Disclosure: I am one of the world’s biggest fans of, and greatest advocates for, IPv6. In the words of rapper 50 Cent, “I love it like a fat kid loves cake.” Anyone I have ever been able to corner in a room knows this to be true. That being said…
I just finished reading [yet another] article trumpeting the reasons to adopt IPv6. And while I appreciate and agree with the authors passion I was left wondering who, from the pool of those not currently inspired to move to IPv6, would be compelled to take action.
IPv6 is twenty years old! The 6-bone, an IPv6 test network, was decommissioned almost a decade ago (6/6/06). And, as of today, Google is reporting that just under 10% of the traffic on its servers is using IPv6. For now, I’m going to assume that Google’s traffic reporting represents a reasonably fair estimation of our planets IPv6 state of affairs.
Twenty years. Ten percent. We suck.
Why, oh why, have we taken so long to do so little (although, admittedly, things have been gaining momentum in recent years)?
I can go on and on with a variety of reasons that are all contributors: NAT & private IP addresses, Carrier-Grade NATs (CGN), continued availability of IPv4 addresses (which is less true every day), the horrific notion that NATs are a mechanism of security, a lack of urgency to fix something that isn’t obviously broken (we’ve been using IPv4 with relative success for this long, right?), fear of the unknown, the lack of a killer app that inspires Earth to migrate, and in the earlier days, the lack of enterprise support (firewalls, messaging services, web servers, etc.).
The future of the Internet is IPv6! Ever heard that one before? That statement is a borderline marketing slogan because of one really important fact: it’s true. IPv4 is not sustainable for the Internet. How many fours do you want to append to NAT (NAT44, NAT444, NAT4444, etc.) before you concede that hiding behind NAT is the wrong direction in which to travel?
The reasons listed above for why we haven’t migrated is far from exhaustive but I left out the biggest problem: Money. Lots and lots of money. Hardware costs, software costs, re-writing old code, training costs. Surcharges, Over-charges. Additional over-charges. It’s a lot.
Suppose for a moment that you and everyone else fully migrated to IPv6 today:
- What would happen when you opened your web browser after the migration was complete? Answer: A web page would open, just like today.
- What would happen when you clicked send on an email? Answer: The email would be sent, just like today.
- What would happen when you needed a site-to-site VPN across the Internet? Answer: You’d set one up using IPSec or TLS, just like today.
- What would happen when you needed to remotely administer that router? Answer: You SSH into it, just like today.
- What would happen when you fired up your game console after a long day at work so you could grab a little Halo action? Answer: You’d play Halo, just like today.
Migrate to IPv6 and your stuff (web apps, VPNs, messaging, web browsing, games, etc.) will, from a somewhat high-level perspective, be just like they are today. So much so in fact that I’ll wager that a good number of those people Google says are using IPv6 don’t even know they are doing it. Their Internet experience is unchanged. That’s good! But wait, it’s also bad. Why? Because the desired result of an unchanged experience doesn’t provide a compelling incentive to migrate. Imagine this discussion:
Customer: “My stuff works today using IPv4 and some varying number of NAT’s, right?”
Me: “Yup.”
Customer: “And if I migrate to IPv6 it will still provide me with the same overall experience, correct?”
Me: “Yup, that’s right, too!”
Customer: “So why is it that I should want to migrate?”
Me: “Because the future of the Internet is IPv6!!!”
Customer: “Yeah, but I have to pay for it!”
And there it is. The future of the Internet is IPv6. But individuals, companies, in particular, have to pay for the migration. And the end result of the migration is a network that works pretty much like it does today. But the company will have to spend piles of money with varying degrees of enormity to get there. To the immediate and direct benefit of whom? The Internet, not the organization. That’s altruism! Altruism is the sacrifice of self for the benefit of others. For-profit corporations don’t work like that. So are you really surprised that twenty-years into it you still have huge swaths of the world staring at you with a stolid, bovine indifference at the prospect of an IPv6 migration? It seems perfectly rational to me.
I know, in part, what you are going to say. “They won’t be able to communicate with the rest of the world if they don’t migrate.” Eventually, that will be true. But they’ve made it this far without migrating. Kicking the can down the road for another decade (or longer) doesn’t seem unreasonable. After all, as much as I love to hate it, NAT works. Private IP addresses work. Tunneling works. Proxying works. We’ve got so much stuff that we’ve come up with over the years that another decade of foot dragging is hardly out of the question. And by that time IPv6 will be way older than IPv4 was when we decided it needed to be replaced. That’s so funny it’s not funny.
So what’s the solution? Sadly, I don’t have one. I’m not a fan of altruism but I sure do love IPv6 (and want to see a native v6 world sooner rather than later). Some people will surely suggest legislating the change. I’m not an advocate of that. Not even a little bit. I love IPv6 because I love the technical nature of it all. But router advertisements, extension headers and fancy new mechanisms of duplicate address detection don’t motivate the money people to start writing checks. We need something to better help us force the issue; a lack of IPv4 addresses isn’t working (yet!). If Facebook, Twitter, Google, Youtube, LinkedIn, Amazon, Yahoo, Pinterest and Wikepedia all came out and said, “Starting on 1/1/2017 we will only be reachable via IPv6!”, you may think the world would line up to follow. But I’m more inclined to think that replacements, still accessible via IPv4, would rise in their stead and they would just lose market share. Don’t count on that. Getting the big boys to lead the way in such an aggressive manner isn’t a solution, either.
What do you think the solution is? We need something better than a relatively hollow “future of the Internet” battle cry (despite its truth). We need better marketing. Apple has gotten really good at getting me to buy stuff I don’t really need. IPv6 needs some of that marketing magic in its corner.
Cheers,
Colin Weaver
