Comments Elsewhere

Jason:

Peering agreements often do not, currently, require the various sorts of “neutrality” rules in how the packets are delivered once they are handed over at the peering point. get dropped for all kinds of reasons all the time, and nobody guarantees that every packet handed over to it will be delivered to its intended destination. Currently a lot of agreements offer a lot of leeway as to why packets might get dropped, or slowed down, or whatever. Craig is right that part of the reason for this is that FCC regulations controlled these agreements more tightly in the past. On the other hand, it has been years since those regulations were dropped and the companies have had plenty of time to renegotiate the arrangements if they didn’t like their network peers having that kind of leeway.

Personally, I would like it if non-slimy infrastructure providers would demand a fairly robust set of neutrality policies in their peering agreements. If such contracts were made, there would be no regulatory issue at all, since violating those neutrality policies would be a civilly actionable breach of contract. But until such a time as they do, I think there’s no justification for having the government force the kinds of peering agreements that I’d like to see on carriers that aren’t yet willing to make them. I wish I could change Comcast’s idiotic Acceptable Use Policy, too, but since I don’t own Comcast or their cable I’ve got no right to force them to provide better service.

Me:

[H]ow, exactly, does a content provider pay a carrier company for “access to the customers,” except by paying the carrier company for the use of its bandwidth? Isn’t the use of their bandwidth precisely how Internet providers provide “access to their customers”?

Craig:

The customers have already payed for the bandwidth.

Whom have they paid for whose bandwidth? We’re talking about interconnected networks here where not all of the infrastructure is owned by the same group of people.

In the scenario I sketched above, Craig is paying Nicenet for access to their bandwidth, but Craig has no agreement at all with Scumnet which would guarantee him the use of their bandwidth just by paying Nicenet what he pays them. Nicenet and Scumnet have an agreement, but the terms that they agreed to (like many actually existing peering agreements) makes no particular guarantee that Craig can expect any particular quality of service from Scumnet. There’s no apriori mandate that carriers must charge at only one point in transit; that’s often the wise thing to do, and the convenient thing for customers, but it’s not an injustice against drivers that you have to pay to drive on I-90 in Ohio, and then pay again to drive on it in New York.

What in the hell am I paying Verizon for if not access to content? Their service is useless without content.

You’re paying Verizon for access to their network. Paying them for access to their network does not entitle you to also access AT&T’s network unless AT&T has made an independent agreement (either with you or with Verizon) that would guarantee that. If there is such an independent agreement, then this is a civil matter rather than a regulatory one. If there is no such independent agreement, then you’ve done nothing which would entitle you to force one on AT&T.

Of course, it may be that part of the reason you find access to their network useful is that you also get access to AT&T’s network as part of the package, thanks to the current set of agreements between carriers. But so what? AT&T does not owe it to you not to renegotiate those agreements if they see fit. And if AT&T and Verizon agree to different terms, which makes it so that access to Verizon’s network is no longer useful to you, then the thing for you to do is to stop paying Verizon for a service that you no longer find useful.

The telecomms want to artificially degrade that already paid for service so they can charge content providers for restoring it.

You’ve paid your telecom for this month’s service, not next month’s, and if they decide to change their service or to scrap the whole damn thing at the end of your next billing cycle then you have no claim against them.

“Artificially degrading service” is, incidentally, a very nearly useless term in this discussion. Every ISP in the world has an Acceptable Use Policy that “artificially degrades service,” in the sense that it imposes restrictions on what you can and cannot do on their network, and to what degree, which have nothing to do with the technical capacity of their networks. Most broandband ISPs, for example, prohibit residential users from running servers on their network, even though the line is perfectly adequate for casual web hosting needs. They do so because they want to sell their business service, which allows for running servers. Is this “artificially degrading service”? In a sense, but so what? It’s their network and they have a right to set the rules — even if those rules turn out to be inconvenient, foolish, avaricious, etc.

Think the telecomms don’t like the FCC? You are dreaming.

Who made this claim? I’m well aware that the telecomm companies are government-backed oligopolists and that they are quite happy with restrictive regulation as long as it comes from an agency they’ve captured. But so what? That doesn’t change the fact that this:

And let me also say that while I am sympathetic to the general deregulation arguments, I am not sympathetic to those arguments when they come from industry players who are quite happy with regulation that creates artificial barriers to entry in their market — exactly like the telecomms.

… is a particularly crass example of argumentum ad hominem. The fact that big telcos, cable companies, etc. are hypocritical in putting forward an anti-regulation line does not make that line false, or the arguments for it any less capable of being assessed on their merits.