Copyright Law Meets the World Wide Web

Copyright protection stems from Federal statutory law. The statute gives a "laundry list" of what is copyrightable subject matter. Included in that list are original "literary," "pictorial," and "graphic" works "fixed in any tangible medium, so there is little question that Web pages are copyrightable subject matter. Unlike patents, the author of copyrightable material does not need to take any actions to obtain the copyright. As the statute puts it, copyrights merely "subsist." This means that every person who has ever written an original Web page or created an original graphic for the Web has a copyright (or his or her employer does) on that page or graphic.

The tough question, then, is what rights are afforded the owner of a copyright on a Web page? This is extremely new subject matter for the courts to handle. As of mid-August the word "Internet" appeared in less than a dozen written opinions of any kind from any Federal court, the courts with jurisdiction for copyright law. Of those cases, all but one were from after October 1994, and only one of those was at an appellate level court (and that one really had nothing to do with the Internet). The Supreme Court has never even uttered the word "Internet" in any form. Even in the handful of trial level opinions that mention the Internet, only one case involved intellectual property on the Internet. That occurred last year when MTV sued former employee Adam Curry to force him to remove the MTV logo from his Web page and to cease using the domain name address "mtv.com." The trial court dismissed the case on the grounds that MTV had licensed Curry to use the trademarks, and it did not matter if that constituted infringement. However, the court did imply in a footnote that although Internic will give a domain name to anyone for free on a first-come-first-serve basis, the owner of a trademark used by another in a domain name could seek relief in Federal court. That footnote is currently the entire body of existing case law on intellectual property on the Internet.

Ordinarily, the law grants the owner of a copyright the exclusive right to authorize reproduction of the copyrighted work. However, the nature of HTTP requires that a Web page be copied at the very least into the memory of the client computer. It is very safe to assume that a court would consider the authorization to copy a Web page (without password protection) into memory as perfectly legal. The existence of other implied authorization is less clear.

Can net surfers save Web pages on their hard disks for their own later use? It would be unreasonable to prohibit Web browsers from storing a copy on a hard disk for caching purposes. Simply as a matter of public policy, caching Web pages must be allowed by implied consent of the copyright owner. Net traffic would get extremely congested if every Web page access that could otherwise be cached had to be resent.

Although copies outside a cache are not necessarily authorized implicitly, courts like to analogize new and technological issues to more familiar parallels. Suppose the copyright owner of a document leaves that document on top of a copying machine along with a stack of dimes. Anyone can imply authorization to copy the document because the owner has intentionally and knowingly made it easy and convenient to copy. It takes almost no effort and a fraction of a second to click on the save icon in Mosaic, and almost instantly the Web page is copied. On the other hand, the copyright owner of a Web page can argue that he or she has no choice in the matter. Once again, that is simply the nature of HTTP and the Internet in general.

Consider, on the other hand, a Web author who places explicit notice in a prominent place on the page, stating that the page was intended for immediate viewing only, and that copying the page even to disk was prohibited. It is safe to assume that anyone viewing the page who clicked on the save button would be infringing the copyright. It is even more likely that making a hard copy of a web page on a printer infringes the copyrights in that case. Indeed, clicking on the print icon is just as easy as clicking on the save icon, but such hard copying is moving the work to a different medium.

Saving the Web page to disk keeps the work in the same form in which it was transmitted -- HTML source code. As mentioned earlier, it is necessary that this code be saved in order for the browser to display the document (which is what was intended by the copyright owner in the first place). Clicking on the save icon merely moves that code from one place to another. Printing a Web page involves making a copy of the document that is not required for viewing the document as intended.

One defense for both copying to disk and printing of Web pages may be fair use. According to the copyright statutes the fair use of copyrighted material is not infringing. The law provides four tests to determine if the use of a work is fair use: the purpose and character of the use, the nature of the work, the amount and substantialness of the portion of the work used, and the effect of the use on the value of the work.

The statute granting the fair use exemption specifically mentions several purposes included in the fair use doctrine, including "teaching (including multiple copies for classroom use), scholarship, [and] research." A commercial use will very rarely (but not never) be found to be fair use. Just because the purpose of the copying falls within one of those enumerated in the statute does not automatically mean that the copying is fair use. For instance, a teacher still cannot make twenty copies of an entire text book and claim fair use. The remaining tests must be examined as well.

The "nature of the work" refers to several factors including whether the work is factual or fiction. Factual works receive less protection from fair use copying than fictional or fanciful works. Application of this test to the World Wide Web will obviously depend on the Web page in question. Infringement is more likely found when a story on the Web is copied or printed than when a research report is copied or printed. Fair use is also more likely to be found when the copied work is published; however, it is likely that courts will consider anything available on the Web as published. One of the results of the popularity of the Web is the fact that now almost anyone can publish a work. It will be interesting to see if that fact will change how much weight courts give to the published status of a work. The commercial status of the work may also make a difference as to whether copying is fair use, though some courts have refused to take that factor into account. If it is going to make any difference at all, a Web page that serves a commercial purpose is better protected than one that does not serve a commercial purpose.

