Microsoft will go open source and Apple will be selling reasonably priced computers before you need to worry about IPv6, so don’t waste your time drawing up plans to implement the next generation Internet protocol.
The principal raison d’etre of IPv6 is that the world is running out of IP addresses, but the truth is that there’s loads of them about – more than enough to last till long after your grandchildren have retired.
At the beginning of the year there were about 925 million free IP addresses. That’s a very large number. And since about 100 million a year are being handed out, we won’t run out for almost a decade at that rate
But that assumes that people are stupid, and fail to alter their behavior when it becomes sensible to do so. Since people aren’t stupid–most people, anyway–there’ll be free IP addresses for far longer. Here’s what will change:
The last time I looked, there was a whole bunch of organizations, like Eli Lilly and MIT to name just two, with Class A (or /8 in Classless Inter-Domain Routing (CIDR)-speak if you prefer) networks made up of almost seventeen million individual IP addresses each. Now you have to ask yourself: do they really need seventeen million IP addresses? Each? What’s Eli Lilly planning – individually addressable Prozac pills? I don’t think so. If and when IP addresses really start to get scarce, many of the IP addresses in these Class As will be reassigned. Perhaps they’ll be appropriated, maybe there’ll be a market and blocks of IP addresses will be bought and sold. Who knows? But something will happen.
And let’s not forget Network Address Translation. Thanks to the wonders of NAT, each IP address can be shared by many, many other machines. A Class A network can connect billions of individual hosts to the Internet using NAT. That’s enough to give every man, woman and child in America an address for a desktop, laptop, network printer, IP phone , cellphone and even a toaster if they want. Russia? China? Brazil? A few billion each should be plenty for them too.
Even using NAT to share a single IP address between just two hosts, that would mean twenty years before we run out of IP addresses at the current rate of consumption, and by sharing an IP address with ten we’ve got a century to go. Now I don’t know how computers will be communicating with each other in a hundred years time, but I’ll bet it won’t be using IP. Robots made with nanotechnology will have devised something better, no doubt.
So forget about drawing up implementation strategies and working out which bits of your hardware and software need to be scrapped or upgraded. IPv6? Put your feet up: it’s never going to happen.