Implementing IPv6 isn't for the faint of heart. Far from being just a technology upgrade, IPv6 is akin to building a completely new network that doesn't talk easily to the existing IPv4 network.
In addition to installing a router fabric that can understand IPv6 addressing, operators have to ensure that there is a market for IPv6 services -- and that's going to be a challenge. Let's look at why.
IPv4 is ubiquitous, and isn't quite dead yet. In fact, all of the hyperbole associated with IPv4 address exhaustion misses the mark by ignoring the fact that there are quite a few unused IPv4 addresses still available. Many network operators acquired massive blocks of addresses years ago and routinely reuse addresses as websites and subscribers come and go.
Also, many private networks use network address translation (NAT) on the interface between their networks and the public Internet. This allows them to dynamically use one address to support many users.
The reality behind address exhaustion is that what is running out are large contiguous blocks of addresses (the so-called /8 blocks) that enable efficient address resolution. So as IPv4 address blocks become fragmented with reuse over time, IPv4 will impose unacceptably higher overheads on routing infrastructures. It is this gradual fragmentation, more than simple IPv4 address exhaustion, which is likely to drive IPv6 adoption.
In the meantime, the IPv4-based Internet is alive and well and continues to deliver interesting content and places to go for the average consumer. So far, consumers have been able to get to what they want using IPv4. The question for operators is: Do consumers want or need IPv6 services to get to content they can't access now? At least for now, the answer is no.