Fred Baker - le (m/j/a) 1/30/09 8:30 AM:
... take a look at this email thread. How much discussion is about NATs, their proposed algorithms, and their proposed uses...How this thread goes is definitely a worry. Trying to get directly a consensus answer to a question like "Do we need NAT66?" h as proved to be a waste of time, and IMHO can only continue to be one. The reason is that this question is much too ambiguous. As some IPv6-specific and new functions are being discussed under the title NAT66, there cannot be a common understanding of what is meant by NAT66 in such a question. A better approach, which I hope can be adopted, would be to start with a serene discussion of user perceivable services that (a) NAT44s contribute to offer, (b) may be desirable in IPv6. They would be discussed ONE BY ONE, e.g. starting with: - Easy renumbering - CPE multihoming - Web-browsing privacy - implicit firewall For each function, available and proposed techniques should be identified and evaluated. Only after that, can it be decided what is worth standardizing (if any), and whether the result should be called NAT66 or otherwise. Regards, PS: For the IPv6 Web-browsing privacy in, I renew the proposal of the following approach. CPEs perform a reversible scrambling, with a secret local algorithm, of: (1) the part of an internal address which is after the ISP prefix; (2) bits 2-15 of the internal port number if it is a dynamic port. Anyone interested in the subject? |
_______________________________________________ nat66 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nat66
