At 9:30 AM Eastern Time on Friday December 4th, the legal gunslingers representing US Telecom and the FCC will show down before a three judge panel of the DC Circuit Court; but you knew that. This promises to be the telecom policy event of the year, anticipated since President Obama made his infamous YouTube program urging his appointees to change course and drop the nuclear option on Internet Service Providers. It’s likely that it won’t be the end of the road for the controversy over how to regulate Internet services, as most outcomes at the appellate level beat a path the Supreme Court. There are some scenarios that don’t, because the appeal has several moving parts.

• One issue is the FCC’s failure to provide adequate notice of the rules it ultimately passed. Before the Genachowski FCC passed its Open Internet rules in 2010, it presented the proposed text to the public for comment. The Wheeler FCC not only didn’t do that, it offered a general approach that it ultimately abandoned in favor of the Title II nuclear option that’s riddled with complexity, enforcement issues, and over-zealousness. It’s interesting to speculate about how the Net Roots activists who flooded the FCC’s inbox with short, highly emotional comments would have responded to the actual rules. They got what they wanted, apparently, but there’s a tendency for activist groups to want more.
• Another issue is the lack of factual evidence of wrong-doing by ISPs on anything like a significant scale prior to the enactment of the new rules. My advice to the FCC before the rule-making began was to be clear about what problem they wanted to solve and to apply the least burdensome solution. Because the final rules weren’t accompanied by a thoughtful analysis of actual harm suffered by innovators, the public, or competitors, the agency clearly didn’t follow my advice (or that of many others) and simply adopted the politically appealing solution to a problem that remains in the realm of the hypothetical. The ISPs haven’t been shown to have abused their power by acting in ways that harm the public interest.
• The FCC also failed on the factual predicate side by not conducting a formal economic analysis of the status quo and the likely consequences of imposing the most restrictive set of regulations the US has ever seen on Internet services. Instead of an economic analysis, the agency simply offered conjecture about incentives for misbehavior.

The parts of the order that concern me the most may not even be discussed at all in the oral arguments because they relate to the actual technology the agency seeks to constrain. On this front, the order relies mostly on unstated assumptions about the technology arc, the importance of competition between mobile and wired networks, and the nature of the service provided by ISPs. Even if the FCC accidentally found the right answer to all the pressing problems in Internet markets without presenting a thorough analysis, the nature of technology is such that the market will be very different five to ten years from now.

So how do these rules look if we begin to see a rash of cord-cutting by Internet users who drop wired broadband for mobile-only services that are much richer and more powerful than they already are? The general purpose of the Open Internet order is to preserve the Internet status quo, and even the FCC doesn’t have the power to prevent technology from improving and user preferences from changing.

So there is one scenario in which an appeal of the DC Circuit’s ultimate decision doesn’t go to the Supreme Court: If the court upholds the FCC’s hardline on wired broadband services but vacates the portion that applies Title II to mobile service, there probably won’t be an appeal. The FCC will be able to declare victory over the cable companies, which accomplishes the political purpose that motivated the president to speak out. The carriers who provide both wired and mobile services will be pleased because they realize that their future is mainly on the mobile side, so they won’t be inclined to rock the boat.

If the rules are either uphold for the most part or vacated for the most part, an appeal becomes more likely. But the court can also rule that the FCC failed to provide proper notice, in which case the FCC will probably issue a further notice, take comments from the net roots, and then impose exactly the same rules with extra vigor. Agencies can be vindictive, so I would expect this possibility to include additional restrictions meant to punish the appellants for causing the FCC to lose face. That sounds petty, but it’s what I would expect. They’re only human, after all.

The issues I have with the order pertain mostly to the ways that the FCC defines Internet service and the representations it makes about how the Internet works. These are important because they undermine the enforceability of the rules in the real world. If a service provider were to be sanctioned by the FCC for failing to operate its business the way the FCC wants it operated, the provider could very well claim that it doesn’t provide the service the order defines as “broadband [public switched network] access service.”

The order doesn’t use that phrase, of course; it says “broadband Internet access service” (BIAS,) but it defines the Internet as a “public switched network” virtually identical to the traditional public switched telephone network that was the target of Title II when it was written in the 1996 Telecommunications Act. One way to argue that is to question the “access” part of BIAS. There’s a long history of meaningful distinctions between “access” and connectivity in the Internet space. Web services don’t purchase access to the Internet, they pay for membership in it, for being present to those who desire to use their services. If they pay for access at all, it would be access to users rather than to the network itself. This is because the Internet is composed of computers with multiple paths to each other and not simply of wires and switches.

So ISPs don’t really sell “access”.

ISPs also provide services that go beyond simple connectivity between end users and web servers. They facilitate peer-to-peer transactions in which there’s no line between clients and servers as there is on the Web. In this scenario, where each connected computer is both a client and service, the assertion of access makes no sense.

And there’s the question of DNS, a service that’s inextricably bundled with connectivity but performed in the interest of both the ISPs’ customers and the third parties with which customers communicate. DNS allows domains to register not only their Internet Protocol addresses but a host of additional information such as mail server addresses and security information. DNS validates the authenticity of at least some of the information it serves up, a task that has nothing to do with access or simple transmission.

If the FCC had to create a fictional network in order to impose Title II, it’s going to have a nightmare enforcing its regulations on real networks. This is probably the reason that the FCC has given passes to Comcast and T-Mobile for exempting certain video-related services from Title II regulation. If the FCC can’t prove that ordinary Internet services fall within the scope of its rules, it certainly can’t prove that video streaming does.

Some have argued that the 2015 Open Internet order protects incumbents from competition by upstarts. That’s a bit cynical, but its kernel of truth is that the order protects at least some incumbents: Those who operate substantial private networks were given a gift.

The courtroom will be packed, but there will be an archived recording available to the public by 3:00 PM. The DC Circuit has been known to provide real time audio streams as well, which would be available here if one exists.