the tickets, the arrangement threatened advertising income. This case is still pending. The same problem was litigated in Ticketmaster v. Microsoft, but that case was settled confidentially in late January 1999 so the terms are unknown.

Another reason that webpage owners may object to deep linking is potential loss of computer capacity and bandwidth. In eBay v. Bidders Edge,[4] the court issued a preliminary injunction after finding that the auction aggregator Bidders Edge (BE) was accessing the auction site EBay about 100,000 times per day. BE was using "robots" to access EBay's bidding site and provide users with a comparison of all of the Internet auction sites. Robots or spiders are software programs used to search the Internet and copy or retrieve information from websites. The robot can perform thousands of instructions per minute and can consume a large proportion of a site's resources making those resources unavailable to other users and slowing access to the site. By creating a large enough number of connections, it could even cause a server crash. EBay asserted that BE accounted for a huge number of requests and total data transferred by eBay during certain periods.

At first glance, observers might have expected copyright challenges to deep linking. After all, a website is a collection of ideas, text and graphics and the creation of a website can be expensive and time consuming. Certainly, if the defendants copied the site and placed it on its own server, the copyright argument would be persuasive. Even in most framing cases, defendants are not downloading the material; they are only creating an access point. Copyright claims are more likely where the defendant appears to claim credit or ownership for the linked site.

Some people argue that it is not in the public interest to permit website owners to restrict access. Proponents of this view maintain that the Internet will cease to function if material published on the Web cannot be universally accessed. In opposition to this view, site owners maintain that the Internet cannot function if intellectual property rights are not respected.

Linking cases have included a wide range of complaints brought under unfair competition, contract violation, unjust enrichment, trespass, misappropriation, passing off, false advertising, tortious interference with prospective business advantage and trademark violations. Most of the early cases were settled and it is difficult to predict how the courts will ultimately rule. Although some courts are accepting a "subtle harm analysis," others have refused to permit an injunction without evidence of irreparable or tangible harm.[5]

The likelihood of litigation depends on the type of sites involved and the content used. Library websites that interfere with another sites’ ability to attract advertising is risky. Further, robots that place a large demand upon resources may invite litigation. When creating library websites in the corporate environment that link to commercial sites, it may be best to ask permission to deep link to those resources. This is especially true when using frames. Deep linking on webpages for nonprofit libraries presents less risk of litigation. All site owners should avoid webpages that confuse visitors as to ownership or that take an unfair advantage of another site's content. If your company does not have a linking policy, it may be worthwhile to create a clear linking policy for the library’s website.

---

[1] Judge Harry L. Hupp of the U. S. District Court for the Central District of California, recently observed that "[t]here are virtually no cases on deep linking, particularly none on linking to particular interior pages of a website." 2000 U.S. Dist. LEXIS 4553, at *12, 54 U.S.P.Q.2D (BNA) 1344 (C.D. Cal. 2000).

[2] The Business Law Section of the American Bar Association published a book entitled Web-Linking Agreements, Contracting Strategies and Model Provisions in 1997.

[3] 2000 U.S. Dist. LEXIS 4553, 54 U.S.P.Q.2D (BNA) 1344 (C.D. Cal. 2000).

[4] 2000 U.S. Dist. LEXIS 7287, 54 U.S.P.Q.2D (BNA) 1798 (N.D. Cal. 2000).

[5] Each of the cases discussed here involved situations where contract negotiations between parties had broken down.