Secretary's Notice | 19 July 2001

At its meeting in Melbourne on 13 March 2001, the ICANN Board adopted the following resolutions:

RESOLVED [01.24] that the President and General Counsel are authorized and requested to complete negotiation of the remaining unsponsored top-level domain appendices as soon as feasible and to post the resulting appendices on the ICANN web site, along with any minor corrections or adjustments to the base agreement and appendices as already posted;

RESOLVED [01.25] that the Board shall be notified of the complete posting of the agreement and appendices for any of the four unsponsored top-level domains (.biz, .info, .name, and .pro) and after that notification seven days shall be allowed for Board members to make any additional comments to the President and General Counsel;

RESOLVED [01.26] that in the absence of the request of any Board member to the contrary based on policy considerations, the President is authorized to sign the posted agreements after the conclusion of those seven days; and

RESOLVED [01.27] that upon signature of the agreements the President is authorized to take such actions, including causing reports to be made to the United States Department of Commerce, as appropriate to implement the agreements.

On 3 July 2001, the posting of the agreement and appendices for the .name unsponsored TLD was completed. On 10 July 2001, an objection on policy grounds was received from ICANN Director Amadeu Abril i Abril. A statement of that objection appears below.

Louis Touton
ICANN Secretary

Note: Global Name Registry submitted a response to Mr. Abril's objection on 26 July 2001.


Subject: .name agreements. Policy Objection

Stuart, Louis and all,

As I mentioned in Stockholm I have a serious policy issue with these .name agreements. It refers to what is now called SLD Mail-Forwarding service. As described in Louis' report:

  • E-mail forwarding service. Domain-name registrations are made only at the third level in this TLD. To make second-level-domain e-mail addresses such as "helmut@schink.name" available, the Registry Operator will delegate all second-level domains to itself and offer an e-mail forwarding service. Because this is a service that the registry operator can exclusively provide, a competitive environment is introduced using the registrar-registry model.

This is both an unnecesary and unacceptable new monopoly service and something that was not approved by the Board last November.

This means that anyone registering under .name may have a delegated string such as amadeu.abril.name, which (this is critical) may be serviced by any registrar, any ISP in the largest sense, any connectivity provider, any anything. The registry only maintains the "database" (from the customer point of view). From that point of view, the registry guarantees global resolvability; any other service is provided by whoever the registrant deems appropriate because of price, service, bundled offerings, reliabitlity, location, or any other consideration.

If we look closer to the so-called E-mail Forwarding Service we see that it is REALLY different. On the "base" registration service Andy Mueller has the chance to register, say, andy.mueller.name, while Milton Mueller may register milton.mueller.name. None of them can actually register mueller.name as such. If we come to the E-mail Forwarding Service, Andy is offered andy@mueller.name and Milton is offered milton@mueller.name Both MUST use for their email purposes, if they choose the service, a SLD that is not delegated to them but reserved by . . . the Registry itself.

The implications are clear: both Andy and Milton WILL BE CONSTRAINED to use the same email provider, the same email servers: the registry's. No choice on their hands. No domain name portability at all. Oh, yes, this is disguised as a service offered by each registrar . . . and this only will add to the confusion (I contract with, say, Nominalia or register.com a service that in fact is provided by a third party, described in the fine print indeed, but that nobody will understand . . . ). Registrars, because of their economic incentives to market that product, will help to the general confusion.

I don't know if you realise what is going on here: the whole email exchange load will be NECESSARILY circulating through the registry's servers. This is certainly bizarre in itself. And dangerous if you think about the security, control, censorhip, data protection, spamming, data mining implications . . . . . . . . .

What's more, we are not talking about a marginal service here: email is the KILLER app for personal TLDs much more than web-based or ftp-based uses of a domain name . . . . . . So we are turning the real HEART of the domain into an unneeded monopoly, departing for no good reason from our repeated objectives of establishing a competitive market wherever possible within domain registration services, making sure domain-name portability and ensuring that the DNS is managed as a public trust.

Now, we can go to the next step and check the usual argument: we are not a consumer-protection body. Correct. But we are not overseeing market behavior, here. We are just SETTING the market itself, defining the rules of the game . . . and we should avoid setting rules that are, in themselves, necessarily, unavoidably, anticompetitive, no matter how the "market players" behave afterwards . . . . The .name registry is also proposing some added services regarding hosting and other services that raise serious concern from the antitrust perspective. But what they do there is "not" our responsibility: we are not granting them a monopoly regarding hosting services. And we do have no responsibility and therefore no control regarding their commercial behaviour as such. But we do have complete responsibility for the services we "create" ourselves (the new TLD) and the way we, ICANN, decide it will be managed. And we do have full responsibility for the allocation of a completely unnecessary "legal" monopolistic position that arises not from market forces, but stems simply from our own "policy decision".

