What is Linking?
The Internet, a vast network of computers, makes it possible to access content (i.e., text, graphics, audio, video, etc.) stored in the files of millions of individual computers. A link is simply a connection between the content of two different files (or between different parts of a single file). A link may lead either to another file in the same web site, or to a file on a different computer located elsewhere on the Internet. Internet browsers automatically decipher the instructions given by links and retrieve the specified file.
The HyperText Markup Language (HTML), used to program pages on the web, allows two types of links. The first, an HREF ("Hypertext REFerence") link, instructs a browser to stop viewing content transmitted from one location, and begin viewing that of another. The link can bring the viewer to a different point on the same page, such as the index links on this top of this page, or to a different page in the same site. For instance, the index links above can be view by following this link (which keeps the viewer on this page), while the BitLaw home page can be viewed by following this link (a different page in the same site). Alternatively, it is simple to use an HREF link to refer to a site that is not on the local web site at all (such as this link to the U.S. Patent and Trademark Office). A single web-page may contain dozens of links to other web-pages. That same page may itself be the "destination" of dozens (or hundreds, or thousands) of other links on other pages.
Although HREF links perform the same basic function, they may appear to the user in many different forms depending upon how they are "embedded" within a web-page. For instance, a graphic may serve as a HREF link. Pressing on the small blue home icon appearing between the different sections of this page returns the user to Bitlaw's main page; pressing on the down arrow icon takes the user to the bottom of the page being viewed. Links also commonly appear as highlighted, underlined or otherwise prominent text. Text with embedded links is called "hypertext." As with an image link, clicking on hypertext instructs the visiting browser to begin displaying different (usually related) information.
The second type of link in HTML is an IMG ("IMaGe") link. An IMG link instructs a visiting browser to supplement the text on the page with an image contained in a separate image file. For example, the small book at the top of this page is actually a graphic file located elsewhere in the BitLaw web site. In this case, the image file is found on the BitLaw web site. However, like the HREF link, an IMG link can also reference a file from a completely different web site. An example would be a web page on art that contains images stored around the world. The web page could contain the following text: "Here it is, my favorite painting, 'Girl With a Pearl Earring' by Vermeer." Using an IMG link, the web page could then direct the visiting browser to retrieve an image of the painting from one of the National Gallery's web-pages, and place it immediately below the text. To the end-user, the integration of the two pieces of content (text and graphic) is seamless, despite the fact that they were taken from two very different sources. The user would never know (though she might suspect) that the image was not created or stored locally. In this respect, an IMG link is different from an HREF link; a user following an HREF link is usually aware that he has "changed pages," either from the different appearance of the newly accessed page, or from the change in the URL address display in the web browser.
Linking is the sine qua non of the Internet's most popular information access tool, the World Wide Web, and there are millions of links in place today. Links allow quick access to information that otherwise could take days or even years to find. Linking also permits the user to determine how deeply to explore a particular topic. As an example, Bitlaw contains an overview of U.S. patent law, with links to the actual text of relevant federal statutes. A particular link may or may not be followed, depending upon the subjective importance of the information to the user. In most cases, the operator of a web-page will desire her page to be the destination of as many links as possible: More links means more hits, and more hits means wider dissemination of whatever information the page is designed to get across. Since most links do not (or cannot) conceal from the end-user the source from which information is being provided, there is little danger that the source of the content on a web-page will be misrepresented.
Problems with Linking
As explained above, linking is the heart and soul of the World Wide Web. Consequently, if linking were disallowed or made illegal in the abstract, the Web would no longer exist. Clearly, no court or legislature would ever go so far as to outlaw all linking. Practically, then, linking is permissible because otherwise the web would end.
Analytically, however, the question requires more thought. When an HREF link is followed by an end user, the user's browser is told to go to a new source for information. In such HREF links, the browser redraws the user's screen from scratch, based upon the information found at the new, linked-to location. In effect, the HREF link merely provides the browser with the Internet address for a new page. Providing this address to an end users web browser should be no more an infringement than providing that web address in written form (as would appear in a newspaper article). The provision of this type of information is similar to providing a library user with the Library of Congress (Dewey Decimal ?) location for a book already in the library. The address is itself is pure information, and is not protected by copyright (see the BitLaw discussion on ideas for more information) or by any other intellectual property regime.
Hence, there would appear to be no legal means for preventing someone from including a link in one page to another. Unfortunately, life is never that simple. In certain situations, links can be used in a way that may violate copyright, defamation, or unfair competition laws. With a little care and planning, the problems described below can be easily avoided.
Consider a hypothetical site designed to present popular comic strips to the public. To incorporate the strips into the site, several IMG links are utilized. These links instruct visiting browsers to retrieve the images comprising the strips from the official comic sites, such as the Doonesbury web-site maintained by Garry Trudeau. The images retrieved in this manner are grouped together on the new hypothetical site, creating a "Sunday comics" page. Although none of the comic images are stored with the hypothetical page, the resulting page arguably creates a new derivative work that is based on those preexisting images (for more information on derivative works, see the BitLaw discussion on copyright scope). Although this argument is not clear-cut, the creation of any derivative work of another's image without permission is copyright infringement.
When the content of another page is incorporated into one's own page by means of an unauthorized IMG link, there is no direct copyright infringement by the creator of the link. This is because the link's creator never copies the pirated content; she merely provides a visiting browser with instructions to retrieve the image, which is then incorporated into the overall page on the user's machine. Thus the only person who copies the protected image is the final user. This does not mean that the creator of the link escapes liability, however. Copyright law provides that one who knowingly makes an infringement possible can herself be held liable under a theory of contributory infringement.
