Steven Blake wrote: > don't think it is necessary to be unable to guess the date or > relative order in which a particular ID was allocated, however.
Someone will have a reason to want such unguessability. That is sufficient. This is one of those cases where making it work in 100% of cases is worthwhile because it means we don't need to keep in mind the limitations of the compromise that we chose. >3. I don't believe it is essential to have alternative registration > methods besides web and e-mail. I think this should be decided by the other issues that determine whether it is necessary to charge for allocation. If it works out to be feasible, I think we should have no-cost allocation, via an automated web/email system -- no need for distribution of the registration function, no need for the mechanisms to handle payments, and so on. It'd require an investment by someone of the hardware (roughly a PC plus 256GB disk space plus backup provisions -- total cost a couple of thousand dollars), and would lead to a fantastically hassle-free registration process. I think it's worth sacrificing offline means of registration in order to get this simplification. It is only if this really cannot be done that we should accept a non-zero-cost registration process: the need to handle a payment complicates the process for everyone involved, and makes it a lot less accessible than the free-to-end-user system would be. In this system, offline registration mechanisms would effectively come for free alongside the payment-handling infrastructure. > In a dispute one can prove that he or she owns > an allocation by producing a non-repudiatable (e.g., signed) message > from a registry. I note that the automated system I suggest above can do this admirably. -zefram -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
