Published on October 14th, 2014
I am back from ARIN34 in Baltimore. The best outcome of ARIN34 was the passing of policy so that the ARIN manual no longer allows for the reclamation of IPv4 addresses, which is a tremendously good thing.
The hot debate at ARIN34 was around various forms of policies to lessen the requirements for ARIN IPv4 transfer requesters to perform IPv4 needs justification. This is two years after my partner Jeff Mehlenbacher first proposed removal of needs justification in 2012. I have to say it is really quite tiring to hear repeated arguments about why companies in ARIN region must continue to justify requests for IPv4 addresses.
Most individuals and companies opposed to loosening of needs justification have weak or self-serving arguments: it would be better for small companies than big companies (needs justification is better for big companies at present because small companies are sent to their ISP’s and don’t get any IP’s of their own) and it would be better for companies with deep pockets because they could buy a multi-year supply (not considering that perhaps part of that supply would become worthless as IPv6 becomes more pervasive). And the arguments in favor of needs justification I heard are also uneconomic: it creates jobs at ARIN! And manipulative: the ARIN community is the group of stewards responsible for protecting the internet, for prevention of hoarding and speculation.
The reality is that in most cases the transfer market frees up IP’s that would otherwise not be used and kept dormant.
RIPE NCC has had no needs in place for a year now, and prevents speculation by not allowing re-transfer of a purchased block for two years. And it would be incredibly difficult to buy up all the /8’s and hoard or control the IP markets. There are just too many players in the marketplace.
ARIN Policy 2014-14 will allow for removal of justification for blocks of size /16 and smaller. I heard some say that a /16 is too large: that we should experiment with a smaller block such as a /20 because we can adjust it to a larger block in the future, which is basically admitting that a /20 is the wrong size and they would rather lose their argument a bit at a time. Others say that putting a limit of a size /16 means that it is not a level playing field for companies that want more than a /16 and policies are supposed to be fair to all. Of course it is fair in that all companies can get up to a /16, if their business can use it over two years, is it not? It seems we cannot win at either end of the spectrum.
I personally support removal of needs justification for a different reason: because I see too many IPv4 requesters who face a 3 to 5 day turn around per ARIN interaction, and it takes them six to seven iterations with ARIN to get IP’s, with a few days between each iteration to gather data. This translates into a two to three month cycle to perform needs justification. When this is contrasted with a two day turnaround in RIPE region, it seems like a competitive disadvantage to North American business. I also feel that the business decision in the form of the cost of the IP’s is needs justification enough.
Other than my rant on needs justification, ARIN34 also had an interesting discussion on the future of Internet Governance. With the NTIA stepping away from IANA stewardship, the question is how to propose a new model of governance. My opinion is that North America should form its own opinions and contribute them to the global community, in order to expedite the process. Gathering of requirements in a bottoms up approach by RIR will be needed. ARIN has proposed a survey, which in the form presented in the meeting, had leading questions. Hopefully the input at the meeting will allow enough write-in information that new thoughts will be gathered and acted upon, and ARIN’s predisposed ideas won’t distort the input.