At 03:28 AM 4/20/2003 +0700, Robert Elz wrote:
Date: Sat, 19 Apr 2003 15:13:29 -0400
From: Keith Moore <moore(_at_)cs(_dot_)utk(_dot_)edu>
Message-ID:
<20030419151329(_dot_)66daf0de(_dot_)moore(_at_)cs(_dot_)utk(_dot_)edu>
| > No, it isn't. It is a cleaned up replacement for 1918 addresses.
| which by itself is reason enough to kill it.
Nothing of the kind. 1918 addresses were created because there was
demonstrated demand for stable local use only addressing. Nothing
has changed in the Internet to cause that demand to go away.
We either provide a mechanism, or the users provide one of their own.
As John Klensin pointed out on this same list several weeks ago
(and I'm sure he said it better than I will), the decision to
use ambiguous local addressing in IPv4 (i.e. RFC 1918 addresses)
was partially motivated by the desire to conserve IPv4 address
space. In IPv6, we don't have an address space shortage, so there
is no reason to introduce architectural complexity to conserve
address space.
Margaret