[arin-ppml] IPv4 is depleted today

[Ted Mittelstaedt]

> -----Original Message-----
> From: Scott Leibrand [mailto:sleibrand at internap.com] 
> Sent: Tuesday, September 02, 2008 2:51 PM
> To: Ted Mittelstaedt
> Cc: arin-ppml at arin.net
> Subject: Re: [arin-ppml] IPv4 is depleted today
> 
> 
> Ted Mittelstaedt wrote:
> > 
> >> Ted Mittelstaedt wrote:
> >>
> >>> It is imperative that our policy making strive to increase ease of
> >>> moving to IPv6 and decrease ease of staying with IPv4.
> >>>
> >>> A liberalized transfer policy does nothing to increase ease
> >>> of moving 
> >>> to IPv6, all it does is make it easier to stay on IPv4.
>  >>
> >> I think you just identified a key point of disagreement /
> >> philosophical 
> >> difference.
> >>
> >> To my mind the idea that "we need to adopt policy to make
> >> people's lives 
> >> harder", in the interest of promoting our idea of their 
> >> long-term best 
> >> interest, is an extraordinary claim that requires 
> >> extraordinary evidence 
> >> that the benefits outweigh the harms.
> >>
> > 
> > But nobody who is against a liberalized transfer policy is 
> asking that 
> > we adopt ANYTHING.  They are asking that we DO NOT adopt anything.
> 
> Understood, but didn't you just say that we need to "decrease ease of 
> staying with IPv4"?
> 

Our current policy naturally decreases ease of staying with IPv4
post IPv4 runout.  You might say that we already adopted an
anti-liberalized transfer policy years ago by not permitting people
to buy and sell IPv4 from day one.

However this is beside the point.  The fact is that some opponents of
a liberalized transfer policy wouldn't agree with my assertion that
our policymaking should decrease ease of staying with IPv4. The ones
that would agree don't want to mix opposition to a transfer policy with
advocacy of more stringent requirements for IPv4.  (at least, no one
who has opposed it as cited a reason for opposition is that they want
to put in a policy change making it more difficult, so I have to
conclude this)

Thus, the set of liberalized transfer policy opponents contains a lot
of people, some more and some less anti-IPv4, with different ideas on
how to do this, and therefore as I was explaining in the first place, you
cannot simply lump it all together into a big mass.


> >> I'm all in favor of easing the move to IPv6.  However, I
> >> think we also 
> >> need to do what we can to ease the transition across the board.
> >>
> > 
> > Your right.  I hope you define "transition" as I do, 
> meaning in this 
> > instance CHANGE from IPv4 to IPv6, and NOT meaning continue to use 
> > IPv4 and ignoring IPv6.
> > 
> > Would you be willing to tie a liberalized policy to a 
> requirement that 
> > the transferee (ie: the party that is obtaining the IPv4 as 
> a result 
> > of a liberalized transfer) submit a transition plan to IPv6 
> to ARIN, 
> > and some reasonable proof that they were actually doing some of the 
> > items on it?  Such as obtaining IPv6 allocations from ARIN? 
>  And such 
> > as providing the date their feed is going to natively route 
> IPv6 and 
> > the date that they plan to do so as well? And maybe some mandatory 
> > followup by ARIN a few years later to insure these things are being 
> > done?
> 
> I supported such conditions in the Soft Landing proposal, and 
> would not 
> oppose them in a transfer policy.  However, the consensus 
> seems to be that 
> we should be removing restrictions from 2008-2, not adding 
> them, so I'd 
> need to see some community support for such conditions before 
> I'd consider 
> adding them to the proposal.
> 

That's fine, but if we fail to reach consensus to adopt many 
different "son of liberalized transfer policies" time after time, then
I think after about the 3 or 4th iteration fails, your going to have to try
something different, regardless of what people seem to indicate.

I recognize the logical position for someone advocating a liberalized
transfer policy is to start by trying to push such a policy with
as few strings attached.  You don't compromise right out of the
box because you don't know yet if you can push it through without
the strings.  You only compromise when you fail.

> > If not, what a transfer policy does is allow the status quo to 
> > continue UNCHANGED.  That is NOT transition.
> 
> IMO transition is not something we can force.  Either it will 
> happen or it 
> won't, and at best we can only nudge things along slightly.  
> What we can 
> do is make it easier to adopt v6, by lots of hard work fixing 
> interoperability problems.  We can also take policy action to 
> ease the 
> transition and buy a little time while those problems are fixed.  IMO 
> *not* doing everything we can on both fronts is an abdication 
> of ARIN's 
> stewardship responsibility.
> 

The argument that we need more time assumes:

1) IPv6 isn't ready now
2) More time will result in it being ready

Both are huge, unsupported assumptions.  I often see these claims 
about IPv6 but I never see the people making such claims responding
with concrete, specific problems with IPv6 itself.  The closest
they usually get is in listing problems with deploying it, not
in using it after it's deployed.

Kind of like arguing against a specific model of tire for your
motorcycle because you have spoked rims, and require an innertube
to put the tire on, rather than because of the mileage, treadwear,
material, etc. etc. of the actual tire itself.

> -Scott
> 
> P.S. This will be my last post on this topic today.  If we 
> haven't both 
> made our positions clear, more message won't do the trick.  :)
> 

yep! :-)

Ted