Re: Support for 240/4 and 0/8 addresses in NetBSD

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

>>> Personally, I think that this whole excercise is a waste of time
>>> (because "on the wild Internet", 240.* has no chance to realiably
>>> work *to all destinations*, due to too many old routers being
>>> somewhere in the path)
>> Today?  Yes.
>> Tomorrow?  Next year, even?  Yes.
>> A decade from now?  Not so much.
> I sincerely hope we overcome IPv4 by then...

That would be nice.

I have low-to-negative expectation that's how it will play out.

> running dual-stack is a major pain,

It is?  I wonder what's different between us, because I find it quite
the opposite.  I've been running dual-stack on all of my machines which
have publicly-routed IPs at all for about 20 years, and I find it not
only non-painful but actually helpful.  (I spent the second half of
2002 working at a job in Norway which set me up with a home netlink
which was v6-only.  I got a sharp lesson in how many of my tools were
not v6-ready.  By the time I came back home six months later, that was
almost entirely fixed.)

> So getting rid of IPv4 should be the plan for the next 20 years, not
> "invest time to make it last 6 months longer".

If that's what it were, I think I might agree with you.  But running
out of addresses will not cause IPv4 to suddenly become useless; if it
did, as I understand it v4 would have died quite some time ago.

What freeing up 240/4 - and, to a lesser extent, 0/8 and 127.x/16 for
x!=0 - _will_ do is allow v4 to support about eight percent more hosts.
(I think dereserving the all-0 host address in each subnet might be an
even bigger win in some senses, but that's a separate discussion.)

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B