Re: Why do we need to go with 128 bits address space ?

["John Levine" <[e-mail]>]

In article <[e-mail]> you write:
>-=-=-=-=-=-
>I just wonder, over two decades ago when the discussions happened, whether the question that
>was asked most was " Why do we need to go with 128 bits address space if whatever is been
>trying to achieve with the existing approach of IPv6, can be achieved by 64 32bits (IPv4)
>address space as well?"

The short answer to your question is "No".  That was not the most
asked question.  The mailing lists are all archived at the IETF, and
you can go back and read the documents if you want.  There was a lot
of debate about what address size to use, and whether to used a fixed
or variable address size.

There are things we do with 128 bit addresses that would be difficult
to do with 64 bits.  As others have noted, the large addresses make it
easy to set up network addressing and not have to worry about making
each network's chunk of addresses big enough, or running out of bits
to handle the local network topology inside a single organization's
allocation.  It's really nice that every network is a /64, and we
don't have to fool around with CIDR boundaries.

There are plenty of things wrong with IPv6; it repeats the IPv4 design
of assigning addresses to interfaces rather than hosts, the plan that
the host part of an address (the low 64 bits) would be based on an
interface's MAC address was misguided, the transition from IPv4 was
and is very painful, and packet fragmentation works badly and not at
all in multicast networks.  But the address size wasn't a mistake.

R's,
John