Hi Kurtis,
On Tue, 25 Jul 2006 08:09:56 +0200
Kurt Erik Lindqvist <kurtis@kurtis.pp.se> wrote:
On 16 jul 2006, at 00.58, Tony Hain wrote:
Kurt Erik Lindqvist wrote:
I think that over those 500+ years we might even have solved routing.
In the mean time even for new encap types you will not see more than
1-5 networks in a home.
You might be right. OTOH, imagine if your end up being wrong ! Do we
really want to have to go through a subnet renumbering event for the
most likely largest number of networks in the world - all residential
and small businesses, say 5 or 10 years from now ? If we've got enough
IPv6 address space such that we can avoid creating this potential
problem in the future, and we certainly do have enough address space,
then why not future proof the end-site addressing allocations against
this ? The cost of avoiding this problem is virtually nothing.
Along those lines, I'm curious what you (and other people who seem to
be against /48s for end sites) think of the "excessive" 46 bits of
address space that ethernet uses, when the reality is that no more
than
12 bits of address space would probably have been plenty for the
even the biggest LAN segments (I've seen one sadly) ? Bare in mind
that
that addressing size decision was made around 1980, when even LAN
segments with 4000 devices would have been inconceivable, so even 12
bit addressing at the time would have seemed beyond "excessive".