[arin-discuss] [arin-ppml] Fee proposal (was Re: Alternativetoarbitrarytransfers)

Tue Apr 7 23:26:11 EDT 2009

I am going to try and roll my reply into this one note rather than reply to each individually. This is not to dismiss any other comments, but just trying to cut down on the email. John, Joe and Ted all have valid concerns. The evolution of the Internet has resulted in behaviors by some that are less than efficient.

One concern I have is that the conversations need to move past the scope of computers connecting to web sites as the primary consumers of IP address resources. I'm sure I'm not saying anything you all don't already know, but there are far more televisions and video subscribers than there are Internet subs, and even more wireless handhelds than the both. 

Take a look at the efforts in Tru2Way, WiMax, LTE, DOCSIS 3.0, etc. and it becomes very obvious that the landscape is changing. Comcast has almost twice as many video subscribers as they do Internet subs. Sprint/Clearwire has an FCC requirement to try and provide broadband to 30M subs by 2010. Countries like India, China, Russia, Brazil, Mexico, Turkey, and so many others are completely bypassing the legacy problems crippling the United States, and deploying DOCSIS, and 4G solutions to their citizens. 

You all make valid points about better utilizing a /29 or /24, but in the end it wouldn't make more than a few months of difference. 22 /8's have been allocated to the Registries in the past two years. There are 32 /8's left, and the growth rate continues to increase regardless of any recession.

I will stop with one final point. This conversation started with the premise that fees are needed to drive IPv6 deployment. I'm guessing this is from the assumption that organizations are not deploying IPv6, nor do they want to. I am quite confident that this is not true. Of course you are not going to see wide-scale IPv6 deployments. IPv4 resources are still readily available. IPv6 deployment is not a technical issue, it is an issue of financial planning. Most ISP are working on IPv6 deployments. They just aren't going to tell their competition. Raising fees would not avoid IPv4 depletion by spurring IPv6 deployments. IPv4 depletion is needed for IPv6 deployment and raising registration fees would just be passed to the consumer until IPv4 depletes anyways. 

That's just my opinion.
Dan

-----Original Message-----
From: Ted Mittelstaedt [mailto:tedm at ipinc.net] 
Sent: Tuesday, April 07, 2009 8:45 PM
To: Alexander, Daniel; arin-discuss at arin.net
Subject: RE: [arin-discuss] [arin-ppml] Fee proposal (was Re: Alternativetoarbitrarytransfers)

Hi Daniel,

  I work for an ISP and I play an end user on TV too.

  The issue on IPv6 transition is thus:  Yes, ISP's benefit by transitioning to IPv6 because IPv4 is running out.
And I daresay that for most ISP's now, the cost isn't so much in equipment as it is in time spent configuring and testing.  The problem though is that IPv6 transition is not something that an end user is going to want to do.

If you have an end user running a single PC running Vista on a bridged cable modem or DSL line, no sweat - you hand out IPv6, they take it, things work. (on the IPv6 network,
anyway)

  With XP it's a bit more work but still pretty easy.

  With people on DSL modems that have embedded translators, big time problem.  Same for people on embedded translator boxes on cable modems.

  And if you think that's bad, try small business customers who have firewalls and such they bought which don't support IPv6.

  So, we are going to have a situation post IPv4 runout where new customers are going to be selecting their ISPs and if their ISP tells them they must upgrade to get connected, they aren't going to go with that ISP if the ISP down the street still supports IPv4.  Every ISP knows this.

  Now, supposedly the utilization requirements instituted years ago make this fair for all ISPs.  In short, if all ISP's utilize their netblocks then all are screwed over at the same time when the last IPv4 address is assigned, and this forces the end user to upgrade - because no ISP will have available IPv4 to hand out.

  But in reality a large ISP can "hedge" in a variety of ways.  They can assign /29's to customers that ask for a SINGLE static IP number then subnet that later.  They can assign IPv4 to remotely-controlled-and-configured-and-upgradable CPE devices like cable modems with embedded NAT's in them, or cell phones, then later on upgrade those units to IPv6 and install
IPV6-IPv4 proxies.  They can assign /24's to customers who ask for them at extremely low prices then later on raise prices which will cause some customers to resubnet and return some IP numbering.

Granted, for a single customer, this kind of hedging doesn't put a lot of IPv4 in "reserve"  But with a large ISP it multiplies out to quite a lot.

Small ISP's cannot do this and store up enough IPv4 to make it worth while.

