On Fri, Aug 22, 2008 at 11:53:02AM -0700, David Conrad <[EMAIL PROTECTED]> wrote a message of 1187 lines which said:
> It is a bit dated as it was originally written > before the IPv6 glue was added to the roots. If there is interest in > pursuing this, I could take a stab at another revision. If you do so, I suggest a few changes: > Obviously, glue modifications to the existing root servers must not > affect the glue associated the new technology servers. This would be > the only exception to the root data replication procedures. Well, adding DNSKEY and DS records, for instance, would also be a change to the data replicated. At another place, you say that the parroot will not "change the namespace" which seems a better description than to say it will not change the "data replicated". > While not an absolute requirement, maintaining coherency and > reducing the possibilities for fragmentation are goals that are > likely best achieved if the new technology root servers were to be > operated by the current root server operators (or a subset of those > operators) It seems quite questionable. First, I do not know how many of them want to participate. Second, parroot could be an interesting way to test new potential root name server operators (it is hard to believe that the set of root name server operators must be immutable for the next centuries). _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
