@p >As does IPv6.
IPv6 just requires software changes really and the software changes have already been made or are possible - the main blocker is proprietary software as always.

>The auto-gateway proposal is a botch, but would work without breaking existing systems.
Previously, gateway protocols were implemented, but it was soon determined that it native IPv6 was a much better choice.

>The point is to solve a problem (address exhaustion) without requiring a complete overhaul of everything (IPv6).
The designers of IPv6 realized that address exhaustion was impossible to solve without a complete overhaul, so they did so.

IPv4 address exhaustion has been dragged out far too long already via nasty hacks like CIDR routing, NAT and CGNAT - anything more would lead to unusable brokenness, so it's time for IPv6.

>Is the solution some sort of drop in network engineers' salaries?
A network engineers salary should be dropped if they refuse to understand an improved version of a protocol that works as it was originally intended to.

>IPv6 sucks.
IPv6 is great in my experience.

I plug the computer in and it autoconfigures and just works via SLAAC, meanwhile, IPv4 takes ages to come up thanks to DHCP brokenness.

Most complaints about IPv6 seem to be skill issues where people can't figure out how to use the computer in front of them to copy and paste the address or DNS.