[arin-discuss] Suggestion 2010.1 -- Initial Fee Waiver for IPv6 assignments to LRSA signatories

[Bob Atkins]

John,

John Curran wrote:
> On Feb 5, 2010, at 9:23 AM, Bob Atkins wrote:
>
>> Generally, I am pleased with the IPv6 allocation fees being waived 
>> currently - it is quite simply why we applied for an IPv6 block. 
>> However, when I look at what is being proposed for IPv6 address 
>> charges - they are outrageous when you take into consideration that 
>> IPv6 is a virtually unlimited resource. IPv6 is a license for ARIN to 
>> print money. We are all going to be operating with IPv4 and IPv6 
>> address space for a /_*long*_/ time - that very fact will 
>> substantially increase ARIN's revenue based on the current fee 
>> structure. As a small ISP I find this very fact very troubling. Just 
>> as the market is eating itself alive from competitive pressure - when 
>> we must spend more in ARIN fees, equipment and training to deploy 
>> IPv6 and it will be virtually impossible to for us to recover any 
>> additional revenue from the use and delegation (to our customers) of 
>> IPv6 space. ARIN is maintaining what I call an 
>> IPv4 /_*scarcity*_/ fee structure for IPv6 space which is totally at 
>> odds with the monumental amount of address space that is available.
>
> Given the size of the IPv6 allocations, and present need-basis 
> requirements that must be assessed before an *additional* 
> allocation, it is quite likely that an ISP that comes in with an 
> *additional* allocation request would presently need to supply 
> a very significant amount of documentation that would need
> to be reviewed.   I'm not saying this is "good" or "bad", but 
> simply reflects what actions ARIN will need to take based on 
> present policy.  If this is recovered via ongoing fee schedule,
> then everyone will pay their share regardless of whether they
> make additional IPv6 requests or not.  If it is transaction-based,
> then just the party asking for the additional block will be paying
> for those costs.
Given the enormous amount of IPv6 address space that is available I 
don't understand why there would be the need for a rigorous, IPv4 level 
of review, which you seem to imply the need for a significant amount of 
staff to handle such reviews. I would think that basic delegation 
analysis would likely suffice and I do understand that some personnel 
are necessary. However, its not like doling out another /32 to /22 IPv6 
allocation is going to have much of an impact on the reserves of IPv6 space.

With our /32 IPv6 allocation, I don't expect to need another IPv6 
allocation until we have grown about 1000 times larger than we are 
which, (sadly) is unlikely to occur in my lifetime.  :-( I suspect that 
this would be the case for virtually all regional ISPs the world over 
leaving just a handful of very large telcos that may need another 
allocation after their initial /22 - perhaps by the year 2150...  ;-)

Given the existing 'standard' that almost 99% of enterprise customers 
have of using NAT for IPv4 based on the 'security' benefits that NAT 
offers - we rarely assign much more than a /29 of IPv4 space to our 
enterprise customers. The typical IT geek is often horrified by the idea 
of having 'real' IP address space internally. While we ISPs are being 
asked to head down the IPv6 path, I find it likely that we may end up 
using _/*microscopic*/_/_**_/ amounts of it assigned to customer router 
interfaces that just want to NAT everything internally to private IPv4 
space. I don't think I'm alone in this observation and I really think 
that it may be /_*decades*_/ before IPv6 utilization rises to the level 
of present day IPv4 utilization so I kinda doubt there are going to be 
very many additional IPv6 allocation requests any time in the next say, 
20 or 30 years.
>
>> IPv6 fees for a /32 should be about $100/year - at most! Larger 
>> blocks could be sold on an economy of scale basis rather than a 
>> linear multiplier of the cost of a /32 block. The entire process 
>> should be fully automated - requiring minimal support staff. An even 
>> better solution would be to create some competition in the IPv6 
>> address management arena by allowing the existence of multiple IPv6 
>> registrars in the same way that is done for domain names.
>
> One of the reasons that ARIN is investing significantly 
> in automation is precisely to aim for lower costs (and 
> hence lower fees) for those items which can be 
> automated. The current roadmap is available online 
> here: <https://www.arin.net/knowledge/roadmap.html> 
> Automation of maintenance will go far in reducing the
> ongoing costs and allowing fee schedules which meet
> expectations of the community.
Thanks - its good to see this happening.
>
> /John
>
> John Curran
> President and CEO
> ARIN

-- 
*Bob Atkins * 
/President/CEO/

*DigiLink, Inc. <http://www.digilink.net>*
Business Inter-net-working
*/The Cure for the Common ISP!/*

	

Phone: (310) 577-9450
Fax: (310) 577-3360
eMail: bob at digilink.net

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-discuss/attachments/20100205/835b103a/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DigiLink_esig_logo.jpg
Type: image/jpeg
Size: 23605 bytes
Desc: not available
URL: <https://lists.arin.net/pipermail/arin-discuss/attachments/20100205/835b103a/attachment.jpg>