Site Map

Please note:

You are viewing archival ICANN material. Links and information may be outdated or incorrect. Visit ICANN's main website for current information.

ICANN Yokohama Meeting Topic - Introduction of New Top-Level Domains: Expression of Interest #3

Posted: 9 July 2000


The working groups effort reported in the document for the preparation of the Yokohama meeting is based upon the current use of the DNS, because no one with a competitive vision of new services based on an innovative use of the DNS will want to disclose it in an unprotected manner.

This does not help identifying comming new needs and may lead to wrong decisions with a dramatical impact where simple no cost solutions and rules decided today could save the Internet, commerce and industry billions of dollars in reducing future complexity and allowing new projects to develop.

I happen to have such a need, to be able to document it and its magnitude. Also to have some real experience on public international addressing plan management and commercial impact and a realistic and simple proposition to match my generic need.

General formulation of the need

The overlooked need is very easy to understand. Today the hierarchy in addressing is handled through levels: 1st level : tld, 2nd level domain name, 3rd level, subadressing.

There is a used and growing alternative to 3rd level and even further down for subadressing, it is the semantic of the second level.

First example: let suppose a company (there is one) offering area code related portals. They may use either or formats. There are several good reasons to use the second one:

- is more compact and easier to memorize

- is easier to recall than - there are 300 area codes, which means a subscription of $ 6000 while buying back the code "" would probably be 50 times more.

- it easier to franchise since the webs sites are independent

- e-mail service for each domain name may be better managed and protected but most of all this is necessary when the different sites are not managed on machines in the same place (growth, access protection, distributed commercial agreements, local services, international localion, etc..).

Second example: let suppose I have a concept for a new service named "abc" I may apply to different thematic web sites and I wish to impose a commercial image with the semantic "" :,,"... Will I have to sue everyone picking an name and wanting to sell it back to me with a big premium?

Third example (mine): I currently develop a simple and robust automation VPN management technology. The target is to install, operate and supervise large networks of automated boxes through the Internet. The boxes may range from web servers, to advertising screens in shops, traffic lights, building appliances, alarms, industrial maintenance, private PABX, franchised sites, local e-commerce and services, etc.. etc.. All of them needing at least one DN per machine. If I sell a traffic light management system to the city of London the need will be to reserve from http://londoncitytrafficlight00000.tld to http://londoncitytrafficlight99999.tld and to register a few of them initially.

From these examples one sees that the need is eventually for hundreds of millions of domain names which cannot be registered on an individual basis but have to be reserved by billions on a semantic basis.

One also see that 3rd level subaddressing my give less lengthy adresses and decrease the need for special domain names, decreasing the number of DN actually needed.

Limitations, costs and non legal protection of the standard solution

The existing today solutions address the need of registrants with one or a few domain names (being accepted that 'DN investors' are not realy a need!).

If I take my own case, I own 500 domain names and I look for many others for my own operations according to a semantic which is a trade property I cannot protect. Famous names are usually backed by large organizations with resources and staff to protect their rights. I have none. This makes my DN management and tracking a complex and unduly over costly operation.

I have no protection against 'investors' or naming conflicts resulting from uncoordinated plans with others. While these people and lack of organisation will cost me far more than any spammers or hackers.

Opening new gTLDs or ccTLDs with the same rules may or may not help existing common needs, but it will help neither me nor my customers and operations like mine. While--if not coordinated with a solution for us--it will increase our difficulties, decrease the public recognition of our ideas and globally cost our operations a lot more to achieve less. Considering that many e-commerce, franchises, communities, developers will share my needs, not to address the domain name semantic issue would actually represent a huge loss in money, innovation, and simplicity of use. It would probably be at this stage of the internet development a real mistake. It would fight an international self coordinated development of the network and of its usage..

A proposition

The target is to permit and protect innovative and simplifying ways to address virtual systems through the semantic of their domain names. To that end I am interested:

- to setup or to share in the set-up of a non-profit organization under the control of the ICANN for a cooperative management of a dedicated gTLD.

- to have this gTLD assigned a clear meaning to anyone and to be chosen in order to have a good potential commercial impact and public recognition.

- to possiblt use the gTLD ".sys"

- to restrict the use of the ".sys" TLD to projects having already subscribed more than 100 domain names in gTLD or ccTLD according to a semantic rule which will serve as a reference in case of conflict (example "area100.sys" may conflict with foot100.sys from a system "xxxx100.sys" dedicated to the leading 100st). This rule will be described in way to Unix wild cards or database format descriptors.

- to use the rule of the first ordered first served for the reservation of formats with the provision every registrant will be informed and in case of conflict of interest (not of trade mark) the earliest date of the first 100th domain name registration in other TLDs will designate the legitimate owner (here we deal with innovation)

- to create a working group to study how the DNS system could at some stage handle formats of domain names rather than domain names to decrease its processing load.

- to establish this non-profit organization as a cooperative operation reserved to registrants paying a yearly contribution in proportion to their turn-over and to the size of the reserved semantic (characters and position). The collected money will go to the management of the organization and to the promotion of the ".sys" gTLD. The Members will obtain their domain names at the ICANN rate plus a small markup. There will be NO transfer of formats without a charged or notarized letter and a delay of 8 days. All management will use crypted mails.

I think I match the criteria of competence to establish and manage such an operation. I plan to proceed and to rise the necessary funding in three steps:

- phase 1: to call upon all the organizations desiring to participate (they are protected by the rule of the 100th domain name registration) with a initial assembly.

- phase 2: to register Members, elect a board, collect Memberships and proceed within two months to the necessary development and organization.

- phase 3: to start operations

I would rise an initial funding estimated to $10.000 to prepare and held the initail meeting and to write the contractual terms, with an international law firm and the ICANN.

Discussion of the proposition

This propostion has no impact on the current ICANN discussions but it remove a key, urgent and not discussed issue which would otherwise interfer with the general assembly.

It has to be addressed now:

- because the impact on the public awareness will be optimum - to avoid any dillution

- to defeat the 'domain investors' who consider the concerned projects as priviledged preys. There will not be 'domain investors' in ".sys": They would need to invest 100 domain names with a concept before anyone else. They would then become real innovators selling their own idea.

- to allow new technologies and e-commerce developments based upon semantic format registration which will be in demand in the comming months (from our prospects).

There is no risk of confusion. The ".sys" TLD must be general and not contrry specific since many projects will span around the world (there are chances than most of the multinational projects use it). It corresponds to a new need and brings a simple adhoc response: only concerned organizations will want to use it.

There is no risk of initial rush. Registrants will first having furst to develop a project.

The proposed rules are simple initial rules the users will be able to adapt through a democratic process. An creation assembly will be held where everyone will be able to propose, discuss and vote.

There is no new special load on the DNS system (the addresses will be needed anyway). The specific developments in registration, accounting, control are more simplifications than anything else. It will be necessarily reserved to professionnals with a lesser demand for support. An existing organisation approved by the ICANN could be retained to handle its operations.

This means that the opening of the ".sys" gTLD under such terms will have no more impact on the internet operations than the opening of the ".arpa" TLD. Experience from this opening will be seeked.

It is likely that this gTLD gathers hundreds of formats this year representing the reservation of millions of domain names, with formats like "my_company-country.sys" to complete "my_company.ccTLD" and the "", all of them having their implies a central management some national market resent.

Organisation of the proposition

This is only a contribution, eventhough I think I have experience, interest and competence to chair or advise the initial effort. This document is not a final and documented statement. If needed I look for assistance in discussing and writing it according rules and procedures of the ICANN I ignore.

Jefsey Morfin

© Internet Corporation for Assigned Names and Numbers