Who Controls the Internet? Illusions of a Borderless World

by Jack Goldsmith and Tim Wu

Who Controls the Internet? : Illusions of a Borderless World

“Who Controls the Internet? Illusions of a Borderless World” is a frustrating book. You’ll stumble across something on every second page which will infuriate you, but it’s also true that almost every page discusses an important legal case, raises an interesting question, or presents an important problem. By describing recent Internet cases and the international legal environment in which they have been resolved, Goldsmith and Wu have illuminated an area which deserves clear and systematic analysis. But the Internet is not a unitary thing to be controlled, and the authors don’t clearly distinguish its various protocols and services.

Disclaimer: This is a book about the law. I’m not a lawyer, although I am an Australian living in the United States who has sent email from China, all of which are relevant to this book.

Goldsmith and Wu’s focus and principal conclusion is (p 180): “What we have seen, time and time again, is that physical coercion by government — the hallmark of a traditional legal system — remains far more important than anyone expected.”. The situations and cases in Who Controls the Internet? clearly prove their point. They demonstrate that national laws are important to large companies like Yahoo!, but I don’t think they prove their case with respect to individuals on the Internet.

The first part of the book, “The Internet Revolution”, does an excellent job summarizing the early development of the Internet, including the extravagant claims of the early Internet and the Internet boom. People said some crazy things. Did John Perry Barlow really write (p 20), “I come from Cyberspace, the new home of Mind”. Ouch!

The authors put a lot of emphasis on the importance of control of the root authority, but I’m not convinced. It’s true that the the right to register domain names within a particular top-level domain (like .com or .tv) means that one company rather than another will make a lot of money, and it’s certainly important to countries like Burindi how much of the revenue from its .bi domain they keep, but what about this from later in the book (p 168):

For the Net to work — for computers all over the world to be able to communicate with one another — the root authority must reliably correlate IP addresses with domain names and uniquely match up both with a particular computer.

It seems to me that Goldsmith and Wu are confusing the legal authority with the technical mechanism behind the domain name system. Computers all over the world communicate using IP addresses. Domain names are an important convenience, but only a convenience. Also, I’m not sure the authors appreciate the role of local (ISP) DNS resolvers as a cut-out between the user and the root. If the United States government turned off the root server, Comcast has a big financial incentive to make sure that my service isn’t interrupted.

As lawyers, Goldsmith and Wu do an excellent job summarizing some important legal cases. On October 20, 2000, Barron’s published, on a web site in New Jersey, an article accusing an Australian billionaire, Joseph Gutnick, of tax evasion and money laundering. Gutnick sued Barron’s in Australian court and won. Comparing the libel laws of the US and Australia, Goldsmith and Wu say (pp151 – 152):

It reflects deeper disagreements between the United States and Australia about the processes that best secure truth, and about the relative value of robust speech versus reputation and uninhibited debate versus order.

When did uninhibited debate become the polar opposite of “order”? In the US, the libel laws are limited by the First Amendment. In Australia, the libel laws are limited by what the people in power think they can get away with. It’s perfectly reasonable to compare the libels laws of the United States with those in Australia, but it doesn’t make sense to assume that the differences reflect something intrinsic about the preferences of the people in each country.

On page 158, Goldsmith and Wu summarize their agreement with the outcome of Gutnick:

Since Barron’s chose to continue to do business in Australia, its consumers in the United States and Japan cannot legitimately expect to receive news from Barron’s that runs afoul of Australian law.

But earlier in the book (p 1!), they discuss a suit in French court against Yahoo’s auction site for selling Nazi memorabilia. There they argue that Internet companies should apply the laws of each country only to users in that country, through the magic of geo-coding. So which is it? Can a US web site avoid being subject to Australian courts by tailoring content to Australian subscribers versus others? What if its geocoding algorithm isn’t perfect? What if a US subscriber visits Australia? What if an Australian subscriber visits the United States?

Another theme of the book is that governments are not just necessary and effective, but also legitimate (p 153):

Even acknowledging that in places that in places like China the laws will often not reflect the wishes of people who live there, differences among laws in the many democratic governments in the world … are presumptively legitimate.

Arguably, there are more “places like China” than there are “many democratic governments”. And it’s common for laws not to reflect the wishes of the majority. But the biggest problem is with the presumption that laws can achieve legitimacy through democratic government. I prefer the presumption that individuals have rights, and that the legitimacy of the law flows from those rights

Goldsmith and Wu convincingly lose the argument about legitimacy when they discuss music copyright (pp 105ff). My issue with their point of view is that copyright is (or should be) a balance between the rights of the user and the copyright owner. If I buy a music CD, I believe I own it, and I should be able to transfer it to my PC or my iPod without the interference of a legally arbitrary DRM mechanism. Goldsmith and Wu mention only one side of this equation, the rights of the copyright holder. Whatever the legal analysis, Goldsmith and Wu are surely wrong about the popularity of pirated music (p 123): “A minority, the Slashdotters, with all the time and expertise in the world, have disappeared into darknets, and won’t pay for music.”

So why bother with this book it all (and why give it a four and not a zero)? Lawrence Lessig, who knows a smidgen more than I do about Internet Law, says this:

It is time that America learn an important lesson about the Internet — that however cyber the space is, it is also real, and subject to real space governments. This is the very best work to make this fundamental point. Goldsmith and Wu have made understandable and accessible an argument political culture should have realized a decade ago.

The book describes an important period, and arguably an important phase change, in Internet history. It raises important questions. I just don’t necessarily like the answers.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s