US vs European Peering Models
I am not sure how up to date this information is, but here is an interesting link that highlights how peering is structured in the US vs many European locations:
Some of the terms and relationships look pretty dense and esoteric to my lay eyes, but maybe that's a personal deficiency.
Dan Rayburn asked netflix to put up or shut up when it came to their claims about verizon slowing down traffic. It looked like a seed story for Verizons eventual post blasting netflix for directing too much of their traffic down one of the congested transit providers into their network.
Dan, as you can read from that link, is a very vocal supporter of the current peering models used in the US, and came out and tasked netflix to provide a rationale for why paid peering should be changed to some other model.
THIS is where no public case is being made one way or the other. Netflix has covered the edges of the issue, but I have yet to see people that were deeply familiar with the subject matter dive into the pros and cons of our current model and other peering models.
This seems like it ought to be an empirical question, what type of model produces the best results? Can we even agree on what the best results are? I chafe at Dan's frenzied defense of paid peering to ISPs because he seems to be suggesting that because that is the way the current peering models actually work, that is the way they should work.
It's a pretty standard is/ought leap of logic, and I don't see how one follows from the other. I WANT someone to make a case for or against paid peering vs some other more cooperative model, someone with more depth and breadth of knowledge than most of us have on these issues. Preferably not an industry shill, someone who can give an honest accounting of each model with the pros and cons. I might even be convinced that paid peering to an ISP will lead to better results and more of what we all want, but NO one will make that freaking case, and it's incredibly frustrating seeing all of us get into rages based on incomplete pictures of what is going on. We are all affected by how these deals shake out, but we're staring through a darkened glass.
The am deeply skeptical about the notion that we ought to give large local monopoly/duopoly last mile ISPs the same sorts of carte blanche that other tier 1 transit providers share when it comes to peering, since the latter group is constrained by a competitive market while the former have nothing around to check their activities. If that is an obtuse view, then someone should make a case about why that sentiment I hold is so wrong and misguided.
The response of:
This is the way it works, Deal with it.
Is not a good enough answer, though we may have to deal with it either way.