Lican Huang wrote:
2^128 addresses may be not used all. But I am doubtful of " A more
realistic estimate of address usage would be 100 * earth's population.
" . There are many public equipments with IP addresses in the future,
may be in the street, or any public spaces.
However, this is not problem. If domain names is just combinations of
characters, the domain names may never exhausted forever. My argue is
"giving a new registered domain name some social meaning known by lots
of other people other than just by the owner himself would be better,
because domain name needs some additional value for the corresponding
IP address".
You are overlooking one important fact:
The domain name system is hierarchical, with for all practical terms, no
hard limit on the depth of the tree.
Your argument seems to be constructed on the incorrect assumption of a
flat namespace.
For example, there exists today, instances of geographic naming schemes
in some geographic areas, such as:
.ca (canada)
.on.ca (ontario)
<cityname>.on.ca
And for each of these, it is possible to get a domain name of
appropriate scope. This way large companies get .ca, while small local
businesses get per-city names.
This is also done under the ".gov" domain.
This ability is limited only by the willingness of local communities to
build similar trees in their portion of the domain space, and to
administer the tree through whatever means they choose.
So, the problem you are trying to describe, has already been solved, and
indeed was solved by the first RFC for DNS.
Brian Dickson
_______________________________________________
DNSOP mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/dnsop
