Scott Hogg wrote up a great summary about the North American IPv6 Summit, hosted at LinkedIn in Sunnyvale, CA from April 25 to 26th. I don’t need to repeat any of the great work he did in documenting the event but instead wanted to focus on one presentation that was given by John Curran, the CEO of ARIN.
While John was representing himself and his own opinions I found his thoughts and comments worth contemplating more deeply. John’s presentation was titled “The future of the Internet, IPv6 and the long tail” and he gave a great retrospective of where we started, where we have gone and what the opportunity ahead is for the Internet at large. But there was a specific point that John raised that caught my attention. It starts with the slide titled:
What do we want the future of the Internet to look like?
More specifically, a bullet point which states:
Planning for [a] long-term predominately IPv6 Internet requires providing a differentiation of Internet service despite a lack of differentiation in technical capabilities over IPv4.
What is implicit in this statement is that we only solved one problem with IPv4 in the development of the IPv6 protocol (the size of the address space) and that a clear lack of differentiation of technical capabilities is a continuing challenge. From a big picture perspective, I agree with this statement. There are small technical differences that are unique to IPv6 but nothing earth-shattering and certainly nothing as revolutionary as moving from the much more proprietary Banyan Vines or IPX/SPX to the more open routed-protocol solution IPv4 offered: Back then, moving from any protocol to IPv4 meant you could connect to the broader Internet. You could have access to the great collective knowledge (or wasteland, depending on your perspective) that was made available by having a standard protocol to communicate and interoperate with. In short, there was a lot of value in adopting IPv4 – you got access to the Internet!
The next slide offered the most interesting statement. John states:
“The IPv6 Internet is the seed of a future Internet, and it does not have to replicate all of the mistakes of the IPv4 Internet (e.g. lack of routing hygiene, which enables DDoS and botnet attacks, blacklists-based spam filters rather than whitelist mediated email networks, etc.)”
It is clear that we (the collective “we”, that is) have an opportunity to make the IPv6 Internet superior to the IPv4 Internet and make it have a compelling value in the market, which will lead to greater adoption. Imagine if moving to the IPv6 Internet meant a large reduction in SPAM and a working formal process to resolve the little that remained? What if all services were significantly more secure? That all players participating on the IPv6 Internet had to meet higher standards of routing hygiene and demonstrate better filtering and peering configurations? What if retail and banking systems would feel more confident doing business over an IPv6 network versus an IPv4 one? What if our Governments only trusted IPv6 traffic due to the networks being more secure, reliable and DDoS and botnet free?
John was not-so-subtly telling the IPv6 community what our next cause should be to help increase adoption and to make a real impact. If we want the IPv6 protocol to really be adopted globally then networks that run on it and services that use it have to offer substantial market value differentiation over those remaining on IPv4. We have the opportunity to start making standards and operational practices that would enforce the behavior we expect to see that will make the IPv6 Internet superior to the IPv4 Internet.
While I am typically not in favor of draconian codes and standards, the points John is making are hard to ignore. While the adoption of IPv6 today is driven by those that truly need more address space, it is hard for anyone to effectively argue that most enterprises need to convert to IPv6: most enterprises don’t have a lack of IPv4 addresses! However, most enterprises would gladly adopt a new protocol if it meant significantly reduced DDoS, SPAM, botnets and other malicious traffic. In fact, downstream business needs would drive more and more adoption because the real and virtual bandwidth requirements for regular business operations would be greatly reduced. The value added by such network and operations hygiene leading to a much cleaner network would add great value for enterprises and likely drive more of them to adopt IPv6.
So, the next logical question is how do we get there? There are some clear calls-to-action that John highlights like:
“The question faced is whether the future Internet enabled by IPv6 is simply a network protocol change for bigger addresses or a conscious effort to build a Better Internet... The latter requires less technical work in standards bodies but far more coordination in operational norms in bodies such as this one...”
So how do we accomplish more coordination? Who sets the agenda and sparks the ideas? Will companies like Google, Facebook, Microsoft, Amazon, LinkedIn, Alibaba and others have to set these new improved standards of operation? Will they have to drive the big ideas forward first so that others MUST comply to interface with them? I must admit, that seems to be the easier way. It seems draconian, almost anti-Internet in nature yet if these organizations could truly make a more secure, reliable, trustworthy Internet using IPv6, would the end result be worth the price? I don’t know the answer to that. I struggle to see how technical task forces and working groups can accomplish what these large Internet properties can do with a few quick handshake agreements around a table. Perhaps our goal should be to articulate what we want from our next generation Internet Protocol and push those that operate some of the largest networks to drive toward those goals.
What do we want? When do we want it? There are a lot of questions to be answered and discussion that needs to happen over the next year and into the future but I believe our window to make this sort of change is narrow. IPv6 is growing and soon it will hard to prevent the same poor operational behavior and bad actors we have on the IPv4 Internet from appearing on the IPv6 Internet. So, what is the next step for all of us? You tell me! What feedback do you have about this? Let us know and perhaps the IPv6 COE can help amplify our collective voices for the common good.
You can find me on twitter as @ehorley and remember…