RE: [ga] November 13 GA meeting

[Jefsey Morfin <jefsey@wanadoo.fr>]

On 18:22 01/10/01, Roeland Meyer said:
>|> - short term solution: the end of the root concept and a
>|>   more stable and open resolver system for all as per the
>|>   1978 international public service naming plan.
>
>Could you please mention URL's and document listing when you mention these
>sorts of things? Are you refering to X.500?

Ho! Do you want me to give you a reference for the Independence Bell or the 
Tea Party? I sent a mail on the GA a few months ago quoting all the people 
involved in the naming birth. I will explain again, as this is History and 
also the proof St. Lynn's Good Book is apocryphal.


I am referring to the international public service naming plan in operation 
at every international public data service I stabilized by mid 1978 as (in 
our today words) :
- registry (ISO 3166 - But I chose the Radio 3 letters list or the private 
registry Name)
- domain (name of the site, company, service)
- sub domains (applications, machines, etc).
I think the first two where "itaforsales" and "gerfordsales".

The naming plan supported tens of thousands of public users in 55 countries 
over more than 10 years. This is history. I sold it for the first time as 
such at the IBI (UN specialized agency) meeting in Torremolinos in 1979. We 
switched very progressively from 1982 to the X.121 addressing plan to 
downgrade to the X.75 new standard and I tested numbering for fax and phone 
in 1985. Also the access to Minitel with Intelmatique.

I do not recall for sure when ARPA interconnected. We discussed that a 
short while ago with Louis Pouzin who developed the Cyclades network at 
INRIA (he introduced the end to end datagram and the zones - and permitted 
the TCP/IP to develop [Elisabeth can document that a lot]). ARPANet 
interconnected with the international naming plan in 1984 if I recall - if 
Vint is reading he might remember. I remember we had problems with the 
interface due their IP address structure. I had to approve the solution for 
compatibility: I found absurd to go by IP addresses and proposed to match 
the IP address with international names. But it was not compatible with 
Telenet, making two plans to support.

I quitted the international public services by mid 1986. I understand they 
eventually came to the naming concept but reversing my Latin sequence (it 
made sense once you used the "."  to still match the EDI standard and be 
nearer from the Postal Naming Plan).

At that time b-trees were the big thing in town if you recall. They were 
also easy database. To  get people understand and pay, you had to use a 
tree. But if you look at Postel document he uses ".U" (Universal) as the 
empty root of the tree.

What I can tell you is the network naming plan started with public 
operators (Tymnet, US IRCs, France Telecom, BT (GPO), Telephonica, KDD, 
Televerket, RadioAustra, Italcable, C&W, etc...) plus the European Space 
Agency, HP, Dupont de Nemours, Philips... and international. No one ever 
though that the USA could be the root of the world!!  We had to wait for 
Stuart Lynn to discover it was from every eternity.

The naming plan is not hierarchical, it is sequential. The TLD is not a 
upper level - except in the Bind data base if they want Bind to 
be  DataBase - it is the prefix (now the suffix) of the registry.
Period. Now I understand that to explain the way Bind works the database 
limited analogy helped them.

  In 1978  our boss at Tymnet was Bill Combs, then it was Bob Harcharcik 
who then created MCI Mail and hired Vint and Don Heath. People important 
were Bob Trehin who open the first international services, Jack McDonnell 
who lead most of the US side of the development before creating TNS Inc., 
Warren Prince. Development leaders where LaRoy Times, Joe Rinde, Mike Rude. 
The management of the entire plan was Vida Stafford and Dominique Marchand, 
under Neil Sullivan's management. Bob McCormick developed most of the 
liaisons with US Islands (VI was connected very early), relations with the 
US  IRC. All this was discussed at the ISIS Clubs, the equivalent of the 
ICANN meetings 20 years ago, with bi-yearly meetings on the European and 
then on the Asian side (Australia, Japan, HK, Indonesia, Guam, etc...).

Nothing new with the ICANN, except to create a single point of weakness 
from a distrinuted architecture while we had technically and politically 
succeeded to fully distribute and make resillient and war proof a 
centralized technology.











--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html