Re: DNS under o=Internet

[Einar Stefferud <[e-mail]>]

> By the way, why don't we start a mail campaign to convince the
> presidency of ISOC that we need this registration?

I think it is premature to start a specific purpose campaign, but I
will agree that the topic should be put on some agenda for discussion.

Discussion in the ISOC list is OK, but I think it might better be
deliberated in OSI-DS till we can decide what we really want.

	Sorceror's Apprentice Prayer: "May I want my every wish!"

There are several ideas for things like this to be done.  I certainly
think that the ISOC or someone who can act on behalf of what we call
"The Internet" should act to acquire an ICD (International Data Code?)
from BSI (which is the registration agent under ISO authority for the
{ iso(1) identified-organization(3) } arc.

The Internet is already using OID values under previously assigned ICD
values assigned to the US DoD.  These are used for SNMP MIB OID
values.  I believe is it { 1 3 6 ? ? ? }.

OIW has been assigned {1 3 14 } and has assigned values under this arc
to its SIGs, such as {1 3 14 13 } for the Registration Authority SIG.

EWOS has obtained {1 3 18 }, and AOW is also has an ICD for OID
registration of AOW Technical Objects.

US GOSIP is using an ICD { 1 3 5 } (I think?) for NSAP assignments.
OSINET has been assigned { 1 3 4 } and uses it for OSINET NSAP
assignments.

So, there is good precedent for ISOC to do this on behalf of The
Internet.  They can request it as a truly international organization
for OID registration of technical objects, and for NSAP assignments,
or to graft the entire IP address tree under { 1 3 n m } where "n" is
assigned to ISOC by BSi and ISOC then assigns "m" to the IP address
tree.

Note however that this does nothing to change the way IP addresses are
used in implementations or "listed" in the DNS.  It only provides an
internationally standard way to unambiguously express an IP address as
an OID -- e.g.,	{ 1 3 n m 128 195 1 1 } for ICS.UCI.EDU (128.195.1.1).

Yeah!  Too bad this does not solve the IP address space shortage;-).

However, all this does nothing for AlphaForm registration of
o=INTERNET or O=ISOC because the ICD arc is not permitted to delegate
registration (sub)authority for registration of AlphaForm Names for
use in X.500 RDN AVAs.  Only the { joint-iso-ccitt(2) country(16) }
arc is so empowered (by ISO 9834).

BTW, I think it would also make sense for the ISOC to register all the
IANA number assignment lists/trees and the RFC series under an ISOC
ICD, which will in effect make every RFC into an OID subauthority, and
allow all IANA number assignments to be expressed as globally unique
ISO/CCITT conformant OID values.  If you think about this, you will
see that it can help to unambiguously cite object definitions in
implementations.  And, it is a painless way to graft the whole
Internet IANA registration tree into the international OID tree.

This would not inflict any changes on any implementations, because
they already use the assigned numbers as required, and the only change
is that an "assumed prefix" will be assigned to each number series.
The assumed prefix need only appear in IANA documentation.

Ain't Registration Fun?  Best...\Stef