ICANN should consider this Alternative Top Level Domains model: .ALT

Amid all of the writing about ICANN’s obviously flawed model for top level domains, I have had a little bit of an “AHA!” insight that I want to share with interested parties.

It is so simple, I cannot believe that no one has thought of this before — and in case someone has, then I cannot believe that I have not heard about it yet… because it is (IMO) such a powerful model.

First of all, let me explain the raison d’etre for my view of the whole “top level domain” discussion. In my view, there are two parts to this. The first one — the reason often mentioned in discussions — is the (purported) lack of available of domain names. The second one is actually an altogether different issue — and the one I hope to focus on particularly below: the lack of available of alternative points of view.

Because the first point has been elaborated ad infinitum elsewhere — and also because I consider it to be largely a value judgement — I will simply take it as given, more or less,… or simply say that it is, for me, a moot point. If you wish to argue this point — pro or con — then you should take that argument elsewhere, because I simply don’t care to argue the point either way.

The second issue is far more complicated to explain — but once this is done (and I simply hope I can do that adequately), it ought to be rather simple and straightforward to resolve.

As a preface, let me note that there has been a tradition in the DNS by which two-letter top level domains were referred to as “country code” domains. Several years ago, I noted that the introduction of .EU broke with this tradition — since the European Union is not a country per se. Nonetheless, as far as I know, it can still be maintained that all two-letter top level domains represent some kind of “government” — although some of these governments no longer exist, others may be under water financially, and yet others may soon no longer have any territory above sea level. All of this preface is merely an attempt to demarcate what has traditionally been considered “geographically localized” domains from domains which are referred to as “generic” (and I would consider “sponsored” domains to be — semantically speaking — a subset of generic domain names). In my opinion, the most significant improvement that could be made regarding the entire domain name system (DNS) would be to clarify this classification of domain names into “generic” (i.e., global) versus “non-generic” (i.e. local) domain names — such that users could more easily understand that when they see a domain name ending in “.TV” that the laws of a small-island country in the South Pacific apply to all information passing though this top level domain.

On the generic side, the number of domains number anywhere from about half a dozen to several dozens — depending on how many are considered to be within the “generic” domain name set. A complicating factor in this regard is that some of the top level domains which I would place into the generic domain names category have quite stringent limitations on their use. This is mostly an odd historical quirk — and I would say that most people who are well-versed on the topic of domain names would admit that these top level domains are by and large failed experiments. As such, I would suggest that such restrictions should be abolished and/or these domain names should be considered as mistakes — and we should simply tolerate them as such, much in the same manner as we accept that other mistakes have been made in the history of mankind — and we should simply move on and not dwell on their existence as oddities that stand out and annoy us.

Let me — for the sake of argument — simply suppose that there are about a dozen generic/global domain names, and that all other domain names are governmental/localized domain names. Also — just for the sake of simplicity — let me propose that the alternative domain name system would also be generic/global in nature — to simplify the hypothetical explanation: consider a generic/global domain name system with a dozen top level domains plus a 13th “alternative” domain name system.

The alternative domain name system would only be used in case one specific string has been registered in all of the “regular” generic top level domains.

Therefore, the alternative domain name system would normally be used to provide a space for alternative viewpoints regarding a particular topic. So, for example, if the string “hotels” (or “brandname”) were registered in all of the generic top level domains, then the top level domain “.hotels” could be created as an alternative top level domain, and it would then be listed in the .alt registry and/or on the .alt homepage.

To register such an alternative domain name, a person would need to have registered some other domain name, with valid contact details — and the person / company would need to declare that they are not affiliated with the registrant of the associated TLD string in any other generic top level domain. If the registrant could not be reached at the contact address (e.g. within 24 hours or within 7 business days), the domain name registration would be immediately cancelled and the registrant would be fined and/or barred from registering another alternative domain name.

This alternative domain name system would, in other words, be significantly different from the “regular” domain name system — most importantly: it would be a single, centralized database.

Another difference would be that the content in the alternative domain name system would not be monitored in any way. Since the registrant could be contacted at any point in time, the validity of the information posted would be entirely the responsibility of the registrant. In order to discourage reckless behavior (and also to cover operating costs), registry updates would require additional processing fees.

The precise configuration of such an alternative domain name system is not the issue here… — nor, for that matter, the question of whether there should be one dozen, or two dozen or even hundreds of generic top level domains. This alternative proposal is intended merely as an “escape valve” to release some of the pressure surrounding the arguments for and against ICANN’s proposed top level domain expansion ideas. As such, I propose that the financing for such a project should come from the present top level domain registry companies, and perhaps these companies should also jointly own a majority share of the alternative registry. In this way, the corporation could operate in a way that might be cost-effective for all parties… and should there be a need or demand for additional “regular” generic top level domains, then this would further dilute the cost to each of the generic top level domain registries.

I don’t know — I’m thinking there must be something I am obviously missing… because otherwise this solution seems so naively simple. The only “problematical” case I can think of is the case in which an alternative registry is set up at time T, and then at time T+x, a regular top level domain becomes available… but even that scenario seems relatively unproblematical to me: the alternative top level domain could simply be continued as long as there are more registrants than there are available generic top level domains for the string in question.

Am I missing something else that is blatantly obvious?!?

Domains Nominations Post Forum