RE: Recommended Draft Policy ARIN-2013-8 My comments:
All of computer science is made up of allocating storage (memory/disk), de-allocating storage, or moving bits around. Like all organizations, the current situation we are all in (exhaustion of IPv4 addresses) is due to an improper de-allocation of IP addresses. The fact that we are in 2014 after a 30 year run talking about what to do means that de-allocation is already good. The current situation is due to desktops/servers/storage units requiring a new IP address (throwing away the old one) while the core routers have the same IPs that were in place when the internet was created. There have been effective solutions put in place by ARIN and others to 'put a thumb in the dike' of this de-allocation issue. There are many possible solutions, but the proposed solution means that ARIN will 'go slow' with allocating the remaining IPv4 addresses stringing out the deployment of IPv4 addresses for as long as possible. It is already economically very difficult for a new entrant to get 'in' and it will be impossible once the new policies are in place. Now, it is not all bad for there not to be any new entrants into a market (it is the heart of standards), and the market gravitates towards three major solutions anyway once something becomes a commodity. The real question is "has the internet become a commodity already, or is there still some juice left in it?". It is impossible to answer this in advance. I do know that when ARIN was formed, the biggest problem was giving everyone internet connectivity, which involved a major expense running wires, buying wireless spectrum, etc and the investors who made it possible deserve to be paid a profit because they were very successful at deploying internet connectivity. 1) It appears that there will be no new ISPs and no one will get into this business. It is difficult already, but if the draft policy by ARIN is put in place, it solidifies and codifies ARIN's ratification of this. Although we all saw the unintended consequences arising when the US Congress made possible CLECs (which were unsuccessful in the market) and new ISPs are very much like CLECs were, it is a very dangerous thing to provide policy that makes sure there will be no new ISPs because there is no economic incentive for one to be created. The opportunity to get ahead by creating a new ISP will soon be removed by ARIN policy. Does ARIN want to enable the entire country to remain a 'banana republic' where the rich are getting richer at the expense of the middle class/small business, or does ARIN wish to be associated with the 'land of opportunity' (not subsidy) by allocating resources to large and small enterprises on an equal basis? 2) There is no need to mess with IPv6 policy. The current situation which we have all been trying to implement for a decade will not be enhanced by this policy change. The change in policy is that IPv4 is getting a lot more restrictive in allocation and IPv6 will be tied to existing IPv4 allocations. It really means that there will not be an opportunity for a new ISP even after the IPv4 addresses are a thing of the past. If it ain't broke, don't fix it. There is ample opportunity for ARIN to create an "intellectual property tax" for payment to ARIN based on existing allocation size and market prices for the IP addresses (separate ones for IPv4 and IPv6). Does ARIN want to make sure only incumbents are able to get IPv6 addresses? 3) If we return to the 'bank of modems' of the dial up modem era, then every modem has to have its own separate dial tone. There may be a way to use one phone number (like IP addresses), but the modem pool still has to have an isolation mechanism per modem. The policy as written will specify that someone getting into the 'dial up modem' business can't deploy but a handful of modems at a time, that all modems must be 80% utilized before any more can be bought, and that the phone number will change for all modems on the modem bank if more modems are deployed. An ISP ensures that a customer is able to put their own phone number on the banks of modems while a large enterprise means that they have to control the phone numbers too. It is a subtle distinction but it at the heart of the question "Does ARIN wish to have a more relaxed policy for large Enterprises than ISPs?". 4) It is important for ARIN to maintain the existing internet policy thru allocation. It is hard to see how the existing policy change will enhance an accurate allocation other than there will be less players to watch after and the expense will be known in advance. Does ARIN want to 'remove the band-aid slowly' which the proposed policy change does, or does ARIN and others involved undergo less pain if the IPv4 band-aid is removed quickly? 5) Doing something now is akin to 'closing the barn door after the horse has run off', similar to anyone that gets broken in to buying a burgler alarm system after they were robbed. In an effort at fairness, because ARIN must serve both large and small internet clients and because of the huge allocations in place in 2012-2013 (.5% of the companies got most of the IP address allocations from ARIN), the attention has been to be fair in administration of ARIN policies. Will the existing policy change enable ARIN to be more or less 'fair' with the remaining IPv4 allocation? Mike Mazarick _______________________________________________ PPML You are receiving this message because you are subscribed to the ARIN Public Policy Mailing List ([email protected]). Unsubscribe or manage your mailing list subscription at: http://lists.arin.net/mailman/listinfo/arin-ppml Please contact [email protected] if you experience any issues.
