Home / Blogs

Interview with Avri Doria on the History of Community gTLDs

This article is published on CircleID by Jacob Malthouse on behalf of the Community gTLD Applicant Group (CTAG).

Community gTLDs play an interesting and even unique role in the ICANN new gTLD process. They reflect the community-driven nature of the Internet. Indeed the story of how Community gTLDs came about is a fascinating example of the how bottom-up process can give rise to innovative policy outcomes.

It has been over six years since the community gTLD concept was first discussed. In the mists of time, it’s easy to forget the deep foundations upon which this concept is based.

This February, Avri Doria joined The CTAG for a discussion and reflection on the role, history and background of community-based gTLDs. A summary of that discussion follows.

* * *

Q. What is your background?

A. I have spent most of my career working on Internet issues as a technologist. I was attracted to ICANN after participating in the Working Group on Internet Governance. One of the structures we reviewed as part of this work on multi-stakeholder governance was ICANN. The Nominating Committee brought me into the Generic Names Supporting Organisation (GNSO) council. I was on the GNSO council for five years and was chair for two and a half years. This period, coincidentally, covered the time when the new gTLD recommendations were made to ICANN from the GNSO.

Q. How long have you been involved with the Internet Community?

A. I have been involved with the Internet since before there was an Internet community. I worked on protocols starting the eighties and attended Internet Engineering Task Force (IETF) meetings until recently when ICANN began to take up my time. I still participate in the IETF on mailing lists and hope to get back there someday. I’ve been directly involved in various parts of Internet policy for a very long time. My last real job, until the bubble burst, was as Director of Advanced IP Routing at Nortel. Since then I’ve been doing technical research part time and research and consulting in Internet governance part-time.

Q. What have your key roles been within the Internet Community?

A. Protocol developer and policy analyst; I work as a systems architect, whether it’s protocols or governance structures. These are the things I focus on and that interest me the most; the technical and governance systems of the Internet and how they fit together.

Q. What are you up to now?

A. At the moment I am working part-time for dotGay LLC, on their policies and community information. I also teach Internet governance in a few places, and am doing some research on Delay and Disruption Tolerant Networking (DTN) when I can find the funding; something I have been researching for ten years now.

Q. How did the ‘community-based’ TLD concept arise?

A. I came to ICANN in 2005, and the ‘sponsored’ gTLD policy making was over by then. All that was left was a few of the finalizations and the .xxx saga, which of course continued until last year.

In looking at it now, the sponsored gTLD concept was certainly part of the history that we looked at in terms of designing community gTLDs for this program and was part of the whole notion of how the community gTLD concept evolved. Those who had worked on sponsored gTLDs were part of the discussion in developing the current new gTLD recommendations.

The concept was part of the overall discussions that the GNSO was having. We were doing it as a committee of the whole—that is all of the GNSO council members were involved in the process. There was the notion that we have to defend communities that may want a gTLD. This encompassed both preventing gTLDs from being ‘grabbed’, but it also involved engaging more communities—a broader notion of support—that would help spread awareness further about the possibilities of a community using its name for a gTLD. We almost always put something up against the ‘community’ test when we were discussing policy (e.g., how would this work in the case of .bank?).

It’s very important for the Internet community to go back to the policy recommendations that formed this program*. It’s what we are rooted in.

* The ICANN Generic Names Supporting Organisation Final Report on the Introduction of New Generic Top-Level Domains was released on 8 August 2007. It is available for download in two parts here as: Part A and Part B (PDF).

One of the recommendations was about substantial opposition from a significant portion of the community. Implementation guidelines F, H & P explain how one follows and understands the support of the community. What is defined there is a very broad notion of community. It was the recommendation of the GNSO that ‘community’ should be interpreted broadly and include cultural, economic and other communities. The recommendations are quite specific about what community meant in the ICANN sense. For example, recommendation H—community contention for a name—calls out the guidelines and definitions and P explains how the objection and application and evaluation all use the same notion of community that is explicit in the recommendations.

Indeed, one of the things we learned from the sponsored gTLD round is that we needed to be a little broader in our definition of community. That is reflected in the GNSO’s report.

Q. Who was involved in that process?

A. I think everyone—there were people who brought it up in terms of the .bank notion, it was one of the favourite examples. The GNSO looked at banking and the need to protect it from being used in an abusive manner.

Another example that was often raised was .Cherokee (e.g., minority that is also a brand). I used to have some involvement in theatre, so looked at the cost of a .Cherokee application for Ford Motor Company as costing less than the catering lunch on a commercial shoot. We knew that brands were going to come in and we talked about Ford grabbing Cherokee to put on the sides of buses, so we wanted to protect the Cherokee nation. We also looked at the example of Lakota as a community that isn’t associated with a particular brand.

We engaged a range of voices from people who thought community gTLDs were good, to people who thought community gTLDs were bad, to people who thought that free speech would be a victim of community objections. Everyone engaged in the discussion and many stakeholders had different views. Eventually we came to ICANN consensus on encouraging and protecting communities.

Q. Where there any challenges in developing the concept?

A. Yes, there was a whole range of issues. We came up with questions such as, “If it’s a community but I have a trademark on it, then who has rights?” Potentially both of them could preclude the other through an objection process. If you have a community and trademark you can try to stop a non-community bid through the objection process. There was general acceptance—rough consensus—that we could never create explicit lists of things. Any kind of controls had to be objection based. The world is too big and broad for policy to say “this is the list” that’s why objections figure in everywhere as an alternative to lists. Everyone should be able to see the names that have been applied for and objected to. Our view was: no explicit lists and no expanding the reserved names list. That was explicit—if anything, people wanted to remove names from the reserved list.

