[ppml] those pesky users...

Ted Mittelstaedt tedm at ipinc.net
Tue Mar 27 17:42:37 EDT 2007

>-----Original Message-----
>From: ppml-bounces at arin.net [mailto:ppml-bounces at arin.net]On Behalf Of
>Howard, W. Lee
>Sent: Tuesday, March 27, 2007 1:51 PM
>To: ppml at arin.net
>Subject: Re: [ppml] those pesky users...
>> -----Original Message-----
>> From: Ted Mittelstaedt [mailto:tedm at ipinc.net] 
>> Sent: Tuesday, March 27, 2007 4:03 PM
>> To: Howard, W. Lee; Johnson, Ron; ppml at arin.net
>> Subject: RE: [ppml] those pesky users...
>> Lee, ARIN and the other RIR's need to admit a huge mistake 
>> was made over the IPv6 allocations and then go forward with 
>> correcting it.
>There's still a basic pronoun problem.  The RIRs do what the
>public dictates in the form of policy passed by community 
>consensus.  The exception (for ARIN, I assume for the other RIRs 
>too) is that fees are set by members.  If we have made a mistake,
>then it's "we" the "community of internet number wonks," not "we"
>the "people behind the curtain at ARIN."  

:-)  Well, I could have said "we need to admit" but I wanted people
to actually read through the trial balloon before talking about it.

Too many folks brains knee-jerk shut down when told that they
need to admit they made a mistake.  It's easier to use a straw
man, then let the people come to the conclusion that you were talking
about them, later on.

>> What you should have done was for ALL ipv4 assignments you 
>> should have AUTOMATICALLY made an IPv6 assignment of a number 
>> block.  
>That would be possible.  Should everyone get a /32 or a /48 or
>some other number?

That is a good question, and I think any proposal to do 
an allocation like I mentioned is going to have to set a figure.
I deliberatly said 1:1 because I knew it was probably not workable
and it would get some people to pipe up.

Of course it will be subject to lots of arguments, no doubt.

>> The best thing going forward would be for ARIN and the other 
>> RIR's to drop the IPv4 and IPv6 fee schedules (the wavier is 
>> a joke anyway, what is the point of an IPv6 fee schedule with 
>> a fee of $0) 
>The point was to let people know what the fee would be, someday,
>so they could plan for it.

I disagree, I think the real point was to attempt to influence
IPv6 uptake by adjusting fees.  You can argue all you want on this
point but it is silly to claim that anyone can predict what
ARIN's or anyone elses costs are going to be in the future.  Without
knowing the future costs how can you claim that you know now what
the fee is going to be, someway.  Come on, now.

In some things, fee adjustment
does cause behaviour change, but I think the existence of the
wavier proved it didn't in this case.

>> and replace it with a single IP allocation fee 
>> schedule that applies to both kinds of numbering, then for 
>Noted for future FinCom discussion, as with your earlier 
>suggestion to increase fees "parabolically." 
>> all current IPv4 holders that the numbering authorities have 
>> assigned numbering for, just go ahead and assign IPv6 
>> allocations at a 1 to 1 ratio.  (for every single IPv4 
>> address you get an IPv6 address) It's not like there's any 
>> shortage of IPv6.
>Er, sorry.  A /21 in IPv4 translates to a /117 in IPv6?  Or
>is that /53?  Should we ignore nybble boundaries?

Let me think about this and see what the responses are to the
trial balloon.  Keep in mind that adopting such a proposal
carries very far reaching implications.

If we adopt it we are essentially saying that IPv6 is the
future, and it also knocks out much of the justificaton for
IPv4 reclamation efforts.

But also on the dark since, while it will immediately and
drastically increase the IPv6 uptake, it also is going to
really stretch out the IPv4 rundown.

Look at it this way.  We all agree to say that next year, bang -
everyone has IPv6.

So that means from that date, IPv4 by definition, is going to
be in a "rundown" state.

>> Then what you do is get rid of the separate IPv4 and IPv6 
>> numbering requests and replace them with a single numbering 
>> request that is used for both IPv4 and IPv6.  Then just 
>> adjust the justification requirements so that IPv6 hardly has 
>> any, and IPv4 has harder requirements - that way, 
>> organizations that only need IPv6 will just leave the IPv4 
>> justification requirements blank, and only get IPv6.
>The justification for additional address space in IPv6 requires
>a much lower host density than IPv4 (if we take "host density"
>to be analogous to "assignments").  
>See the Number Resource Policy Manual, specifically
>This was based on a recommendation from the IETF, though adjusted 
>last year to be slightly tighter (policy proposal 2005-5):
>And I have to tip my hat again to the Member Services staff who
>maintain the NRPM for making it easy for me to find (using the
>change log) what policy proposal changed the policy and when.
>I'm having trouble envisioning a combined justification form.
>Would you be willing to cobble one together, based on existing

Yes, I would.  I'll include a sample in a proposal.

The ideal thing would be a simple formula that would spit out
an IPv6 allocation, unless the IPv4 requestor would want for some
reason to request a small amount of IPv4 and a large amount of IPv6
at the same time.


More information about the ARIN-PPML mailing list