December 1, 1998


by Andy Oram
American Reporter Correspondent

CAMBRIDGE, MASS.—The new Copyright Act was meant to promote electronic networks by protecting content placed online. It was not meant to erect barriers against site administrators and service providers by imposing extra requirements on them. But its “directory of agents” section, barely noticed by interested parties before passage of the act, could create an annoying and—in my judgment—unnecessary burden on a wide range of Internet services.

The framers of the Digital Millennium Copyright Act of 1998 wanted to increase the penalties for electronically distributing software, music, and other mass-market content without permission. But at the behest of librarians, educators, researchers, service providers, and other people who depend on the dissemination of information over the Internet, the bill picked up a number of sections over time to protect legal, innocent copying.

For instance, it would be unfair to sue a university because a student put unlicensed software up for file transfer, or to sue a service provider because it rented a Web site to someone who was offering commercial music recordings for downloading.

Indeed, any content that appears on the Internet crosses a number of systems whose owners know nothing about it, just as an infringing book passes through distributors, retail stores, and public libraries. Although few people know it, these institutions can be sued for books that infringe on copyright. But they are well protected because they deal with only a few well-known publishers, whereas Internet providers accept material from a relatively enormous range of sources.

So a “safe harbor” was created in the copyright act for sites that host content put up by other users. But Congress thought that it might be difficult to determine the owner of a site from its URL or other identifier (an assumption I will challenge later). So they inserted clause 512(c), requiring the Copyright Office in the Library of Congress to maintain a “directory of agents” that lists the owner for each Internet service provider, and allows them to charge a fee to cover costs.

What is a “service provider”? If it’s simply a business offering Internet service, like the familiar dial-up access most of us use, the directory would contain about 5000 agents and would burden only a narrow range of people. But the law defines “service provider” in relatively vague terms: “a provider of online services or network access, or the operator of facilities therefor.” An analysis by the law firm of Lutzker & Lutzker in Washington, D.C. says this definition “is intended to be broad.”

The law could be interpreted to cover a school or university that gives Web sites to students, or a non-profit community network that gives free accounts to low-income people. My own company, the publisher O’Reilly & Associates, allows its employees to put up personal Web sites and even at times to run servers of their own under the domain.

Will all these myriad organizations have to file with the Copyright Office? And could it handle the onslaught of filings if we did?

Sometimes the chain of ownership extends through several levels. For instance, an Internet service provider may provide a section of its Web site to a community group. That community group may then offer its users accounts and posting privileges.

If a user posts something infringing on copyright (say, an internal document from the Church of Scientology), who should be liable? The user, the community group, or the service provider? The Copyright Act protects the community group and service provider—but only if they file.

But this is not the worst of it. Suppose you put up a Web page with links to sites you like—and unbeknownst to you, one of those sites has an unlicensed recording or piece of software? Could a link on a Web page be considered contributory copyright infringement?

The idea is by no means ridiculous. People have been sued for making links to other sites without those sites’ permission. The legal status of Web links is still unclear in the courts, but according to my (legally untrained) reading of the Copyright Act such links are protected. (You are transmitting the material without modification, and are not making a copy.) But still, you may have to file with the Copyright Office to be protected!

Robert Raben, Democratic counsel on the House Judiciary Committee, says that filing is required in order to be protected by the new Copyright Act, but that site administrators could take their chances by not filing. They would then depend on whatever protections are offered by common law and tradition.

The directory was added fairly late in the life of the Copyright Act—first in the Senate version, then on the floor of the House—and seems to have escaped the notice of all the affected parties, such as service providers and universities. This oversight is not surprising because those institutions were busy fighting so many other provisions. The Copyright Act was a controversial bill opposed by communities ranging from librarians to cryptography researchers; it changed considerably during the two years it needed to pass Congress.

The Copyright Office is now trying to disentangle the rules and find the least burdensome way to implement section 512(c). But their interim regulations, for starters, lead one to groan. They provide a (mercifully short) form that must be sent in to the Office with $20 for handling.

Twenty dollars doesn’t seem like much of a burden for any organization that offers Internet accounts to users. But the law is just one more burden that they don’t deserve.

Furthermore, the United States is only the first of countries to adopt such laws. The 1998 Copyright Act was passed in order to bring us into compliance with an international treaty. So other countries are sure to follow in our path—and fees for filing in those countries may truly be a barrier to sites going online.

Bookstores don’t have to file with the Copyright Office in case someone puts pirated books there for sale. Nor do television and radio stations have to file just in case they broadcast a pirated film clip or recording. The Copyright Act places a unique burden on one of the most democratic media, one that has historically faced no regulations at all.

Enough griping—what can we all do to fix this problem? The Copyright Office is open to suggestions and will soon start a formal comment period before making a final ruling.

In my opinion, tracking down copyright infringement online is by no means as hard as the people who wrote 512(c) think. I would like to modestly propose a solution that grew out of an exchange I recently had with a lawyer colleague.

One of his clients had a copyright problem: some employees left to start a competing business and placed online some documents they stole from the client. Using the simple advice I gave him, he found the infringers and resolved the problem within a few hours.

What I told him to do was take the host name of the site (for instance,, strip it down to the last two parts ( and enter a whois query on a machine offering standard Internet commands (“whois”). This command would reveal the administrative and technical contact people, who can find out what user owns the infringing files.

Different databases have to be searched for different domains, but the contact people of any domain can be found. And since these contact people give out user accounts, they can find out who put any document or program on their systems.

Using this principle, the Copyright Office could offer a “directory” that is simply a Web-page form leading to a program that interfaces with the “whois” command. (In fact such interfaces are already available). They may thus be able to follow the letter of the law while almost completely eliminating administrative costs for themselves and Internet services.

Such a solution would require a questionable definition of “service provider” as “owner of a domain name.” It may not fully protect administrator in the middle between the domain name owner and the users—for instance, the community group mentioned earlier.

But we all need to work together and think creatively to prevent this law from running away with us. Nobody wants to put unnecessary barriers in the way of offering Internet content. And it would be anathema to most Americans to force anyone to file a form just in order to provide information.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Editor, O’Reilly Media
Author’s home page
Other articles in chronological order
Index to other articles