On Level 3’s blog yesterday, the topic was “chicken” as it relates to traffic interconnection, net neutrality, and the last mile. The chicken part was in relation to the game Level 3 says some last mile operators are playing when it comes to upgrading their interconnections with other network operators. It’s a nice analogy that reminds us all how easy it would be for runaway egos to bust the internet these days if we let it.
If there’s any doubt that the network neutrality battleground has moved upstream, it ought to be gone by now. By looking to recoup some costs while bypassing the need for silly things like interfering with specific traffic, network operators that fought against the FCC’s net neutrality regulations can do battle in the murky world of peering and transit where those regulations did not even try to apply. Comcast’s dispute with Level 3 a few years ago was an early case, and the Netflix/Cogent/Verizon news recently was another, but there are more we don’t hear about – or haven’t yet.
That doesn’t mean it’s evil necessarily, because in the world of peering and transit there have always been shifts in the balance of power. Who should pay who to send bits where? If the last mile operators flex its muscles on the top of the hill, those who used to flex theirs can’t really complain too much. As long as the system works, that is… meaning as long as the deals made aren’t fair and thereby damage investment in other parts of the global internet infrastructure.
So, how do we make sure this process stays fair? Do we ask regulators to step in and work their magic? I rather doubt we’ll get anywhere that way and even if we did it might not be as useful as we might think. But maybe there’s a better way…
It’s the same way that network neutrality has really been enforced these past five years, through public scrutiny. Let’s make it all public. Take peering and transit, and open it to public review – not entirely but in some form where we can monitor the bits and the money that changes hands. I’m not saying it should be rigid tariffs or try to reproduce legacy regulation in any way, but surely we can find some metrics that would work for the purpose that wouldn’t be a huge burden for network operators of all types to collect and report. Bit-miles, traffic ratios, other stuff, I don’t care — you tell me. Or since nobody likes to work under a microscope perhaps we only make things public when there’s an unresolvable dispute, where the side claiming unfairness can threaten to turn on the spotlights.
Just a little transparency would let us channel the outrage to where there is abuse, while still letting internet infrastructure evolve without ham-handed intervention — the way it always has. Where the market is limited and abuse is a valid fear, visibility is the only real cure. And perhaps along the way we’ll all learn a few things about how all our bits really traverse the world’s fiber.
That’s my game, I call it “Naked Chicken”. It’s just like chicken, except neither side has clothes and there are bleachers. Anybody up for it? Nah, I didn’t think so.
Let the people be the watchdogs. We know how to bark, and maybe even bite when we have to. And we don’t get gourmet cupcakes delivered to our office inside the Beltway by lobbyists. Let us watch.
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!
Categories: Government Regulations · Internet Backbones · Internet Traffic
Join Our LinkedIn Group
In one of the recent articles about the Cogent-Verizon peering dispute, a Verizon rep was quoted as saying they pay some ISPs for transit, and some ISPs pay them. If that is true, and if Verizon is getting full transit, then Cogent should be able to do the peering -transit end-run. i.e. if Verizon gets transit from NTT (just as an example – I haven’t checked any route servers to see if it is likely true), and Cogent peers with NTT, then Cogent can send most of its data for Verizon through NTT.
Peering disputes provide economic incentive for networks to get bigger. That generally means scale and efficiency, and in turn lower cost.
Agreed Rob. Sunlight is a powerful disinfectant.
This basically comes down to a powerplay Eyeball Networks vs. Transit Providers (especially the ones that have a CDN product).
Shouldn’t the eyeball networks which are getting paid by their (access)customers being able come up with a business case for providing the appropriate connectivity to their users ? They have provided high bandwidth access to their users and are now complaining that it is to expensive to handle the traffic on those pipes ?
And shouldn’t the combined Transit Provider / CDNs have a case that takes into account that they are getting paid by the content owners for their content to be distributed ? Akamai and LLNW at least peer on most exchanges while L(3) wants to see $$$ for Transit.
Back to the roots ! Peering seems to be the easiest way for everybody. Not just that it is easy – it scales pretty well and we have many locations across the globe which are ideally to implement it. Dear Transit providers and Eyeball heavy networks – please fix your business model in order to support this.
All those “special deals” are potentially disruptive to our Internet Ecosystem. The next Netflix can’t be created w/o having all packets treated the same.
Uhm, if i read that correctly this Level3 lawyer said it was my fault, as the Eyeball ISP, that my customers experience poor performance to (netflix) for instance. I provide say a 5mb circuit to my customer ~ i Purchase 1GB from level3 as my Transit provider, i use for example, 300mb on the 1gb to Level3 ~ so i have 700mb “in essence” available, and its My fault the customer experience is bad to (netflix)?
I have to be reading that wrong 😉
Yes, you are reading that incorrectly.