Internationalized Domain Names (IDN) Committee

Discussion Paper on Non-ASCII TLD Policy Issues

Summary of Public Feedback

[ General Comments & Responses | Specific Comments & Responses | Structured Question Comments ]

NOTE: This page quotes excerpts from comments submitted via email in response to
the IDN Committee's Discussion Paper on Non-ASCII TLD Policy Issues.
The quoted excerpts are presented in a structured manner, are meant to be illustrative,
and are not necessarily full quotations. A revised version of the paper is available here.


Response from Kilnam Chon

"sudden introduction of a massive number of new TLDs would be a bad idea"

I don't agree with this statement.  We introduced the massive number of ccTLDs in mid 1980s, but did not have any serious problems.  The important issue is "how to introduce TLDs".

Response from Danny Younger

    “a new TLD should only be introduced if user demand can be demonstrated to exist.”
I question whether in fact we have the metrics to gauge the degree of user demand for products that are not yet on the market, and further whether it is  up to ICANN to become the barometer of the market.  Clearly there was no pressing "user" demand for .museum (as compared to what-seemed-to-be a market demand for .web), and yet that TLD was introduced without such a constraining principle on a proof-of-concept basis.

Why not allow for the marketing of any such IDN string by any responsible registry operation that is willing to take the risk, and thereby allow the market to determine if there is sufficient user demand to sustain the operation.  Some registries may indeed fail as user demand might not materialize... but this is what proof-of-concept is all about.

Response from Mr. S. Maniam, Chairman WG03 of the International Forum for IT in Tamil (INFITT) and Chairman MINC Tamil Language Working Group, Approved by Executive Director, INFITT.

“TLD expansion should occur in a careful and controlled fashion, with regard for the overall stability of the DNS”

    Limiting TLD expansion with the unfounded fear of affecting the overall stability of the DNS has over the years, been often invoked as a stockphrase that has not been backed by experimental data. Using this as a basic premise to slow down the implementation and deployment of a new technology and a new namespace would be unacceptable to the Tamil speaking community.

”(T)he sudden introduction of a massive number of new TLDs would be a bad idea”

Re-using this oft-quoted premise is generally a bad idea as the history of ICANN has shown, and merely reinforces the misconception that ICANN is a bureaucracy unable to respond quickly enough to the pace of Internet growth and innovation. It is our opinion that ICANN be seen as a proactive organization, rather than have an international organization tarred by this kind of obscurantist and unnecessary phraseology.

”(A) new TLD can only be created if there is a willing and able registry operator to run it”

This principle is generally sound but safeguards should be instituted to prevent this principle from being abused as an excuse to shut out new namespaces particular in those places where a language namespace may be associated with developing countries, where the standards and yardsticks used to assess willingness and ability to perform registry operations are those of developed countries, and thus become disproportionate and not commensurate with the stage of evolution of Internet in these countries and communities.

”(D)ifferent categories of new TLDs may require different contractual, policy, and selection frameworks”

While this principle may be true generally speaking, it belies the underlying assumption and premise that is fundamentally flawed because that was not how the Internet became such an unprecedented success in the history of mankind.

By having an open framework, the unfettered growth of the Internet has been unprecedented. Anyone who wanted to set up a network could do so by connection to a service provider, without having authorities to evaluate contractual, policy or worse, selection criteria and frameworks.

Basing the future process of the non-ASCII domain name space on such a counter-Internet culture is detrimental to the growth of the Internet in non-ASCII using communities and places unnecessary strictures on Internet accessibility in the name of care, control and stability. Instead, one strong recommendation would be the lightest touch of policy and regulation, so as to facilitate growth.

”(T)he selection process should be transparent, allowing key stakeholders to participate in it.”

As the ICANN's recent history has demonstrated, the inability to properly define stakeholders generally leads to ineffectual execution of well intentioned policies. Transparency is sound and desirable. Essentially, participation in language or script-specific name spaces should be left to the community of language/script users to decide for themselves. Non-language speakers/users should not interfere with other language/user groups and allow them self-determination. The market forces and the evolutionary pressures will do the rest.

”(T)he core purpose of introducing non-ASCII TLDs would be to make the DNS service easier to use for Internet users whose native languages include non-ASCII characters”

This principle is sound, fair and magnanimous and should be actively pursued for a level playing field for language and script using communities world wide, so as to build an Internet for all peoples everywhere.

”(A) new TLD should only be introduced if user demand can be demonstrated to exist.”

