@Kasperd I am surprised, how strongly you like ipv6. I can't see, why would be the ipv4 address space depleted. I think, NATs and application-level proxies solved the problem decades ago. They work also as pretty good firewalls as side-effect. Furthermore, the NATting firewalls, mixing many unrelated sessions, make the work of the network eavesdroppers harder. Furthermore, I can't see why the 128 bit address space wouldn't be depleted just as the ipv4 "did".

@Kasperd There are no 4billion computers on the whole world, and there is absolutely no need for all to be able to address eachother concurrently, rather it would be a serious security problem. I think it is pretty okay, that the majority of them are behind different NATting firewalls.

@Kasperd I think the problem of the IPv4 address space depletion is long solved by VPNs and NATs. I've read Tanenbaum's network book and I know his arguments, how and why was the ipv6 created and why so as it was. From his arguments it is very clear that he never ever understood the concept of the NATs and the VPNs, and nobody understood it in the whole committee. Furthermore, his arguments are totally on a theoretical level, he thinks in the born-dead ISO/OSI model, and it is very visible that

@Kasperd probably he never had to configure a router or to design a company network. On his whole book I see a strong theoretical attitude, with only a very few practical affinity - now the problem is that it is hardly engineering thing, and the theory, the science simply doesn't exist. You can have very high level and useful theories about the elemental particle fields, but not from the network addressing standards, it is simply not a theory. I think, Tanenbaum should have gone to a theoretical

@Kasperd physicist and so he became probably a very successful one.