So, after IPv4 runout, FOR A WHILE the large ISP's who have been hedging will be able to institute reclamation programs that will free up IPv4 and will thus be able to avoid deploying IPv6 for a longer time than the small ISP's who aren't able to hedge.  So the small ISP's end up stuck with serving out NAT to their customers and begging their upstreams to please for God's sake start selling me native IPv6 - and the large ISP's they are buying service from are fiddle-faddling around and telling them to go to tunnel brokers.  And in the meantime the business customers who find it cheaper to buy Internet service by buying it on routable IPv4 since they don't have to upgrade, are being forced into going to those large ISPs.

Obviously, all good things come to an end and even if every ISP pulls these tricks, eventually all will run out of
IPv4 no matter how much hedging they do, end users who want a routable IP address will have nowhere to go and must go to IPv6.  But, damage will have been done.

I am not accusing large ISP's of doing this right now, or even ANY ISP's of doing this right now.  I am saying that the ARIN fee structure encourages IP consumption in that the more IP you consume, the less you pay per IP.

See https://www.arin.net/fees/fee_schedule.html

/16, 2^16  65535 numbers, $4500  6.8 cents per IP per year

/14  2^18  262144 numbers $9000  3.4 cents per IP per year

/13 2^19 524288 $18000      (same)

/12  2^20  1048576   $18000  1.7 cents per IP per year

and so on.

Thus, an ISP that decides to hedge is effectively being rewarded financially for doing it.

