Re: [sig-policy] prop-091: Limiting of final /8 policy to specific /9
On Jan 24, 2011, at 7:21 PM, Matthew Moyle-Croft wrote:
>
> On 25/01/2011, at 12:28 PM, David Woodgate wrote:
>
> ...
>
>>
>> So this is another issue I have with the current final /8 policy; I
>> think it's a reasonable idea to have enough addresses for specific
>> purposes for a couple of years while the transition to v6 gets under
>> way properly, but if either it promotes a lengthy availability of
>> IPv4 for new content servers (thus potentially removing IPv6 adoption
>> drivers and increasing IPv4 address value)
>
> I don't think the "supply" side (Content) is or is going to be the issue. The issue is the eyeball/access side. Reducing the space to a /8 to allow the eyeball side to continue to avoid the IPv6 issue does more harm. The issue the content providers have with turning IPv6 on is the concern about the 0.05% of people with broken IPv6 connectivity due to broken CPE. If the eyeball side gets on with providing working IPv6 connectivity then the content side seems willing to turn it on quickly.
>
The /9 is a piddly drop in the bucket for the access side. It won't change the
runout date for access by any significant amount of time. There's no significant
harm here in terms of allowing access to continue avoiding the issue.
> The IPv6 day coming up shows that a majority of content is capable of turning IPv6 on - by the time you have Google, Yahoo, Akamai doing IPv6 then a big percentage of an eyeball's content will be IPv6 capable. Yet, in my country, only one of the top 10 providers appears to be interested in dual stack to the eyeballs. Indeed, the largest 4 have negligible or no IPv6 deployment.
>
Um, no, it shows that a collection of content providers are capable of
turning IPv6 on. So far, of the Alexa top 10, 5 are planning to participate
in IPv6 day. When you can show 100 of the top 100 or 1000 of the
top 10,000 you start to approach the "majority of content".
>> , or it causes an explosion
>> of requests artificially tailored to bypass the intent of the policy
>> (thus potentially causing an inappropriate distribution of addresses
>> anyway, with unnecessary hostmaster workload and impacts upon routing
>> tables, etc.), then I feel it may do more harm than good for the industry.
>
> Maybe prop-91 should give way to a proposal to tackle that issue? Reducing /8 to /9 doesn't seem to be a plan or a solution to that problem. It appears to be a way of extending IPv4 at the expensive of transition later.
>
I'm not seeing what transition you think is somehow enabled by addresses
held on a shelf where nobody can reach them. Can you please explain
what aspect of transition is improved by this mechanism?
>>
>> So the question in this context is what the correct balance in size
>> of space to reserve? The current policy suggests a /8; prop-091
>> suggests another view.
>
> The issues for me are:
>
> Changing the amount of space, without regard to how it's fairly allocated is an issue for me with prop-91.
>
It would be fairly allocated just like all the other IPv4 space not subject to this relatively new special
policy. It would be fairly allocated just like the majority of IPv4 space. That how is a solved problem.
> Changing the amount of space based on past behaviour without considering how behaviour of APNIC members and new members will change over the next few years is an issue with prop-91.
>
Considered... The existing policy will force them to behave in a manner contrary to policy if we
don't amend it. I would rather see policy address the needs of the community to the extent that
it can and avoid creating a situation where the community has to work around or against policy
out of necessity whenever possible.
> Mean that prop-91 isn't supportable in it's current form.
>
Sure it is... I support prop-91 in its current form.
Owen