The Top Level Domain Evaluation Project

Jonathan Zittrain* and Benjamin Edelman**
Berkman Center for Internet & Society
Harvard Law School


One of ICANN's primary tasks since its 1998 inception has been to break a de facto deadlock over the introduction of new top level domains for the Internet's default domain name system -- the DNS. The deadlock existed because various stakeholders had widely divergent views about the merits of expanding the top-level namespace -- and widely varying interests that would be affected by any such expansion. Many holders of famous marks opposed expanding the namespace, since it would entail that much more policing of names they wanted; at the very least, they opposed expansion without some means of assuring their ability to stake a claim to new names apart from first-come, first-served. Others favored expansion either to dilute the near-monopoly in public domain registration (and attendant high registration fees) enjoyed by Network Solutions, which was the registry for names in .COM, .NET, and .ORG, or because more TLDs meant greater availability of mnemonic labels for Web sites and other Internet content.

Arguments for and against expansion that relied on predicting the behavior of registrants and registries were largely conjectural, since apart from the opening of certain ccTLD registries to public registrations, there had been no expansion of the namespace in years.

As ICANN now rolls out new TLDs amidst a variety of models -- some generic open, like .INFO, some limited in intended registrants, like .MUSEUM, others limited in purpose, like .NAME, the authors have embarked on a continuing project to collect and analyze data about their growth and usage. We seek to go beyond theoretical arguments relying on anecdote or assumption about what is and is not taking place -- an approach perhaps especially helpful in an environment in which ICANN policy decisions typically entail significant economic consequences for stakeholders. Accordingly, we seek to inform our analysis with data on the actual documentable usage of specific domains, and when practical we provide our raw data for inspection so that others can verify our results and continue discussion on their own.

Key among questions of interest is the level of use of new domains: To the extent that a TLD is put to use for the distribution of Web content, especially content not available elsewhere, that domain is likely to be evaluated positively. Correspondingly, all else equal, when two TLDs of a given size are compared, the more favorable evaluation is likely to go to that TLD with the greater number of functional registered domain names. To the extent that cybersquatting, domain warehousing, and defensive registrations can be identified, these phenomena are also likely to be associated with somewhat less favorable evaluations of a new TLD.

In this project, we begin to survey usage of all gTLDs and, over time, major ccTLDs. Our research seeks to combine qualitative discussion of motivations, perspectives, and policies with quantitative data and analysis to document specific actual behavior.

The purpose of this work is primarily academic -- to document the activity at issue for the benefit of those who seek to help shape policy decisions on related matters. We aim to be helpful to consumers, providers, scholars, and regulators. For example, consumers may take from our research a sense of the registrars and registries with which they wish to do business. Providers may examine the successes and shortcomings of others as well as, in some instances, themselves. Finally, ICANN is evaluating and will continue to evaluate the rollout and effectiveness of its New TLD Program, and other potential regulators of the activities at issue may also find these resources helpful in the actions they contemplate.

All data provided in our projects is intended for private non-commercial use, and our data is not provided for commercial use or redistribution of any kind. Please contact the authors for question about appropriate and acceptable usage of our work.

Our recent work on domain names is linked from the following table of contents.



* Jack N. and Lillian R. Berkman Assistant Professor of Entrepreneurial Legal Studies, Harvard Law School.
** J.D. Candidate, Harvard Law School, 2005.

Last Updated: October 6, 2003 - Sign up for notification of updates and related work.