A separate issue is that in the global namespace there is not an available protection for users. I mean that .fr users, most of them French people, may rely in the French legal system, and even the political one, to control what happens within .fr and around its registry . . . . But only UK, at most EU residents could dream of having any real "legal remedy" against future abuses of such a dominant position . . .  if any. And, once again, remember that we are experimentg with a personal domain name space here . . . .

If we think about the current discussions involving new.net, the recurrent mantra is "what you do is OK, except that we all want and are committed to a DNS that is A) universally resolvable and B) works independently of the hardware, software, connectivity, registrar, service provider . . .  you choose." The .name proposed SLD Email Forwarding Service does not affect global resolvability, indeed, but it does not respect the independence of service provider principle . . . .

Furthermore is an unacceptable distortion of what was submitted and selected. The base for .name is that it has a 3LD.SLD.name structure, where the 3LD is delegated to the registrant (and therefore she is free to service it through the registrars and ISPs of her choice). This service is in fact moving the VERY CORE of that domain (email for personal use . . . !!!!!) to the SLD directly, and controlled (from service to traffic and content) by a single company. Isn't this a significant change????? Just have a look of the examples they were giving in their application:


There will be no restrictions on how a name is articulated and, for example, a person with the name Steve Martin Anderson could register any of the following variations. Of course, these are only examples and the possibility of choice need not be limited to these:

1) Steve.Anderson.name
2) Martin.Anderson.name
3) Steve.Martin.name
4) Steve.Nickname.name
5) Anderson.Steve.name
6) Anderson.Martin.name
7) S.Anderson.name
8) SM.Anderson.name
9) Steve.MartinAnderson.name
10) Martin.SteveAnderson.name

This, in turn, would lead to the possiblity for this individual to have any of the following examples:

1) mail@Steve.Anderson.name
2) Steve@Martin.Anderson.name
3) Anderson@Steve.Martin.name
4) Martin@Anderson.Steve.name
5) private@Anderson.Martin.name
6) email@S.Anderson.name
7) mail@SM.Anderson.name
8) mail@Steve.MartinAnderson.name
9) private@Martin.SteveAnderson.name


Well, you see: all email examples follow the orthodox submission: two dots at least . . . .

Indeed they mention, but just as a "future development" the SLD email forwarding service. The reason why this is mentioned only as a future possibility is clear: there had been a lot of discussion about this issue and the general feeling was that this could not fly under the current technical constraints where this services REQUIRE a single, monopoly, ISP/mail server service provider. If this changes, and someone may devise a truly "shared" solution, then this would be a great addition. If you want more details on this, you could check also CORE's .nom application, where the exact issue is discussed. And, I do insist, this was clear to everybody, even Global Name registry (so they only mentioned this as "possible future development").

I hope that we all see the difference between say Microsoft marketing hotmail and related amadeuabril@hotmail.com addresses and an ICANN-selected Registry doing the same as an established and ICANN-blessed monopoly for a new TLD, the first for personal use, doing the same for andy@mueller.name. A registry is a natural monoply. But this also means that we have to be three times more careful allowing them to run additional monoply services related to the "natural" one that watching private companies doing similar (but thoroughly distinct and much less relevant) services in a competitive environment . . . .

Louis told me that, as this is a monopoly service, it will be regulated, ie, price will be fixed by contract with ICANN. But price to registrars or even to end users is not at all the only question at stake, nor the more relevant.

To sum up, please consider carefully what is being proposed here: we create a new TLD under the concept that SLD will be "shared" among registrants and only 3LD will be delegated to them. In order to make sure this works, we say that SLD will be "reserved" by the Registry itself (so it can guarantee the sharing among registrants acting as a neutral party in order to manage a resource for the public trust). Then we allow an apparently trivial mail forwarding service . . . which in fact means that the SLDs previously in hands of the Registry as a "neutral party" guaranteeing sharing among registrants turn into a private resource of a single party–the Registry. We have moved from a conception where names were serviced by a registry and delegated to individual users who enjoyed domain name portability, to a proposal where the registry acted as a neutral party facilitating sharing among a number of registrants to a system where ALL NAMES ARE IN FACT DELEGATED to the Registry, not in order to be offered for delegation to users, but as a monopolisitc (by our own contractual terms, not by any other reason) asset of a single entity for their commercial purposes. So long for shared SLDs, domain name portability, and management of the DNS as a public trust in the interest of the whole community :-(

I therefore formally request the Board to instruct Stuart NOT to sign the .name agreement until such time that the so-called SLD email forwarding services is deleted from it, or at least delayed until such times as there is a technical solution that does not prescribe centralized service provision and email channeling through a single provider, be it the Registy or any other.

Amadeu

(typos corrected)