This article is more than 1 year old
IPv6 is built to be better, but that's not the route to success
Why won't you love me, sobs perennially spurned protocol
Opinion In the World of Tomorrow that's always 10 years away, Linux dominates the desktop, quantum computers control the fusion reactors, and all Android phones receive regular system updates. And the internet runs on IPv6.
This sort of talk irks IPv6 stans, mostly because it's true. They are serious-minded, far-seeing, sober engineering types who are both baffled and angry that IPv4 still rules the world in 2022. This is not how it was supposed to be.
IPv4 was designed by expert prophetic dreamers more than 40 years ago to be future-proof, but the future it actually created outstripped their dreams. IPv6 was the engineers' answer, born from a decade and a half of experience, and solving IPv4's undeniable routing, addressing, security and performance problems at the unprecedented scale it was being asked to support.
They built IPv6 to be everything the world needed in a 21st century network. The 21st century network was too busy to care. Nearly 25 years on since the protocol's launch, it carries maybe 25 per cent of global network traffic.
It has not swept all before it. For those who follow such things, one by-product has been two decades of unhappy articles – the latest heart-rending missive is here, berating the makers of networks for not embracing the scheduled future. "It's as if they don't see the need," is the mournful cry.
Well, they don't. And that's not their fault.
IPv6 made two very important, interlinked assumptions from the outset: that the world would need more than the four billion addresses IPv4 could support, and that this plus a lot of less important factors justified an incompatible new stack that fixed everything. The first was correct, but irrelevant, and that made the second just plain wrong.
The irrelevance of the first observation came through the clever bodge of NAT, Network Address Translation. It's like a local switchboard for telephone extensions: a company with 10,000 employees can have just the one public telephone number. If you can do that, why rip out the whole phone system to give everyone a personal number? In retrospect, relying on IP address exhaustion to make IPv6 happen was doomed anyway, as it implied a user base far too huge to move. Making IPv6 incompatible with IPv4 just made that mountain higher, with the result that there are now tens of billions of IPv4 devices to convert. Not happening. Not without a really good reason – and more efficient routing tables ain't it.
The race does not always go to the swiftest. Many senior netheads will remember the tech wars of the 1980s and 90s, when wave after wave of superior computer technologies tried and failed to bring down the oafish monolith of the IBM PC architecture. Call it the Amiga Syndrome.
In the end, not only did innovation fail to overcome inertia, but the crude, neo-brutalist 8086 evolved to take over every specialism from video editing workstations to supercomputers. Incompatibility equals obsolescence.
- Can Rust save the planet? Why, and why not
- It's time to decentralize the internet, again: What was distributed is now centralized by Google, Facebook, etc
- APNIC: Big Tech's use of carrier-grade NAT is holding back internet innovation
- OK so what's going with these millions of Pentagon-owned IPv4 addresses lighting up all of a sudden?
But IPv6 isn't quite the network Itanium. There is no doubt that an IPv6-only planet would be superior, more efficient, support a bigger variety of services and have better security. The trouble is, until you get there, the opposite is true. If running an IPv4 network implies a certain amount of resources required and a certain threat landscape to manage, then adding a parallel IPv6 network means adding to those costs and liabilities. It doesn't get you much in return.
There are plenty of places where running IPv6 as a core protocol makes sense, especially if you're managing a huge estate where you make the rules, but they get less sensible as you move towards the edge.
It's not impossible, even then: some mobile operators have an IPv6-only network that does IPv4 translation and tunnelling – although it's messy and hardly confers a compelling commercial advantage. You have to dig very deep into your phone settings to find if your carriers' APN is pure IPv4 or not. And do you know which UK ISPs do IPv6, which do not, and which are dragging their feet? You do not, nor do you care, nor should you.
Until IPv6 confers a palpable advantage to end users, which means compelling services that cannot be replicated on IPv4, it can only hope for a tap-drip of growth as each of a thousand business decisions on cost versus benefit goes its way.
If IPv6 fans want to change that, writing regular letters of unrequited love will do nothing but evince wistful pity among the rest of us.
Instead, they must build the services for which their protocol is the essential motor. Demonstrate what we could have if only we saw the need. Create the world they want – not give us the bricks and say "There you go."
It's harsh, it's unfair. It's reality. ®