Sascha Lenz wrote:
c) don't need NAT
Of course i'm aware that such things as 6to4 etc. might be called "NAT" too and might be needed indeed :-) It's just not the point of this thread, don't want to complicate it now.
I'm afraid it's you who wrote:
b) don't like NAT c) don't need NAT d) don't use NAT
I think the more important thing is to show that there is little to no support for his specific approach (i hope).
My apprach is to accept NAT, including legacy ones. Note that, as is described in the ID, end to end NAT can and will be upper compatible to legacy NAT with ISP and user opt-in. If both GW and an end host deploy end to end NAT, the host can enjoy full end to end transparency. Otherwise, the host is as if it is behind legacy NAT. Masataka Ohta