Search RPD Archives
Limit search to: Subject & Body Subject Author
Sort by:

[AfriNIC-rpd] Updated Version of the "IPv4 Soft Landing Policy"now Available Online

Owen DeLong owen at delong.com
Fri Feb 25 18:29:45 UTC 2011


On Feb 25, 2011, at 10:04 AM, James Blessing wrote:

> On 25/02/2011 17:23, Owen DeLong wrote:
>> 
>> On Feb 25, 2011, at 2:04 AM, James Blessing wrote:
>>> 1. Increase the income into AfriNIC, this could be used to fund educational projects and consultancy to those networks wanting to transition but without the experience or skills.
>> You're talking about increasing the income by a very small amount in the proposal you submitted.
>> I don't think the income you would produce would provide for more than one or two educational
>> events.
> 
> Er, do you mean Andrews proposal? I did comment that he did appear to undervaluing the potential of the last /8
> 
Sorry, yes. It's been a long week.

>>> 3. Helps other regions who are trying to transition but are growing faster than they can migrate (Apnic)
>> It  really doesn't. It's such a drop in the bucket compared to the consumption rate in any one of those
>> regions, let alone their combined consumption rate that it provides almost no relief whatsoever.
> 
> For Apnic 2-3 months *might* be enough to take the edge off the impending doom of September, I know a single /8 is a mere a drop in the ocean, but a few drops of dew in the morning might just save the life of a man in the desert...
> 
ROFL... I expect APNIC to be out in June. The amount of space proposed to be provided by AfriNIC would
not last APNIC more than 2 weeks, let alone getting anywhere near 2-3 months.

>>> 4. Stops Africa becoming a dumping ground for all the hardware that isn't v6 capable. Potentially increasing the time before depletion as networks continue to use the 'cheap' discards from the other regions (including the address space that is handed back once v6 is the defacto choice)
>> That's an interesting theory. I would have expected people in Africa to realize that buying gear that
>> doesn't support IPv6 isn't in their best interest already.
> 
> Really, has the procurement strategies deployed in the rest of the world passed you by? How many networks are still buying Cisco 7200s as they're cheap? Those with the tightest budgets will always 'look for a bargain' and with other networks upgrading and trying to get anything they can for what is useless kit I can see many hardware resale opportunities *if* networks don't mandate IPv6 support. Comments about having enough IPv4 space for 5 years will make people think that non-V6 stuff is suitable...
> 
I was running IPv6 on a Cisco 7200 for quite some time. If you're going to cite examples, please cite examples
of equipment that doesn't support IPv6 that are still being widely purchased.

Frankly, I don't think that AfriNIC's IPv4 free pool will last 5 years. That's a linear projection of current
consumption and includes the following erroneous assumptions:

	1.	The policy will be 100% effective at preventing the pillaging of AfriNIC space.
		I would anticipate more like 50% effective.
	2.	Consumption in the AfriNIC service region will not accelerate (it already is accelerating).

I believe that without this policy, AfriNIC runs out this year along with everyone else. With this policy,
AfriNIC has IPv4 addresses available to facilitate IPv6 transition in a smoother way over the next
one to two, maybe three years.

>>> 5. Demonstrates the Africa is forward looking, maybe the policy needs to actually put a hard stop on the allocation of IPv4 addresses (say 1st Jan 2013) pushing networks to deploy IPv6 at a greater rate than they are today
>> Again, that's an interesting approach, but, it has nothing to do with the policy you've proposed or with your
>> proposed changes to the soft landing policy.
> 
> Its a view point, the soft landing policy appears to be trying to extend the life of v4 for as long as possible which (to me) appears to be counter intuitive.
> 
I don't see the soft landing policy as doing so in any way. I see the soft landing policy as attempting to do exactly
what its name implies. Allow AfriNIC to preserve the ability to make a smoother transition to IPv6 over the
next couple of years rather than facing the abrupt smack-into-the-wall transition that the other regions have
chosen.

>>> We (Limelight Networks) have been adding v6 capabilities to our services for years, our JNB pop had v6 added to it about 30 minutes after the final v4 addressing plan had been completed (only because I wanted a cup of coffee) but its only *now* that people are beginning to ask for v6 services to be turned on.
>>> 
>> I'm not sure I see a correlation between this and any meaningful part of the policy debate.
> 
> Quite simply its taken large amounts of press coverage to get to the point where people are thinking about turning on v6, it still a very small number and many companies are still failing to consider it. By adopting policies that try to extend the life of v4  its having the opposite effect.
> 
We can agree to disagree about this. I don't know how much IPv6 advocacy you have done. It's what I do
for a living. (http://owend.corp.he.net/events)

I think that having a policy that facilitates Africa continuing to work towards a smooth transition to IPv6,
Africa has an opportunity to learn from the mistakes of the rest of the world and gain some ground by
spending less and moving forward faster.

>>> Its true AfriNIC needs to have a expiry policy, but for one I think it should actually be more about a setting a (close) hard expiry date and getting whatever value is possible out of a scarce resource rather than trying to arrange he deck chairs to make sure they don't get wet until th last second.
>>> 
>> ROFL... I would support policy to do that. I don't think that transferring the space out of region (throwing the deck
>> chairs at the other boats to follow your analogy) is the best way to gain value.
> 
> Throwing the deck chairs at other boats so that people stop thinking about them and concentrate on either the life boat or how to get to the rescue ship...
> 
>> Remember, I am the one who said:
>> 
>> IPv4 has a TITANIC address shortage.
>> IPv6 is the rescue ship.
>> IPv6RD is a life boat.
>> NAT is a seat cushion used for flotation.
>> NAT444 is a deck chair.
>> Deck chairs don't float.
> 
> If you did congrats, I've been using it for a while...
> 
I did. I will not pretend that I am the first one to use "rearranging the
deck chairs" to refer to IPv4 policies that don't make sense for the
end-game. But I was the one who put that combination of analogies
together.

Owen




More information about the RPD mailing list