Nick.. I don't disagree with any of your arguments for IPv6. They're all valid.
There were great arguments for BetaMax. There were great arguments for AppleOS.
Neither of them were compelling enough to cause end-users to move away from VHS or Windows
Cloud providers....? Here's a scenario that bears watching vis-a-vis Software Defined Networking. (btw, have you noticed that all of the SDN examples still use IPv4?)
If a cloud provider says.. "Yep, we can do SDN with your data center, but you'll have to use IPv6." ..... what are the odds that cloud provider will get the business vs another provider who doesn't have that requirement?
Now. compare that with "Yes, we can DO IPv6, if you need to at some point in the future, but it's not required today"...
Well, then, we're back to the end-user being the driving force. Even though the provider "can do", that doesn't mean that the end-user "ever will".
The problem is, IPv6 isn't a viewed as an "upgrade" to IPv4, it's a competitor. There is zero perceived value in "upgrading to IPv6", but there is significant challenge viewed in "converting to IPv6".
While that wasn't the original design, it's become the de facto reality.
My point is simply one of a more practical nature.
Until there exists a compelling reason to abandon IPv4 in enterprise networks... it's never going to happen.
That simple it is.