Orbit

Toshi,

Thank you for your note.

In response to the point you raise, I observe that what we have learned with IPv4, and have learned from other related technologies that if we start designing a successor technology at that point in time when scarcity is already a factor in the current technology, then the successor design will naturally tend to focus on that aspect which is scarce, rather than attempting to design a technology which is better/faster/cheaper/more useful/.

I would put forward the proposition that if we embark on IPv6 with an address plan that will work across our anticipated lifecycle for Ipv6, but will encounter address exhaustion at that point then we have failed in the IPv6 address plan, as we would not have learned one essential lesson from the IPv4 experience. What would be a far better approach in the view of the authors of this proposal, and one that underlies this proposal, is to be able to say with some confidence that addresses will still be abundant at the anticipated decline point in IPv6 technology, and that lack of IPv6 addresses will _not_ be a reason why we will move on from IPv6.

We should not be in the business of built in obsolescence, and certainly not if we can buy additional time without undue pain. We've looked at the HD ratio and the subnet boundary as potential points of variation in the IPv6 address plan that could admit more efficient utilization without substantial alteration to the overall IPv6 architecture and without undue need to alter existing equipment, software or current deployments, such as they are today. What we buy back is a greater level of assurance that IPv6 and its address plan can readily encompass even the most optimistic of expectations of IPv6 utility and deployment.

regards,

Geoff

At 11:44 AM 1/09/2005, Toshiyuki Hosaka wrote:

Dear Geoff,

Thank you for your proposal submitted. Please allow me to ask you one point on behalf of JP community, that is "How many years should IPv6 lifetime be?".

There was a comment in JPNIC Open Policy Meeting in July this year, that he doesn't think there is any problem because *your* forecast shows IPv6 will live for 100-120 years, that would be enough lifetime (from his view).

According to your forecast referred here, and presented in the URL below, a total of 1/2 of the available IPv6 address space would still be unused after 60 years from now, even in the worst (most consumed) case.

<URL> http://www.ripe.net/ripe/meetings/ripe-50/presentations/ripe50-plenary-wed-ipv6-roundtable-report.pdf

Furthermore he pointed out that Paul (Wilson: APNIC DG) had commented IPv6 should be live for at least 50 years time, which has no inconsistency with your forecast.

<URL> http://www.apnic.net/docs/apster/issues/apster4-200208.pdf (page 8)

If I summarize his view, that would be "Current situation is within the scope of original IPv6 distribution design because we will have IPv6 for 100-120 years. So what is the problem?".

How do you respond? I appreciate your comments.

thanks and best regards, Toshi

-------- Original Message -------- From: Geoff Huston gih@apnic.net To: sig-policy@lists.apnic.net Subject: [sig-policy] IPv6 Policy Proposal - prop-030-v001 Date: 2005/8/11 10:26

...

Attached are text, pdf and word versions of a IPv6 policy proposal for consideration at APNIC-20 regards, Geoff Huston Stephan Millet

<snip..>