[arin-ppml] IPv4 Transfer Policy Change to Keep Whois Accurate
cengel at conxeo.com
Fri May 13 13:23:47 EDT 2011
Thanks for your explanation. The approaching run-out of IPv4 certainly brings alot of complications to the table for ARIN and I don't envy those charged with administrative or decision making duties in the coming months.
> On May 13, 2011, at 8:21 AM, Chris Engel wrote:
> > Actually, I would argue that "manufactured" need would be functionally
> identical to "genuine" need from the point of anyone forced to make
> approval decisions based upon that, unless the organization attempting to
> game the system was incredibly un-artful in their attempt. How would
> someone charged with deciding to approve the "IP-Enabled virtual pet rock
> program" that was simply instituted to grab and hold IP addresses for a time
> until they were actually needed for a legitimate project or could be sold (at a
> profit) to someone who had a legitimate need for those addresses..... from a
> company that legitimately wanted to enter the "IP-Enabled virtual pet rock"
> business and were simply bad at business planning? On paper, they'd be
> identical....the only thing that would be different was the motivation of the
> organization acquiring the resources....and how could anyone discern that
> without being a mind-reader?
> Both receive approval for a small initial allocation (unless the one
> which was serious could show a history of subassignments from their
> existing IP-enabled "digital paperweight" program...)
> As soon as either reached 80% utilization, we'd be happy to approve
> a significantly larger allocation (and so on)
> Does this clarify how these two organizations would be handled, and
> how gaming the system works in the small but has diminishing returns?
Sure but how do you determine genuine utilization from manufactured utilization. Does giving away a 3 or 6 month free subscription to a "digital paper-weight" with every happy-meal purchase count as utilization? How about having another company which is a legally separate entity (but just so happened to have the exact same private ownership/board membership) purchase 80% of the virtual pet rocks for their "virtual pet rock garden" count as utilization?
Those sort of shenanigans may sound rather absurd now, because there would be a real cost to enacting them in a convincing manner. However IF the real value of an IPv4 address increases beyond the cost of such ploys, suddenly they become worthwhile doing. Would not the cost/difficulty of effectively differentiating between gaming and the real thing also increase for ARIN under such circumstances?
> > Note that speculators always risk being left holding the bag when the
> bottom falls out of the market they are playing with. Risk is inherent in what
> they do...unfortunately it doesn't seem to curtail their activities very much in
> most scarcity markets.
> Agreed, although far fewer participate in situations where the risk
> is self-created due to operating contrary to the system. It's one
> thing to point out to investors that you misread the market, it is
> another to explain that you were doing something that was technically
> contrary to market rules and that resulted in the loss.
> > Functionally the only things I see a needs based requirement accomplishing
> in a transfer market is artificially inflating the cost of transfers,
> It definitely increases the net transaction cost.
> > increasing the cost for ARIN to administer transfers
> To be clear: it results in transfers having overhead similar to
> the assignment approvals which we are presently doing.
As I see it, as the rewards for gaming the rules increase, people will be willing to invest more resources in doing so. Unless you are willing to devote more resources to enforcing said rules, your effectiveness in enforcing them diminishes accordingly.
> > and creating a new job position among certain companies of "needs
> justification designer".
> That position or at least job task (IP address management) already
> exists in ISPs today (or at least those ISPs which are growing and
> need additional allocations on a routine basis)
> > I don't think it's going to accomplish what many proponents might hope it
> would very effectively.
> For clarity: there is a big difference between "effective" and "very
> effective"; do you mean that you don't think it will curtail routing
> growth and prevent speculation at all, or simply that it will provide
> less than ideal protection from such?
I think you are not going to end up without very much "bang for your buck". Either you are going to end with a rubber stamp system which people with some knowledge of the system and resources to spend in gaming it will increasingly be able to subvert or you are going to end up having to increase your enforcement efforts (and the resources devoted to them) significantly in order to maintain a similar degree of effectiveness as you do today.
It strikes me as a pretty unenviable position for ARIN. Perhaps the point David Farmer raised in his post about dropping the needs based justification for transfers below a certain size threshold (or perhaps annual quota for a single organization) and preserving it for ones that exceed it would be a good compromise position? After-all, small transfers would likely be approved on a trial basis anyway....and likely would do limited harm even if done without real justification.
Anyway, the above assumes that the scarcity of IPv4 addresses results in a significant increase in the value of such addresses. Perhaps if IPv6 adoption see's more of an uptick.... that situation will be significantly mitigated and the impact of all this won't amount to very much.
> > In a market where acquiring IP addresses is relatively trivial (i.e. unassigned
> space from ARIN), I can certainly see why there has to be some artificial
> barrier built into the system to prevent people from acquiring IP space for
> trivial reasons....with a scarcity market, the cost of acquisition through
> transfers itself should act as a barrier to that. Anyway, that's how I see it...for
> whatever it's worth.
> Good comments; hopefully this discussion will help folks in consideration
> of the large set of policy proposals before us.
> John Curran
> President and CEO
Thanks again, appreciate the ability to be able to participate in such discussions as just a member of the interested public. The openness and level of communication are refreshing.
(representing only my own views)
More information about the ARIN-PPML