We've spent the past few weeks with at first some knowledge of the FCC's "Slow Lane" plan (to call it "Fast Lane" is to be disingenuous), and later with the full plan. In that time, there has been a massive outpouring of disagreement -- clearly plenty of it on this side of the screen. But with more time spent thinking about this awful plan come greater distaste and outright concern that this terrible set of guidelines may actually end up as the law of the land.
For instance, this plan turns ISP peering points and their own internal network interconnections into straight-up revenue generators in the worst possible way. Say that Comcast's customers see sluggish performance and potential timeouts when trying to access resources on the other side of a peering point or even when accessing an internal interconnect into large portions of New England. In a normal, free-market situation, Comcast would negotiate terms with the Tier 1 provider for an additional 10G port if it's a peering point, light up more fiber, or otherwise upgrade its network to alleviate poor performance and congestion.
[ Also on InfoWorld: The next weapon in the War on the Internet: the brownout. | Pick up expert networking how-to advice from InfoWorld's Networking Deep Dive PDF special report and Technology: Networking newsletter. ]
This is a normal cost of doing business, and those same fibers have carried 10Mbit through 10G just fine throughout the years. Remember, bandwidth is cheaper now than ever before.
However, under the new FCC proposal, upgrading those connections would work against Comcast's financial interests. The proposed rules would in fact reward Comcast for letting its networks become congested. The more congested the pipes into Comcast's network become, the more money Comcast is likely to make.
Why? Because once Comcast customers begin experiencing problems with accessing resources on the other side of that congestion, Comcast can demand a ransom from the content provider to give it "Fast Lane" access to its customers. This is straight-up extortion, but it would be fully permissible under the FCC's proposal. Comcast can continue selling that congestion over and over until the only traffic that makes it through the 10G link has been paid for on both ends. Any small blog, small service, VPN connections to the corporate office, or an innovative startup company will only be accessible at modem speeds, if that. Those who have paid for clear passage will be consuming 99 percent of that 10G pipe.
Then, and only then, will Comcast add a second 10G peering connection, and it will potentially sell access to that peering point to both sides of the connection. "Want faster access to the wider Internet? Only $10 more per month!" will be the pitch to its downstream customers. On the other side, Comcast will sell "Fast Lane" access through that same 10G pipe to the content providers as fast as it can, continuing to squash the rest of that traffic into the ground.
And so on and so forth. The FCC may have vague language about charges being "commercially reasonable," but we all know it will have zero impact on reality. We only have to look to Comcast's refusal to participate -- at no cost -- in Netflix's Open Connect program that would significantly reduce the bandwidth used by Netflix and provide better service to its customers. Comcast very clearly can and will harm its own customers in order to extract large sums of money from companies like Netflix. This is what happens when you have a monopoly.