On 02/10/2012 16:12, Chris Grundemann wrote: > On Fri, Feb 10, 2012 at 15:13, Doug Barton <dougb@xxxxxxxxxxxxx> wrote: >> On 02/10/2012 10:22, Chris Grundemann wrote: >>> This is not about IPv4 life-support. >> >> Seriously? > > Seriously. > > The birth of a shared CGN space in no significant way extends the life > of IPv4. It does provide the best possible solution to a necessary > evil (CGN inside addresses). Ok, now THIS is a snark: Dude, don't bogart the good stuff, pass it around man. </snark> We were specifically asked not to reopen the debate on the necessity of CGN to start with, so I won't go there .... feel free to dig through the archives on all the reasons why this evil isn't actually necessary. >>> This is about providing the best answer to a difficult problem. >> >> The best answer is to make sure that CPEs that will be doing CGN can >> handle the same 1918 space inside the user network and at the CGN layer. > > Are you volunteering to buy everyone on earth a new CPE? If not, who > do you suggest will? My bet is that no one is willing to drop the > billions of dollars required - if they were, we could just sign > everyone up for IPv6 capable CPE and skip the whole debate... ;) Everyone doesn't need a new one. In fact, the only end users who need new ones are those behind ISPs that chose to ignore IPv6, "need" to deploy CGN, and can't actually get CGN done with existing 1918 space. I strongly suspect that the latter number will be very small, and what this allocation request is really about is to make lives easier (read, cheaper) for those that didn't want to invest in IPv6, and now don't want to have to think hard about how they do their network design. It's also in no way clear to me how many extant CPEs would break if CGN were deployed with 1918, and/or how many of the extant CPEs will have to be replaced anyway to do CGN in the first place. Repeated requests for hard data on this (by me and others) have gone unanswered, I suspect because collecting such data would also be expensive. Doug -- It's always a long day; 86400 doesn't fit into a short. Breadth of IT experience, and depth of knowledge in the DNS. Yours for the right price. :) http://SupersetSolutions.com/ _______________________________________________ Ietf mailing list Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf