New.net Distends Domain Dynasty

The fiesty alternative domain name registrar oppugns ICANN's insolence toward determining new top level domains. But ISPs hold the key to making potential TLDs universally resolvable. Could a single-minded ISP collective make ICANN capitulate?

by Julie Wheeler [November 14, 2001]

Though no one really expected it to survive in the beginning, New.net has now been around for a year. If you haven't heard of New.net, it is the company that launched its "alternative" top-level domains last November. It debuted directly after the Internet Corporation for Assigned Names and Numbers (ICANN) made its own decisions on which top-level domains should be introduced into the market.

ICANN, the DNS governing body, chose seven; New.net has, to date, chosen thirty. Of the company's launch, and its timing, David Hernand, New.net's CEO, had this to say: "It is clear that the [ICANN] staff and board are waking up to the fact that if they keep moving so slowly, market forces will take over." And that seems to be exactly his plan.

But if you haven't heard of New.net, the company will be most displeased. That's because, if it is not able to enlist the support of ISPs, it will not be universally resolvable, as it is not part of the A Root, which is what is conventionally thought of as the database that houses the Internet.

New.net relies on ISPs to make minor changes in files accompanying BIND that allow users' browsers to see New.net names. These changes are at the point of the ISP, though interested users may download plug-ins that add the trailer "new.net" after any top-level domain (TLD) that is not recognized as one of ICANN's stock.

ICANN's not very happy about this new development, either, as it considers a single, authoritative root an absolute necessity to maintain the stability of the Internet. While most alternative root schemes die a slow and painful death—or spend their years flopping around just outside the Internet, and outside the public's radar—New.net has garnered a fair amount of interest from some large ISPs, including EarthLink, and some ICANN-accredited registrars, most notably BulkRegister.

In fact, in June after ICANN's meeting in Stockholm, Sweden, where New.net made a big, splashy and controversial appearance. Joe Sims, independent council for ICANN, said, "They came in with a business model and financing, unlike some of the others, so they have the potential for more impact." However, he also made the claim of the company that either it would survive, or ICANN would. "There is no way to work it out; there is no middle ground."

Parlez vous DNS?
But it seems like it's full steam ahead for New.net—ahead and abroad, that is. Multilingual domain names have been a topic of some interest within ICANN, as well as the Internet Engineering Task Force (IETF), for some time. After all, even if the Internet was born in America, it has grown up and begun to travel the world. But it seems translating the DNS is more complicated than expected ... depending on whom you ask.

In October, New.net launched its foreign language services, full on with a slue of in-language "top-level domains." That's something ICANN has failed to do, instead getting all tangled up in the nightmare of actually making the cultural intricacies of language jive with the technical complexity of the domain name system.

It seems New.net is conducting its own multilingual, and multi cultural, experiment, without ICANN's consent. The newest additions to this experiment include German and Italian. Of the services, New.net claims, "Most of the new extensions correspond in meaning and relevance with English, Spanish, French and Portuguese extensions released thus far by New.net, but appeal to the unique cultures and local communities represented by each of the new languages that New.net serves."

Dot-come-ons?
New.net's extensions in English, anyway, range anywhere from .family and .kids to .xxx and everything in between. Of course, you can visit New.net's website to check out its full list of extensions. They are designed to appeal to the public, and to attract the most registrations possible—along with registration fees. For example, the company's most recent introduction of .movies fits into its theme.

Have you ever been searching for the website on the newest movie, only to find that the most logical guesses, such as the title of the movie, all lead to dead ends? This is the sales pitch New.net is using for .movies. And though I'm sure that's true, I'm not certain that it's a serious problem facing the world today. Should .movies warrant its own top-level domain? Will it be followed by .anything, .everything, and .nothing? When will they decide enough is enough?

Dot-co-operation
More importantly, when will ICANN decide enough is enough and approve a registry to launch a TLD that New.net is already in the process of selling? Hernand doesn't think that's going to happen. "We think ICANN would be foolish to release new names that conflict with New.net." Though he didn't mention it, it seems clear that the inevitable legal hassle would be enough to prohibit such a move on ICANN's part. But the organization might just decide it's worth it to firmly establish its authority in managing the DNS.

The strange thing is that it seems everything would likely be fine if ICANN didn't approve any TLDs in the same spaces as New.net's ventures. According to David Conrad, CTO of Nominum, the company that authored BIND 9, the system will work fine if there is cooperation. "Moving a single authoritative root to a cooperative root is not a problem. The problem is when they cannot agree on who should have the TLDs." So that makes this problem one of economics and of power.

And that, of course is the problem. Who has the authority to release new top-level domains? ICANN? The U.S. Government? Any company with the funding and the determination?

Though still an unanswered question, you can check out the opposing views on the subject. Hernand wrote a white paper called A Proposal to Introduce Market-Based Principles into Domain Name Governance on the virtues of New.net's system, as well as an argument on the acceptability of multiple roots [PDF].

In response, Stuart Lynn, ICANN's president and CEO, wrote a discussion paper called A Unique, Authoritative Root for the DNS, countering New.net's claim to legitimacy, and citing several RFCs and other sources as support.

What is the correct answer? You will have to be the judge of that, as it seems this dueling DNS debate will continue for some time.

—End