It's a kludge. I can't see how it would be preferable in any way to a single global "namespace".
I have serious concerns with IPv6's practicality, even with 6in4 et al. Is trying to implement it just flogging a dead horse?
If somebody comes up with a way to actually get IPv6-only nodes widespread - not just nodes with joint IPv4 and IPv6 machines - then I'll have hope for mainstream IPv6. But while every client and server has to have IPv4 as well to be of any use, then what benefit does having IPv6 connectivity give it?
Adding features to operating systems is easy - IPv6 support has been in all the major OSes for years. Getting it out there wasn't that hard but getting people to use it has proven to be a different matter entirely, largely because it is not backwards compatible with IPv4 and because of the resulting chicken-and-egg situation.
With something like this, you could upgrade your OS, then upgrade you router, and see immediate benefit. Or you could skip upgrading your router and use an in-the-cloud provider instead. It's much, much easier.
I think we're going to see people explore application-layer proxying (things like Tor, my http://pagekite.net/, this SOCKS idea) more and more as the IPv4 shortage starts to have an impact. Sure it's less efficient and it may feel messy, but it works surprisingly well.
And that's it. There's no need for ISPs to route you IPv6, or to set up tunnels. No need for dual stacks and AAAA records. No need for service providers to provide IPv6 access to their servers.
SOCKS is less restrictive than NAT - the client can know about it and find their external IP/port, it allows incoming connections, it lets the gateway device authenticate users and give them differing quality of service, etc.
Just shut up. Do it. Do the transition. Yes, it will be hard. Yes, it might break shit. No, IPv6 is not perfect. But do this transition once and we have enough address space to make it to the galactic civilization level. 2^128 addresses is big. It's a one-time thing.
So it's in my best interests to wait until I have to.
Which means I'll continue to use IPv4.
Which means the sites I want to connect to aren't motivated to turn off IPv4.
Ad infinitum.
What would I propose for that?
* Using them more efficiently by having protocols support endpoint identification within a server (HTTP supports virtual hosting, SSL has spreading support for host identification in the SSL handshake process, SMTP has been fine with it for years, etc).
* Using incoming connection proxies / load balancers to have a small number of external IPs, connections to which are handled by a large number of backend servers
* Perhaps in the longer run, better usage of SRV records so that well-known ports fall into disuse, and server ports can be assigned by the administrator and then placed into the SRV record for that service, in effect making IPv4 addresses be 48 bits long.
