[ARIN-consult] Fee Schedule Change Consultation

Thu Oct 25 23:59:01 EDT 2012

Jimmy,

I think william is speaking to a more fundamental issue that I'll try to
paint more broadly. Getting PO's etc are practical issues and are
certainly significant hurdles in organizations where you have over 100,000
employees let alone 10,000.

I think William's observations are of interest because he is illustrating
a progressive scheme that appears almost punitive to small business while
rewarding the larger organizations and their vast waste of resources. I
would argue it even encourages that waste.

	Looking at this from a broader view I think it's important to take a look
at how were things handled in the past as well as the result. In the past
this same progressive scheme was used and one has to ask what have we
achieved in using it? You have organizations like Sony with a /8 that are
using it _entirely_ for internal addressing. I wish I could say this was
uncommon having worked at Sprint, WorldCom, Sony, and EarthLink I know
it's the norm. I think when you have a sliding scale like this it has an
inadvertent effect of favoring large allocations to organizations who are
extremely wasteful.

	What would happen if this progressive scale were turned upside down this
time around? I think the following would happen:

1. Organizations would be encouraged to use address space more efficiently
2. I think ARIN's revenues would be greatly increased

3. The one drawback is ARIN may possibly have to deal with more people
(i.e. greater ticket volume) but I think this would pay for itself

	Maybe I'm being overly skeptical but I was around when IPv4 was first
being issued and I remember the feeling that there was a limitless supply
of address space. As a result you have organizations like Apple 17/8, HP
15/8, Ford 19/8, Merck 54/8, halliburton 34/8, ely lilly 40/8. The list
goes on and on. Do we want this situation again? I just listed off
100,000,000 IP addresses allocated. Does anyone believe Ely Lilly needs
441.5 public IP addresses per employee to make Cialis?

	I would hate to see us start down the same path expecting a different
result. Are there a lot of IPv6 addresses? Sure, but what good does 4.8x10
to the 28th power do if the underlying allocation policy doesn't trickle
down to end users that way?

Jimmy, I think you and I may see things somewhat differently as to why
things are where they are with IPv4. To me, getting where we are today,
after only about 10 years is an issue of policy. I think at some point we
would have arrived here anyway but I think a large part of the reasons
we've arrived here with IPv4 so soon is because:

1. What I cited above regarding inappropriate allocations and the
assumption that it would never run out (seem familiar?)
2. The larger your allocation the easier it is to acquire additional (and
massive) space (this is backwards imho)

	I think it's easy to be tempted to look at this as a $$ issue and say
well these fees are small compared to enormous company X. That may be true
but I don't think it's the point. I think as organizations start getting
larger and larger allocations we should start taking a much closer look at
"what are you doing" and "why". I think when you have a fee structure that
reflects that rather that a fee structure that reflects the exact opposite
you may get a different result.

-- 
Jesse D. Geddis

LA Broadband LLC
AS 16602

On 10/25/12 8:43 PM, "Jimmy Hess" <mysidia at gmail.com> wrote:

On 10/25/12, William Herrin <bill at herrin.us> wrote:

> question I'm asked: can it wait. The second question I'm asked: do we
> lose anything by waiting. The honest answers: it can and directly
> speaking, we don't.

Well, you could cite RFC6540, and remind that the internet standards
say you can't wait.

A $20 reg fee is ridiculously small,  compared to the human cost of
planning and deploying, in terms of  person hours,  and new training
for technical staff that need to understand the network.  But
'waiting' or trying to not spend the cash, does not make the IPv6
future go away.

By  waiting  and not at least pre planning IPv6  you have a risk of
increased amount of work that will have to be done in the future;
reexamination  and reconfigurations of hosts and networks deployed
without IPv6,  possible requirements to replace equipment,  unless
your  IPv4 network will stay the same size the whole time you wait.

So yeah, there is something quantifiable lost,  and  some significant
risks created
by  just choosing to wait.

>

Rather than try to balance them out at this point;  I do say,  I
believe  the IPv6 initial allocation, transfer fees and maintenance
should be much lower than IPv4 fees, rather than unified.

A zero-cost for the end user option should be offered to receive an
initial allocation of IPv6 resources any time IPv4 resources are being
transferred, requested, by an org with no IPv6 block,  and a  low-cost
option to  pay for an IPv6 allocation at the same time as IPv4
resource renewal.

The work required for ARIN to manage each block of IPv6 resources
should be much lower,  especially with  ARIN's  "ipv4 countdown plan"
and   requirement  for  increased amounts of  review during initial
allocations.

Not to mention the many services which are really only of interest for
IPv4 resource holders: such as the  ARIN STLS.

And of course the opportunity  to  promote using one large block of
IPv6  for new initiatives when possible,  over   multiple disparate
IPv4 blocks which will have to be requested or obtained at separate
times, as the usage need increases.

> Bill Herrin
--
-JH
_______________________________________________
ARIN-Consult
You are receiving this message because you are subscribed to the ARIN
Consult Mailing
List (ARIN-consult at arin.net).
Unsubscribe or manage your mailing list subscription at:
http://lists.arin.net/mailman/listinfo/arin-consult Please contact the
ARIN Member Services
Help Desk at info at arin.net if you experience any issues.