It is typically considered a breach of net etiquette to link to anyone else image through an IMG link without permission. Even more important, unauthorized IMG linking may well be a violation of the right to make a derivative work under copyright law. Consequently, one should always obtain permission (preferably in writing) from the copyright owner of the image prior to creating the link.
Returning to the hypothetical comics page, one could imagine a situation in which the incorporated strips are not properly attributed. Instead, the comics site presents the images as the site's author's own work. Someone unfamiliar with Doonesbury might be led believe that the strips were original to the new site. This is a classic example of "reverse passing-off," in which the work of someone else is passed off as one's own (ordinary passing off occurs when one passes off her own work as that of somebody else). One can also utilize an HREF link to pass off another's work as one's own. For instance, one could tell the reader to click here to see some of Brad Bolin's best original comics. The link leads to a Doonesbury image which is falsely claimed to be original to Brad Bolin. Consequently, the HREF link also is a reverse passing off. Reverse passing off by using a link to pass-off another's work as one's own most likely violates state law governing competitive business practices.
The term defamation refers to a false statement made about someone or some organization that is damaging to their reputation. The law of defamation is complex, as it has been determined by numerous court decisions rather than one national statute. In addition, claims of defamation are subject to first amendment defenses. As a result, a full explanation of defamation will not be set forth here. However, it should be noted that a link to another's page or image could be defamatory, and hence subject someone to legal liability. An example defamatory link would be: This man killed my cat, stole my invention, and threatened to destroy the Internet. The statement itself does not identify the party. The link itself (assuming it actually linked to someone) provides the context that turns the statement into defamation.
Trademark infringement occurs when one party utilizes the mark of another in such a way as to create a likelihood of confusion, mistake and/or deception with the consuming public. The confusion created can be that the defendant's products or services are the same as that of the plaintiff, or that the defendant is somehow associated, affiliated, connected, approved, authorized or sponsored by trademark owner (for more information, see the BitLaw discussion on trademark infringement). As a result, any link that falsely leads the end user to conclude that the web page author is affiliated, approved, or sponsored by the trademark owner could lead to a claim of trademark infringement. An example would be an IMG link that places a trademark of another on the web page as to create such a false conclusion. Such as: Another fine web page by /IMAGE/.
Frames are used to subdivide web pages. For example, the hypothetical site described in the last paragraph could be divided into three frames, with one containing a chronological listing of the various comic strips and their titles, another dedicated to the display of the strips themselves, and a third presenting the biography of the frame's creator. Regardless of the particular strip being displayed in the second frame, the index and biographical information are constantly visible in the first and third frames. As with certain IMG links, frames can mislead the viewer of a site as to the creator of its content.
Moreover, because of their capacity to present data from several different sources as part of one unified display, frames can easily result in the juxtaposition of unrelated, even antithetical, pieces of content: pages from the Jewish Defense League web-site could be made to appear within a frame on a Neo-Nazi site, causing obvious confusion. In this latter case, the problem isn't that the Nazis are passing of the work of the JDL as their own -- the problem is the potential reputational damage to the JDL from having its name associated with a fascist group.
Consequently, framing third-party information into another web page raises issues of copyright infringement (derivative works), passing off, defamation, and trademark infringement, just like the linking situations described above. The solution to this problem is simple: if the use of frames is likely to give rise to the sort of confusion described here, third party pages and images should not be linked into the frame.
The Shetland Times Case
A Scottish case, Shetland Times, Ltd. v. Dr. Jonathan Wills and Another, has received a great deal of attention recently, particularly from members of the Internet community who are concerned about its implications for linking practices. Shetland Times provides an illustration of some of the potential problems with linking that are outlined in the preceding section.
The Shetland Times operates a site on the World Wide Web through which it makes available many of the items in the printed version of its newspaper. The "front page" of the Times' site (i.e., the page which appears upon initial access to the site) consists of a number of news headlines with embedded links to corresponding articles. The intent of the Times, like many other content-providers, is to sell advertising space on its front page once the site achieves a significant amount of traffic.
The Shetland News, a competitor of the Times, also operates a web site, with a layout similar to that used by the Times. For two weeks prior to the initiation of the suit, the News had reproduced, on its own front page, the headlines appearing on the Times' front page. The reproduced headlines were directly linked to articles on pages created and maintained by the Times. Because of their configuration, these links bypassed the front page of the Times. The Times asked for a court order temporarily preventing the News from maintaining the links, arguing that such links constitute copyright infringement.
The presiding judge agreed that the Times had presented at least a prima facie case of copyright infringement based upon the United Kingdom's law governing cable television program providers. First, the judge reasoned that the service provided by the News was sufficiently like that provided by cable television to justify applying the laws governing the latter medium. Second, he found that the articles were being sent by the Times but through the web-site maintained by the News. In terms of the cable TV analogy, the News was interposing itself between the Times and its customers by routing the Times' transmissions through its own web-site. In the process, the front page of the Times' site (on which paid advertisements appear) was bypassed, significantly diminishing the value of the site to potential advertisers. Finally, the judge stated that the headlines may be sufficiently long to have copyright protection, and therefore the copying of the headlines (and not necessarily the links) might be a violation of copyright law.
The relevance of this preliminary decision to linking on the web in general is unclear. The judge ruled without a complete understanding of the technology involved, and based the ruling on U.K. law governing cable television. As a result, it would be inaccurate to say that the ruling represents the position that links to underlying web pages (not the front page of a site) are illegal under U.K. law.