Another challenge we faced was—what’s the final step? Auction or lottery or ICANN Board evaluation? We were basically split on this issue and we all sent our opinions to the Board. Auction as a contention resolution solution was stronger with those who have deep pockets or who believe the market can solve all problems. In the end we left it to the Board and ICANN on how to resolve contention after community priority and such. We did spend a few years talking about it and getting feedback, so it was a thoughtful process, but we could not reach consensus.

We also went through a very strict process with ICANN staff, going through several readings and going through exercises with them. We would say “we think we would do it this way” and staff would respond with comments. It was a really interactive process in terms of coming up with our recommendations, and it took a year or two longer than many would have liked.

When I got to the GNSO, we started working on this. I was chair when we approved the recommendations and I’m now chairing the At-large group that is reviewing the process, so it’s been quite an undertaking.

Q. What is your impression of the community applications?

A. I was disappointed in the number of community applications overall. I was hoping for more from around the world, especially from cultural and linguistic communities. I see this as part of the whole issue with the expense and style of outreach of the application round to other parts of the world. For example, we got assistance for applicants so late in the process that there was no time for outreach. By the time we told them about the money they would have been working on it with the assumption that there is no money. Or rather, they would have long since given up the idea because there was no money.

The At-large working group is looking at the failure of that pre-application round outreach and also at the failure on the community applicant front.

ICANN didn’t make it easy. There was no specific outreach to communities. Many of us in the GNSO thought it would be a good thing but some in ICANN think communities are a bad thing—that people are cheating when they claim to be a community—but if you read the guidelines they were meant to be broad.

You are community because you say you are—and you only need to prove it if challenged by a standard application and/or an objection. If you apply for a standard string, you are implicitly saying there is no community that needs to be served. A standard application for a community’s name is the same as telling a community that they do not exist, or at least that they do not matter. The other way to attack communities is the direct objection.

Q. Would you make any changes for the next round?

A. Review is one of the requirements of the program. We knew we could not figure everything out a priori, that the process was going to teach us. I certainly believe that we have to—have a remedial round—to pick up communities from other parts of the world.

We failed on community, on diversity, on international scope. Most of the IDN applications are the same old incumbents just getting an IDN. That failed and needs to be fixed.

Regarding the community test, I am of two minds. I think the testing idea is good, but I think the Community Priority Evaluation test is flawed. ICANN has improved its way of testing since then. For example, the qualification test it created for the outreach aid was richer—you might be this or that kind of community, the test had different ways to meet the threshold. It was still points based, but the way you built up your points to reach the required threshold was not quite as punitive.

As it stands, communities have to prove it the hard way under trial by ordeal, rather than starting from a notion of trust. It’s “You’re gaming, how do we prove you are innocent by putting you through an ordeal?” We don’t need to wait for things to go further to think that this emphasis is wrong, This notion that communities are not to be trusted isn’t right.

It’s the “hard cases make bad law” syndrome—we can find methods of catching gamers without ordeals—the questions for the applicant support program were more nuanced and included clauses to catch gamers, so its encouraging to see that some learning has been done already.

I believe one of the primary reasons for this gTLD round was communities, around the world, cultural, linguistic, etc. This program has failed at that. We will certainly learn from this program how to allow for more categories of gTLDs—more people wanted this. What categories have been developed, what’s special about them? Brand, community, geographic, etc., not all are community but many of them touch on the concept in a lot of places.

Q. What are you most looking forward to in watching these new community gTLDs?

A. 1,000 flowers blooming. I’m looking for many different kinds of communities finding cool ways to express their identities and creating a safe, useful and meaningful environment for themselves. Each one of them should somehow develop differently, following their own logic for the kind of community it is. I’m hoping that the communities manage to make it through the ICANN community priority gauntlet.

We will see how communities develop these things. The .xxx and IFFOR (International Foundation for Online Responsibility) process is just a start for what we’ll learn as we watch communities try to create a self-regulated environment for their stakeholder groups. Some of those new community gTLDs should be the most beautiful blooms.

Q. What risks do you see for community gTLDs?

A. Many of them have done hand waving about how they’ll really be able to implement and enforce their commitments. It won’t be clear how many of them are really doing their policy work until they’ve won. How can they give metrics to ICANN to serve the process in a bottom-up manner? What are the ‘success’ metrics for community-based gTLDs?

Living up to ICANN commitments and expectations about how a community gTLD should function will be harder than most are imagining. Living up to evolving community requirements will be even more so.

By Jacob Malthouse, Fmr UNEP Staffer, Ex ICANN VP, Co-founder dot-eco domain registry.

Filed Under


Comment Title:

  Notify me of follow-up comments

We encourage you to post comments and engage in discussions that advance this post through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can report it using the link at the end of each comment. Views expressed in the comments do not represent those of CircleID. For more information on our comment policy, see Codes of Conduct.

CircleID Newsletter The Weekly Wrap

More and more professionals are choosing to publish critical posts on CircleID from all corners of the Internet industry. If you find it hard to keep up daily, consider subscribing to our weekly digest. We will provide you a convenient summary report once a week sent directly to your inbox. It's a quick and easy read.

I make a point of reading CircleID. There is no getting around the utility of knowing what thoughtful people are thinking and saying about our industry.

Co-designer of the TCP/IP Protocols & the Architecture of the Internet



Brand Protection

Sponsored byCSC


Sponsored byVerisign


Sponsored byDNIB.com

Domain Names

Sponsored byVerisign

Threat Intelligence

Sponsored byWhoisXML API

IPv4 Markets

Sponsored byIPv4.Global

New TLDs

Sponsored byRadix