[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Question about past advice to RIRs
On Tue, Aug 08, 2006 at 11:15:32AM +0200, Brian E Carpenter wrote:
Note that RFCs 1265, 1787, 2008, 2827, and 3704 are included for the
perspectives they present whether currently applicable or not. The
biggest problems seem to arise from RFCs 1881 and 1887; if they are
obsolete, I have an entire section that might disappear.
7.2. Informative References
[RFC1881] Internet Architecture Board and Internet Engineering
Steering Group, "IPv6 Address Allocation Management",
RFC 1881, December 1995.
[RFC1887] Rekhter, Y. and T. Li, "An Architecture for IPv6 Unicast
Address Allocation", RFC 1887, December 1995.
note that both of these documents predate most of the RIR existance,
when the IETF had a direct role in defining allocation
in todays environment, i would consider both of these documents to be
historical in nature, e.g. not applicable to the current RIR models.
it is useful to note that RFC 1887 is labled "An Architecture..."
which is a clear indication that the authors did not preclude other
valid architectural models.
You're overlooking that RFC 1881 was and remains the basis on which the
IPv6 address space was formally delegated to IANA by the community that
created it. There's no sense in which that delegation is historical.
Certainly we have learned a thing or two since 1995.
not overlooking it. but am not sure that RFC 1881 allows for
the delegation to be revolked. Surely your not impuning that the
IETF could reclaim IPv6 from the IANA and its subsiquent delegations
to RIR's? Are you?
It does say that IANA can revoke allocations if a registry does
something really, really bad - but I agree, any wholesale revocation
would be, let's say, unlikely.
On a point of fact, the RIPE-NCC was established in 1992. It was certainly
my assumption in 1995 that IANA would further delegate prefixes to
the registries, and that the registries would remain active in the
IETF, so that address assignment practices would inform and be informed
by IETF discussions. That still seems to be happening, fortunately.
As I indicated earlier, MOST of the RIRs came AFTER these RFCs
came into existance. And as far as I am aware, there is no special
"RIR" class of participation in the IETF, since IETF activity is
by individual. Certainly the RIR policy process is based on
individual participation. Statements from the IETF about how they
think address assignment practices should occur are, on occasion,
are taken by indivduals and crafted into policy proposals. But then
the proposals are from individuals, not the IETF.
Well, they may, or may not, become statements of IETF consensus. But
my point wasn't about formalism - I think we all win if RIR people
are active in IETF discussions, and we all lose if they aren't.