[arin-ppml] IPv4 Depletion as an ARIN policy concern

[Warren Johnson]

Tom,

I understand your point of view and appreciate the response.  But, your
comment below is very interesting:

" doing what they always do, what their job descriptions and careers and
personal expertise mandates that they do, i.e., plan ahead for potential
problems and potential opportunities as much as possible given the time and
budget constraints that everyone faces"

1) If your comment refers to the executive end of the company then we must
consider that this person's mandate is to make as much money as possible for
the stakeholders in the organization as quickly as possible.    If this
means hoarding IP addresses, exercising monopoly or cartel power, than that
is what is going to happen (provided it is legal in their jurisdiction).  If
you have enough IPv4 addresses to grow, then going to IPv6 simply opens the
market back up to competition and incurs migration costs.  That's a hard
sale to the board of directors.

2) If your comment refers to the technology braintrust end of the company
then we must consider they don't make the business decisions.  See point 1.

I never said people were evil.  I only said that in a market driven
situation, the tendency is going to be to make as much money as possible, as
fast as possible and for as long as possible.  If that means blocking IPv6
transition, than that is what is going to happen.  You're right though, that
encumbants will find themselves in the extremely enviable position of having
acquired their holdings during the happy times.  

What particularly distresses me is that there is a lot of discussion on
minimum allocations, fair distribution right before run-out etc. And very
little discussion about how to prepare for and deal with the cartel-like
atmosphere that is going to develop once Ips run out.  Around 1936, the city
of New York issued about 11,500 taxicab licenses.  You couldn't pick up a
fare without one of those licenses.  In 2000 how many of those licenses do
you think existed?  Same amount.  What you have is a closed system where
there only so many tickets to get in and you need to purchase an existing
one if you want to be involved.  If your family or company happened to be
around when the city issued them for $10 each back in the 30's then you're
doing well now (medallion for a taxi costs about 700k now).  What you get is
a cartel like situation in New York City.  


Thanks,
Warren



-----Original Message-----
From: arin-ppml-bounces at arin.net [mailto:arin-ppml-bounces at arin.net] On
Behalf Of tvest at eyeconomics.com
Sent: Friday, October 23, 2009 10:24 AM
To: ARIN PPML
Subject: Re: [arin-ppml] IPv4 Depletion as an ARIN policy concern

Thanks Warren.

However, to repeat something I've said here many times before, I personally
don't that anything so cynical as this is going on.
Doubtless there are a few parties who are looking forward to becoming (and
remaining) the exclusive brokers of the world's few remaining usable IP
addresses, but I think the majority of people (esp.  
participants in this list / policy process) are just doing what they always
do, what their job descriptions and careers and personal expertise mandates
that they do, i.e., plan ahead for potential problems and potential
opportunities as much as possible given the time and budget constraints that
everyone faces.

People aren't "evil" now, nor are they going to become evil when IPv4 runs
out -- I don't think so anyway. In the end the exhaustion of IPv4 is a
certainty, and nothing that any hypothetical speculator or aspiring would-be
monopolist could do, or not do, will change that fact. The problem is that
once IPv4 ceases to be available through the current mechanism, the normal,
prudential, professional decision making calculus of commercial network
operators will inevitably, unavoidably shift to accommodate that fact. At
that point, it won't matter who wanted or really didn't want to be a
speculator or monopolist. At that point, every IPv4 holder will become an
IPv4 broker, active or passive, whether they like it or not. At that point
the uncertainties and commercial pressures that have resulted in the current
level of IPv6 deployment are only going to become more acute.

As the old saying goes, when life gives you lemons, make lemonade.
In this case, however, making lemonade is not the only option; one could
also cause all of the other sources of potable liquids to become unavailable
indefinitely.

Even the nicest commercial lemonade vendor might have a tough time resisting
that option.

TV

On Oct 23, 2009, at 10:29 AM, Warren Johnson wrote:

> I agree completely.  For the last six months I have discussed and 
> debated privately with individuals about what I consider the probable 
> situation once
> IPv4 allocation requests can no longer be met. The situation we are 
> facing is horrible at best.  We're running out of IPv4 addresses and 
> the world is not even remotely situated to start using IPv6.
>
>
> Let us consider the not-so-meager issue of critical mass.  Consider 
> the unlikely near-term scenario that the world is 80% onto ipv6.  So 
> I'm running a website and I am only on iPv6.  That precludes 20% of 
> the internet from getting to my website.  Am I willing to pay $20 or 
> $30 a month for an IPv4 address so I can capture the last 20%?  If I 
> was a business concern (majority of websites) I would do it of course.  
> Add on to that dual-stacking and you basically have everyone using 
> IPv4 addresses until we reach ultimate critical mass (95%+ conversion 
> maybe?).  And if we're all on
> ipv4 anyway, why bother spending the money on ipv6?
>
> Let us also consider the potential power of the ipv4 cartel.  Right 
> now the big boys in the USA (ATT, Comcast, Time Warner Cable) are 
> among the largest
> non-legacy IP holders.   Officially, these guys all have ipv6  
> gameplans.
> But that is PR in my opinion. I'll tell you why.  Suppose you want to 
> start a new cable internet company.  You figure you can get 1 million 
> subscribers so you go to ARIN and you request 1 million IP addresses.  
> Ooops, sorry none left.  So you have to use ipv6. Well ipv6 isn't 
> going to cut it because the world isn't converted over enough yet.  So 
> what happens? You don't start an internet cable provider company.  Who 
> does that benefit?  Can you imagine going to the board of directors of 
> COMCAST and telling them "let's go to ipv6... Sure it'll open 
> comeptition up again but we'll be promoting the well being of the 
> world".  A  more likely scenario is "Officially, let's have an
> ipv6 policy but let's not really push ipv6 because ipv4 gives us a 
> virtual monopoly on this market, stiffles competition and makes us 
> more powerful and rich".
>
> Here is something everyone needs to consider VERY CAREFULLY:
>
> The current ipv4 stakeholders have an economic incentive to block or 
> delay the transition because it drives up the value of their IPv4 
> holdings.
>
>
> Good-bye IPv6, it was nice knowing you.
>
>
>
>
>
> -----Original Message-----
> From: arin-ppml-bounces at arin.net [mailto:arin-ppml-bounces at arin.net]
> On
> Behalf Of tvest at eyeconomics.com
> Sent: Thursday, October 22, 2009 8:14 PM
> To: ARIN PPML
> Subject: Re: [arin-ppml] IPv4 Depletion as an ARIN policy concern
>
>
>
>>>> Are you in favor of changing anything at all or can you think of no
>>>> better course of action than to continue exactly as is now?
>>>>
>>>
>>>
>>> IMO, it's time now to think about what we do *beyond* the end of the
>>> free pool when IPv4 addressing policy changes to a zero-sum game.
>>> Where giving one org new addresses means taking them from someone
>>> else.
>>> The address market strategy might work. Ought to work. But we should
>>> probably make some contingency plans.
>>>
>>
>> Ration, Reclaim, Return, Reuse.
>>
>> Those are the alternatives to transfers based on market principles.
>> I greatly prefer the market which is why I advocated for it, but
>> policy for what to do with reclaimed space after depletion is still
>> needed and any approach to it that doesnt consist of giving it all to
>> whoever can show need will smack of rationing.
>>
>> And in the strictest sense of the word they are correct, it is
>> rationing. However supply and demand markets are also a form of
>> rationing, so the word in and of itself does not carry automatic
>> negative connotations.
>>
>> Only in a worst case scenario where neither transfers or returns are
>> meeting even a portions of needs and ipv6 is not obviating ipv4 need
>> should any attention be given to reclaimation of non-abandoned
>> resources.
>
> Is anyone else experiencing any cognitive dissonance here?
>
> A. No clear community consensus in favor of mitigating the impact of
> IPv4 runout; many concerns raised about the fairness of depriving  
> current
> IPv4 holders of anything less than the max. IPv4 that they can justify
> between now and runout.
>
> B. No significant likelihood of anything close to IPv6  
> substitutability in
> the foreseeable future; zero probability before
> IPv4 runout.
>
> C. No apparent acknowledgement of what this implies for anyone/  
> everything
> who might need -- and be able to justify -- "usable" IP addresses of  
> any
> kind after IPv4 runout.
>
> D.
>
> ***
>
> IMO, this combination suggests that it would be prudent to anticipate
> that the "worst case scenario" as described above is also the highest
> probability scenario, by a wide margin.
>
> It's still not too late for some version of prior planning...
>
> TV
>
>
>
> _______________________________________________
> PPML
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> http://lists.arin.net/mailman/listinfo/arin-ppml
> Please contact info at arin.net if you experience any issues.
>
>

_______________________________________________
PPML
You are receiving this message because you are subscribed to
the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
Unsubscribe or manage your mailing list subscription at:
http://lists.arin.net/mailman/listinfo/arin-ppml
Please contact info at arin.net if you experience any issues.