ILAW: Zittrain and Fisher on Domain Names

Jul 9, 2003

The final session today @ ILAW: professors Zittrain and Fisher on "Content: ICANN and Domain Names."

JZ: The Internet waits for no one. We are going to talk about Internet governance today--tag team style. Frame: This is something you may or may not have heard of. How many people know about ICANN?
It's always been of interest to me why people find ICANN interesting. CDT and ACM both devote big sections of their websites to ICANN. Why do they care? Why should we care? Who should run the whole outfit?
The mess--how did we get here? The mops--legal and political intervention.
Origins: nerds having a good time before the lawyers and business people came in and ruined everything. Here are the nerds [shows picture of Postel, etc.] Here is the early network [shows diagram]. But first, more about the IETF.
Quote by Dave Clark, RE renaming ourselves so no one would be interested in us.
Describes RFCs...RFC #2, etc. until the final RFC. No more comments needed.
Clark: "RFC not meant to be an assertion of control."
Back to how it works: not good to use only numbers to identify. Use names instead--so they could reach your computer without knowing the numbers. Names kept on a list.
In 1984: new and improved DNS. It wasn't scaling; if you make it big, it breaks. Distribute the directory. Insight was to allow people to maintain their own chunks of the list. Ask the root where the list of .edu can be found. The root tells you who. You go to that entity and so on down the line.
Aaron Swartz: How did they find the root?
JZ: It was a hard-coded bootstrap. Built into the software. Who controlled the root? Jon Postel drew the short straw. Jon eventually ran only the A root. He called himself IANA.
In 1993, Jon's system essentially broke--because it became too hard/boring and routine. The NSF worked to find an entity who would maintain the root for a flat fee: NSI won the contract. NSI had a brilliant insight: we can charge a fee to those registering the name. You rent your name for a year. This became quite a big business.
One pressure: other entrepreneurs were jealous of NSI; they wanted a piece of the action. Another pressure: expand the namespace.
ISO got the job of assigning extensions to countries.
Another pressure: corporations discovering "cybersquatting." The dotcom boom. A guy named Josh had McDonalds.com, etc. Kaplan.com was registered by the Princeton Review.
Terry: So I am going to be talking about the law in this area: first phase--trademark law, 1993-99. Round one: when the companies who were aggrieved looked around for help with domain names, they turned to trademark law.
Four kinds of trademark infringement: four concentric circles. 1/ Identical marks on competitive products, 2/ Similar mark on a competitive product (test: likelihood of consumer confusion?) , 3/ Similar marks on non-competitive products (test: likelihood of confusion?), 4/ Dilution (behavior dilutes power of trademark).
1996/7: Doctorines similar throughout the world and becoming more so. Why? TRIPS. Variations exist but gradually being reduced. Trademark law is not well designed for addressing this problem. Courts adapted. Amadeus Marketing; British Telecom, etc. This wasn't working very well.
Problems and limitations:
costs money to bring suit
time
jurisdictional variations
doctorine "twisted" in this context
Other limits: only available to famous trademarks, requires use of commerce, and sometimes hard to show blurring or tarnishing.
Jonathan: Okay, so this moves us to part 3: taking over the DNS, or How the nerds decided they'd had it w/lawyers and businesspeople and more or less turned the system over.
This is what happened next, even as courts dealt w/this case by case.
Solutions that Jon Postel came up with:
Jon thinks hard, adds new names
Jon thinks hard, does some RFCs, adds names
Jon convenes a committee: the ad hoc IAHC.
Their solution was gTLD MoU. Document anyone could sign. At some point, the theory was, it would become soup. It called for a policy counsel, more registrars, etc.
Then there was the moment of constitutional crisis. NSI was custodian of the A-root, but Jon was still the administrator.
Unhappy with NSI because it refused to make the root larger, Jon took action: he got half of the roots to take a test--take instructions from B root rather than the A root.
The US Government stepped in, in the form of Ira Magaziner. He told the NSF to back off. The DoC then formulated a statement of policy: the white paper. The paper said it's time to privatize the DNS.
IFWP v. IANA--attempts within the Internet community to rise to the challenge posed by the White Paper.
I attended the first meeting, in Reston, VA. Everyone showed up--all there to watch.
Jon Postel went on his own track, separately. He put out new bylaws for a new IANA.
We had a plan to meet at Harvard--but instead, it turned out to be NSI and Jon Postel.
So was born ICANN. They held their first meeting in Cambridge.
Terry: One of the things that were happening during this time is that the types of disputes grew:
"Typo" squatting
conflicts between competitors
conflicts between non-competitors
retailers
commercial v. non-commercial users; reverse domain name hijacking
fan sites
parody and commentary
Two main new systems to deal with this: the UDRP (Unirform Dispute Resolution Policy). How did ICANN leap over the limits of jurisdiction? Contracts.
Registrars and registries must force people to agree to this regime. This enable the policy to be global in its reach.
Ostensibly, the UDRP governs only abusive registrations and use. "Legitimate interest" and "bad faith." These terms are by no means crystal clear.
The big advantage of the UDRP is an expeditious procedure, highly favorable to the complainant. Complainant picks the forum, respondents must respond quickly. Post decision--the domain name is immediately transferred.
Rates of disputes are declining--perhaps because the majority of disputed names may have been registered during the dotcom boom. This may signal the end of the domain "Gold Rush."
WIPO most popular forum--most friendly to complainant.
Second development in domain name law, in the US: the Anticybersquatting Consumer Protection Act (ACPA is the UDRP on steroids).
ACPA has a "safe harbor" provision. More powerful remedies for injury than the UDRP. Courts willing to work hard in favor of plaintiffs.
JZ: Lessons learned. Qs--does ICANN matter? What will happen next?
Penultimate plan for ICANN: no at-large membership. Then the Boston Working Group got angry. So it was grudgingly included.
As it stands now: membership has been excised.
Do you feel outraged by this?
[...missed a bit here...]
Why care about ICANN?
It acts like a government. It can affect your life if your domain name is taken away.
The NSI answer--who is on the other side of my contract?
The historical accident answer
How upset would you be if:
Google delists your site;
MS refuses to cut a deal w/you;
MS announces that Windows will have GUIDs; and
Your high-speed Internet provider says you can't have high speed Internet anymore.
Should we really be more concerned w/ICANN as a "governing body" than we are about these situations?

And with that, it's a wrap.