Margaret Wasserman wrote:
The WG chairs of the IPv6 WG did determine that there was
consensus of those in the room to deprecate site-local
addressing in IPv6. Like all consensus achieved at IETF
meetings, this consensus will be checked on the list.
BTW, I was at the meeting (Tony was not) and do not agree
with his characterization of the discussion. The
conversation did not focus on NAT and, IMO, was no more
confused than typical WG discussions :-).
Unfortunately I could not be there, so I can only go on the reports I
have gotten, which show that at least some of the people in the room
felt this was an anti-NAT discussion. Some have even suggested that
there was no clear indication on what 'deprecating site-local' means. Is
it the address range, or the concept of scopes?
All the arguments seem to be targeting getting rid of the address range
because a few app developers can't figure out how to deal with scopes.
The range doesn't create the scope, the application of filtering does
that. Since there will be filtering going on anyway, are we supposed to
get rid of all addresses? Let's get real here. All SL amounts to is a
well-known route filter.
History shows people will use private address space for a variety of
reasons. Getting rid of a published range for that purpose will only
mean they use whatever random numbers they can find. This has also been
shown to create operational problems, so we need to give them the tool
they want to use in a way we can contain the fallout. Site local is
defined to do that job, and we do not have WG concensus on depricating
At 05:16 PM 3/25/2003 -0800, Tony Hain wrote:
Lloyd Wood wrote:
spaking of which, Noel, where's the obIPv6 whinge on how,
they've deprecated site-local addresses, IPv6 has no redeeming
I don't know who 'they' is, but the WG has not deprecated site-local
addresses. As I understand it a completely confused
in SF which equated SL with NAT, but that does not equate to a WG
consensus to depricate something people are already using.