This is an unnecessary bureaucratic requirement which is superfluous. Creation and destruction of domain names on the Internet are currently dynamic and fluid. Depending on usage, the survival of the domain names obeys the laws of nature and the market forces of demand and supply, that is not easily pre-determined or pre-demonstrated.

The creation of so-called non-ASCII TLDs should also not be artificially constricted by arbitrary concepts of user demand which do  not have technical or engineering bases.

If the Tamil speaking community would like to have Tamil TLDs, and the technology allows for it, this should be facilitated whether or not there is user demand.

The inability to define "user demand" opens it to abuse. Is one user sufficient user demand, or ten thousand or ten million? Or is it relative user demand that this principle is based on? If so, relative to what? the population of China? or the population of Mauritius? Once this principle is invoked, there is no objective yardstick and if adopted, merely opens up more complexity, confusion and chaos.


Response from Stefan Probst

They differ now already between a TLD and say a 3rd-level-domain. I see no reason, why the requirements for e.g. the ".com" registry have to be the same [stringent] like the ones for a small language community in a developing country, which is accustomed to non-reliable infrastructure.

Also operating cost differ. I do not believe that ICANN/IANA is able to judge the financial ability of a registry in a developing country.


Response from Stefan Probst

In addition to that resource requirements, there is also the issue of global equivalences. Just two examples:

- While there might be a fairly clear definition for the .coop in the West, who knows, what kind of cooperatives exist in other political systems? Who defines what qualifies, and what not?

- The .name is based on a Western naming system with family name and surname, where both names can be arranged in either way. There are surely many cultures, where this does not fit. Why should they not be allowed to have their own TLD for their individual-name-based domains?

Again, there should be no automatic linkage between non-ASCII TLDs with the ASCII versions in your category of sponsored and unsponsored gTLDs.


Response from James Seng

I do not believe in any incumbent preference for gTLD or ccTLD or for the government. However, I recognized that they are likely candidate to submit an application and also likely to have the community support for the TLD.

Response from Stefan Probst

I oppose therefore any regulation, procedure, etc. which would give "rich" corporations and organizations in the West any advantage over "poor" ones in the developing world.

This applies especially, but not exclusively, to any linkage between non-ASCII TLDs with the ASCII versions in your category of sponsored and unsponsored gTLDs.


Response from James Seng

Divided by either geographic, language or culture is possibly a bad idea. There is no single well define rule to draw that line. For example, have a TLD for every geographic is easy but it only capture the country level TLD.It does not capture the other generic use or for languages which have no country (e.g. Tamil, Mongolian). having a TLD for every language or culture is also problematic as it does not capture the ccTLD, or at least in the "normal" way. And by creating and delegating a TLD to every language/culture, the problem also exist which group to delegate to and also the legitimacy of the organization, and their process. Delegating only creates multiple mini-ICANN, and worst, one which may not adopts the similar open process.

I suggest looking into a generic creation process. For instance, an application for any TLD would be up for open comments for a period of time (3/6/12 months) and from the feedbacks, an appointed group would make a rough consensus call. The key word is _rough_ (in IETF terms that is around70-80%), because it is unlikely any proposal would receive 100% consensus.

A well-defined (e.g. narrowly defined) TLD is therefore likely/easier to get through this process vs. a more generic TLD. It is both an advantage and disadvantage of such a system.


Response from James Seng

The lack of ISO-3166 for non-ASCII TLD is both a problem for ccTLD and gTLD. And there is not just one firm "translation", even within a single language. Therefore, it is better to consider a generic creation policy for all non-ASCII TLD (regardless of ccTLD or gTLD) rather relying on ISO.

Depending on the TLD requested, the vested interest parties may be within a specific region of a specific language, specific to a country or across multiples locale. This is much like the ccTLD and gTLD interest, one which covers a specific country the other worldwide. In IDN case, there is an additional twist of language/script consideration.

I foresee that the applicant of the non-ASCII TLD would have to submit these (sic) information and demonstrate support from the relevant locale.

All players, ccTLD, government, non-profit, commercial and at-large have a vested interest in the TLDs. Therefore, there should not be any special incentive (or roles given) to any of the vested parties. All comments should be treated equal, regardless whether it comes from a ccTLD, a government, non-profit, commercial or at-large.

Comments concerning the layout, construction and functionality of this site
should be sent to webmaster@icann.org .

Page Updated 26-Jun-2002
©2002  The Internet Corporation for Assigned Names and Numbers. All rights reserved.