Ted

> -----Original Message-----
> From: Alexander, Daniel [mailto:Daniel_Alexander at cable.comcast.com]
> Sent: Tuesday, April 07, 2009 4:08 PM
> To: arin-discuss at arin.net; Ted Mittelstaedt
> Subject: RE: [arin-discuss] [arin-ppml] Fee proposal (was Re: 
> Alternativetoarbitrarytransfers)
> 
> Ted,
> 
> Let me start with the usual disclaimer that this is my own opinion and 
> not the position of my employer. Registration fees are not the key 
> factor in whether or not to deploy IPv6, nor are they the key factor 
> in the efficiency in which IPv4 address space is utilized.
> 
> The largest consumers of IP address space are ISP. The reason they use 
> these IP resources is not because they are cheap, or easy to come by. 
> It is because they are connecting end users (I assume you are one of 
> them) to the Internet. The growth, and very existence of an ISP is 
> based on the services they provide and the revenue these services 
> generate.
> 
> The fact that IPv4 is running out is the single largest incentive any 
> ISP needs to deploy IPv6. The lost revenue, because you don't have the 
> IP resources to add new customers and services, far outweighs any 
> increase in fees that ARIN could impose. Another thing to consider is 
> that historically, an increase in fees are rarely born by the 
> provider, and are more often passed along to the consumer. It is the 
> end user who would most likely bear the burden of the fee "incentive"
> you propose.
> 
> Again, this is my own opinion.
> Dan Alexander
> ARIN AC
> 
> 
> -----Original Message-----
> From: arin-discuss-bounces at arin.net
> [mailto:arin-discuss-bounces at arin.net] On Behalf Of Ted Mittelstaedt
> Sent: Tuesday, April 07, 2009 2:14 PM
> To: 'John Tobin'; 'Lee Howard'; 'Brian Johnson'
> Cc: arin-discuss at arin.net
> Subject: Re: [arin-discuss] [arin-ppml] Fee proposal (was Re: 
> Alternativetoarbitrarytransfers)
> 
> 
> All I am asking is that anyone who cares anything at all about 
> transitioning to IPv6, be aware that a fee incentive exists for the 
> largest holders to NOT transition.
> 
> As I don't work at a large holder I do not know if the fee discount 
> for large IPv4 holdings actually influences decisions.  If it does 
> not, because the IP address registration fee is such a small part of 
> total business expenses, then perhaps the fees should be adjusted 
> until they do start to influence decisions.
> 
> Just a thought.
> 
> Ted
> 
> > -----Original Message-----
> > From: John Tobin [mailto:JTobin at origindigital.com]
> > Sent: Tuesday, April 07, 2009 8:01 AM
> > To: Ted Mittelstaedt; 'Lee Howard'; 'Brian Johnson'
> > Cc: arin-discuss at arin.net
> > Subject: RE: [arin-discuss] [arin-ppml] Fee proposal (was Re: 
> > Alternative toarbitrarytransfers)
> > 
> > 
> >  All, this thread is hard to follow... What are you asking me?
> > 
> > 
> > John Tobin	
> > Director Of Information Technology
> > 
> > 
> > 300 Boulevard East
> > Weehawken, NJ 07086-6702, U.S.A.
> > 
> > E: jtobin at origindigital.com   | C: 732-616-8780 | V: 
> > 201.272.8451 |  F: 201.272.8400 . . . . . . . . . . . . . . . 
> > . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
> > . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
> > This message contains information which may be confidential and/or 
> > privileged. Unless you are the intended recipient (or authorized to 
> > receive for the intended recipient), you may not read, use, copy or 
> > disclose to anyone the message or any information contained in the 
> > message. If you have received the message in error, please
> advise the
> > sender by reply e-mail at jtobin at origindigital.com and delete the 
> > message and any attachment(s) thereto without retaining any copies.
> > 
> > 
> > -----Original Message-----
> > From: arin-discuss-bounces at arin.net
> > [mailto:arin-discuss-bounces at arin.net] On Behalf Of Ted Mittelstaedt
> > Sent: Monday, April 06, 2009 6:52 PM
> > To: 'Lee Howard'; 'Brian Johnson'
> > Cc: arin-discuss at arin.net
> > Subject: Re: [arin-discuss] [arin-ppml] Fee proposal (was Re: 
> > Alternative to arbitrarytransfers)
> > 
> >  
> > 
> > > -----Original Message-----
> > > From: arin-ppml-bounces at arin.net
> > > [mailto:arin-ppml-bounces at arin.net] On Behalf Of Lee Howard
> > > Sent: Monday, April 06, 2009 3:09 PM
> > > To: Brian Johnson; ARIN PPML
> > > Subject: [arin-ppml] Fee proposal (was Re: Alternative to
> > > arbitrarytransfers)
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > ----- Original Message ----
> > > > From: Brian Johnson <bjohnson at drtel.com>
> > > >
> > > > If viewed by cost/IP, then the cost/IP for larger orgs
> > > (ISPs generally
> > > > speaking) is less than for smaller orgs. This has been long
> > > standing
> > > > policy. If you want to change this. Make a proposal and get
> > > consensus.
> > > > Don't degrade one group to make yourself feel better.
> > > 
> > > Probably suggestion process, not policy process.  The suggestion 
> > > doesn't have to be for a specific fee structure; rather,
> you[1] want
> > > to change the principle by which fees are
> > > set: instead of setting fees based on ARIN's cost, you
> want to set
> > > fees based on a per-address cost.
> > > https://www.arin.net/app/suggestion/
> > > 
> > > Probably requires member consensus.  Probably belongs on
> > arin-discuss.
> > > 
> > 
> > Lee,
> > 
> > It was not my intent to trigger a discussion on fees.
> > 
> > But, since we are discussing them, adjustments to the fee
> structure do
> > not have to be made to increase the money paid to ARIN.  You can 
> > reduce the discount to larger players, collecting more money from 
> > them, and reduce the fees for smaller players, collecting
> less money
> > from them, and end up with the same money coming in - just
> a different
> > distribution among the bearers of the fees.
> > 
> > In any case, I will direct your attention to the ARIN staff
> comments
> > on 2008-7, posted to arin-ppml on 3/23/09:
> > 
> > "...An annual re-registration of all POCs (~223,000 currently) will
> >       likely result in a vast increase in workload,
> particularly with
> >       the follow up work and research involved when a POC does not 
> > reply
> >       within 60 days. ..."
> > 
> > An increase in workload will mean having to hire more
> people at ARIN
> > which will increase costs.  Thus increasing fees under the existing 
> > principle.  Since increasing fees to the largest consumers of IPv4 
> > would increase incentive of those consumers to more efficiently 
> > utilize IPv4 and thus defer additional IPv4 requests, which would 
> > affect the largest amount of available IPv4, it would be completely 
> > logical to do this rather than increase fees across the board.
> > 
> > Ted
> > 
> > _______________________________________________
> > ARIN-Discuss
> > You are receiving this message because you are subscribed
> to the ARIN
> > Discussion Mailing List (ARIN-discuss at arin.net).
> > Unsubscribe or manage your mailing list subscription at:
> > http://lists.arin.net/mailman/listinfo/arin-discuss
> > Please contact info at arin.net if you experience any issues.
> > 
> > 
> 
> _______________________________________________
> ARIN-Discuss
> You are receiving this message because you are subscribed to the ARIN 
> Discussion Mailing List (ARIN-discuss at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> http://lists.arin.net/mailman/listinfo/arin-discuss
> Please contact info at arin.net if you experience any issues.
> 
> 