The test relating to the amount and substantialness of the portion copied might be tricky when it comes to the Web. Generally, if someone copies an entire work, fair use is much less likely to be found than if someone only copies, say, a couple pages of a large book. When someone copies a Web page they usually copy an entire page, but is that the entire work, or is the entire Web site the work in question? Web documents tend to be very modular. Only a small amount of information is placed on a single page, and hypertext is used to link to other parts of the Web site. When courts are first forced to consider the question of fair use in copying the World Wide Web there will be much disagreement among different courts as to what constitutes the whole "work" (and many will try to avoid the question altogether by deciding the issue of fair use on the other three tests).

The last test, the effect of the use on the value of the work, is quite simple to examine. Because someone could freely go to a given Web page and copy it, that copy is not going to diminish any commercial value of the page. The fact that making a simple hard copy of a Web page for your own personal use does extremely little or no commercial damage to the owner of the Web page is likely the deciding factor that such copying is fair use. That is especially true when the copying is done for academic or research purposes, or when the information on the Web page is factual material.

What happens when someone uses copyrighted text or a copyrighted image in a Web page or places it on an FTP site? This is a much clearer case of copyright infringement, which unfortunately occurs much too often. The only issue here is enforcement. It is much easier to see when someone is posting a commercial piece of software or a scan of last month's Playboy centerfold than it is to know when someone is sending the same to a printer at home (which is virtually impossible to know). Yet it is difficult for administrators of Internet sites to catch these cases immediately when several hundred people have home pages there. It is even harder when thousands of people are posting to a bulletin board or when tens of thousands of people place files on anonymous FTP servers or Usenet groups. The administrators still have an obligation to remove the material immediately after the problem is discovered, even before the copyright owner notices. If the material is not removed even after the administrator knows about it, the administrator may be liable for contributory infringement. For example, Microsoft asked maintainers of the many "unofficial Windows95 home pages" to remove the logo from the page.

Is it acceptable to place in a Web page an inline image of a URL on another host? The answer is that it is not acceptable without the owner's consent. Netscape provides images on their home site designed for use with their <BODY BACKGROUND> tag which they encourage people to utilize. That is legal and acceptable, but putting an inline image of a photograph from another person's web pages in your own is not. It is true that the copyrighted image is not copied by the referring Web author. For that matter, when a client accesses the page in question, the image never even touches the referring server. The server sends the page to the client and embedded in the HTML source are instructions for the client to grab the image from the second server and place it in the proper place on the page. It is not contributory infringement, because the client user could legally access the image itself if he or she wanted without instructions from the referring web page. Nonetheless, this practice creates what is known as a derivative work. The right to control such derivative works is another right afforded copyright owners. Just as copyright holders are the only ones who can authorize the copying of the copyrighted work, they are the only ones who can authorize the creation of a derivative work, which is any work that is based upon a preexisting work. A comparable situation is where a person buys a book of paintings, cuts out the pages, sets them on mountings, and sells them. That was found to be a derivative work when it really happened. Although the defendant could have resold the book since he bought it and it was his to dispose of as he pleased, remounting the pictures in the book and selling them as individual pieces of art was infringement of the copyrights.

Another intellectual property issue which is unique to the web is where the author of a Web page attempts to prevent others from placing links to his or her page in other pages without consent. A friend placed a notice to that effect on her home page after she had found her page referred to on an offensive page elsewhere on the Web. She asked that others ask permission before placing links to her home page on their own pages. Unlike the other issues raised above, enforcement would be very easy on this issue by keeping referrer logs and writing a script to compare them with a set list of authorized referrers. Unfortunately, assistance is probably not available from the legal system here. The referring page does not contain any of the referred page. The referring page simply has instructions on how to find the referred page. Placed in the form of a hypertext link, the referring author has only made more convenient what he or she could have also done by saying, "And there is this great page over here you should check out." Such a page is also not a derivative work of the referred page, as it is not based on or using the referred page in anyway other than by telling the world where the referred page is. Whether it is appropriate or acceptable to disobey the owner's wishes not to refer to his or her pages is another question (one that I think is rather obvious), but it is not illegal to ignore them.

One issue that is too complex for discussion in this article, but which the Internet community should think about, is the implications of international law on intellectual property on the Internet. Reconsider the example involving an inline image in a Web page based on a URL from another host. What happens when the server with the referring page is in Russia, the client and its user are in Brazil, and the server with the inline image, whose owner's rights are being infringed, is in the United States? Most industrialized nations, including the United States, are members of the Berne Convention. The Berne Convention is an international treaty relating to intellectual property, and for the most part it provides international intellectual property rights protection for anyone in any member country. But even with the Berne convention, the copyright laws still differ from country to country. Even if the only issue was jurisdiction, where does the infringement occur in this example? In Russia where the derivative work was conceived? In Brazil where the derivative work is being created? Or in the United States where the image was copied for illegal purposes? That question is beyond the scope of this article.

It is impossible to tell what the courts will do when first confronted with these issues. It sometimes takes years before courts really understand the issues involved with new technology. Rest assured: there will much disagreement among the various districts and among the various circuits as to the answers to these questions.

Copyright 1995 by Matt Rosenberg

Want more Crossroads articles about Social Issues? Go to the index or to the next one or the previous one.