RE: stable addressing

["Fleischman, Eric" <eric.fleischman@boeing.com>]

>If an organisation wishes to continue with NAT, it might as well stick with
>IPv4 though?   What's the gain then from having IPv6?

Tim,

This is an excellent question. An organization may indeed strive to continue using IPv4, as I stated would probably be the case in RFC 1687 (see http://www.ietf.org/rfc/rfc1687.txt). However, there are many factors that also war against that position such as:

1) Apple has made IPv6 be its default deployment setting. Microsoft plans to do so in a future O/S release. Other vendors are expressing similar intentions. $$ is an important factor in influencing networking decisions.

2) Some governments (e.g., USA's DoD) are apparently enamored with IPv6. Should they eventually require that IPv6 be used to communicate with them, then those needing to communicate with them will comply.

3) As I mentioned in RFC 1687, should a "killer app" be deployed requiring IPv6, then the Fortune 1000 *will* deploy it. Having said this, I personally believe that virtually all of today's articulated advantages of IPv6 over IPv4 are either theoretically-true-but-not-operationally-true or else pure marketing hype. However, should this change, then the whole deployment algorithm changes. There is a lot of positive potential within IPv6 that hopefully will be leveraged (e.g., QoS, IPSec, the issues the Multi6 WG are addressing).

4) Dual stacks are a pain. It pains users, it pains ISPs, and it pains vendors. There are many reasons why the world as a whole will seek to simplify its computer and network deployments. If IPv6 flies then the world will probably standardize upon it. If it doesn't, then a modified IPv4 will probably remain king. It is currently too early to know how things will turn out. When I imagine what each alternative would probably look like in 2020, I hope that IPv6 will pan out.

--Eric