On 05/08/2011 10:32 AM, John R. Levine wrote:
Depository's business model appears to be to set themselves up as a toll booth on the information highway.
I totally agree with you on that point. And yet even accepting that their business model is to make money I do not see why that is deserving of condemnation? I agree that we don't need five RIRs,
but the last thing we need is a freelance RIR competing with the existing ones, thereby ensuring that the IP registration data will be even patchier and less accurate than it is now.
That's the kind of argument that I believe is very constructive when I run it through my "first law". Let me put that another way: My "first law" suggests that Depository's business ought to be allowed unless strong and convincing arguments can be produced to show that it causes a public harm. There *is* a public harm if the tracking of IP addresses back to those who operate them were to be muddied. And as one who uses IP addresses, I feel that such muddying would be a pretty serious public harm. Does that harm rise to the level that Depository should be denied? Perhaps. However, this does suggest to me that if muddy records is the harm that is feared then if Depository makes a commitment to a reverse IP address lookup system that is at least as good as those offered by any of the RIRs then the argument of public-harm dissipates. Personally, I don't understand why anyone would want to create a new RIR - at least not for IPv4 addresses. I do see the need for recordkeeping of what block belongs to whom - but to my mind that ought to be a non-discretionary, clerical, boring, and low cost task that need not be necessarily associated with a RIR - any decent bookkeeping company could do the job. And I do see the need for brokerage functions for IPv4 address blocks - and given the lack of new IPv4 inventory I don't see what the RIRs are going to be doing unless they get into the brokerage business. I have a hard time understanding how this is an issue that goes beyond IPv4. And for those who believe that we will be coerced to IPv6 quickly then those people might feel that if they just let IPv4 address traders exist that if they cause problems that those problems will be short lived. However, I sense that some of the energy being displayed here comes from a fear that IPv6 isn't going to take off. (My own feeling is that IPv6 isn't going to happen and that, instead, the net is going to fragment into separate IPv4 regions, each with its own complete 32-bit IPv4 space, connected by application level gateways.) --karl--