It has been a difficult week for supporters of online neutrality, as the Trump Federal Communications Commission's decision to strip internet neutrality rules was upheld by the DC Circuit Court of Appeals.
It was a fairly narrow victory for the FCC, as the court said it was required to defer to the agency's judgment, and bound by the precedent set in a controversial case in 2005 called NCTA v. Brand X ( or just Brand X in brief). And the court said that the FCC can't block states like California from writing their own net neutrality laws, so that's where the fight moves next time.
But what really and truly stands out about the DC Circuit's decision in this case: federal-level net neutrality has turned into a legal quagmire with almost no relation to the real problems common people face in the Internet access market.
The heart of the net neutrality debate is incredibly simple and easy for almost anyone to understand: Do you think ISPs should have the power to block, pull or otherwise interfere with internet traffic outside of normal network management? Most people don't believe it ̵
But because the struggle has been going on for so long, and the rules have been imposed and taken away so many times under different legal theories, the actual trial and legal questions are miles away from the very simple political question.
Instead, the legal side of net neutrality has become an exercise in lawyers making fine-grained arguments about whether washing machines can call, whether consumers with a single broadband provider still experience the benefits of competition, and whether federal or non-regulations can override state law if federal the regulations do not actually exist. "Is it good if AT&T can comfort Fox News while streaming CNN for free," never actually comes up, though that's the basic political question. It is deeply frustrating.
But it is also revealing, because it makes it clear that net neutrality needs to get away from lawyers and judges and be written into the law itself. And since the court ruled that individual states could enact their own net neutrality laws, it seems like that's exactly what's going to happen.
But let's go through it – you can download a copy here to follow.
Still fighting the ghost of Antonin Scalia
The last 40 years of the history of net neutrality have largely been a battle over whether the Internet is an "information service" (like old-school "walled garden" AOL or Prodigy, regulated under Title I) or a regular carrier "telecommunications service" (such as landlines, regulated under Section II).
The most important precedent here is the 2005 Brand X decision: The Supreme Court ruled 6-3 that broadband Internet was an "information service." Brand X is also known for having a searing dissent from the late ultra-conservative Justice Antonin Scalia, who thought it was astoundingly obvious that Internet access was a telecommunications service. Brand X is the shadow of the entire net neutrality debate, and it seems the DC Circuit feels bound by that precedent, but thinks it should be resumed.
Regardless of the legal history, it really seems obvious to most people that broadband internet access is a telecommunications service that should be neutral. In this case, Ajit Pai and the FCC argued that broadband is in fact an "information service" because access is interconnected with … DNS and cache services. There is DNS, such as in the domain name lookup servers that translate domain names into IP addresses, and caching services that host copies of data closer to your location to speed up access.
No email, no crazy AOL chat room. DNS and caching. And because this argument worked in 2005 Brand X the court in 2019 was obliged to say that the FCC could use the same argument again.
We believe that broadband Internet access classification is an "information service" "Based on the functionality of DNS and caching," is a reasonable policy choice for [Commission] to make …
This claim – that Internet service providers offering DNS and caching are enough to make the right broadband access an "information service" – forcing the right to long digests and metaphors right away. Here's a little about how DNS is like invisible signs on the Internet subway? I have no idea what this means.
While DNS is "invisible" in the sense that it is "under the bonnet", so to speak, it remains "essential to provide Internet access to the regular consumer." Using a certain "Configuration" tool or – Protocols, for example, can speed up or slow down Internet traffic so that a metro's use of different rail technologies can affect train speeds, but a lack of DNS would be somewhat different, even users as they navigate the web, like a complete absence of signage in a subway. Signage, unlike DNS, is of course quite clear, but their user-centric purpose is the same for all practical purposes.
Of course, you can always just use a different DNS service than your ISP gives you, and completely destroy this already mute metaphor, but the court doesn't really think about it.
And here the court says that even though encrypted Internet traffic (you know, like all HTTPS traffic) does not flow through ISP caching servers, it doesn't matter, because, um, the FCC says.
The Commission (without contradiction in the record) found that the cache "enables and enhances consumer access to and use of information online." In particular, "
So, although there is a broad industry push against HTTPS encryption – 73 percent of all Internet traffic is now encrypted – ISP's simple presence of caching servers means that broadband is an "information service." It doesn't make much sense, since an ISP that only delivers DNS servers (which you don't need to use) and cache (not relevant to any HTTPS connection) is obviously not enough to make your broadband connection the same as Prodigy in 1998.
The court raised this argument, saying that DNS and caching were considered information services in Brand X and it would not overturn that precedent. It does so only by using a metaphor of, um, weaving sweaters with golden thread:
The idea seems to be that ISPs now offer fewer "walled gardens" services of the kind consumers generally care about albeit they did in the time of 2002 cable modem order and brand X, so to base an "information service" designation on DNS and cache alone at the moment is as questionable as saying that a few golden threads are woven into a regular sweater makes the sweater a golden garment … But the Supreme Court has never introduced or even hinted at such a quantitative standard for deciding whether inextricably intertwined functionality can justify an "information service" classification.
This is not hard to understand: Providing DNS and cache turns your Comcast connection into an "information service" that dialed AOL in 1998? No reasonable person would think so, but that's Ajit Pai's argument, and he's got the Brand X decision in his pocket, so he won. Again, this is a legal victory, not a logical one.
Landlines and Washing Machines
The Court further argues whether mobile broadband is a "commercial mobile service", which is the wireless version of a telecommunications service, or a "private mobile service", which is analogous to an information service. I will spare you the details of the long, lengthy discussion that follows, except to say that telecom legislation in 2019 is such that the court winds up making its decision based on the fact that smart washing machines cannot call.  You think I was joking.
The proliferation of "smart" devices with IP addresses, such as "servers, thermostats, washing machines and scores of other devices on the Internet of Things," threatened such a definition with a new complication. If these devices were part of the public wired network, it could give the dubious view that mobile voice will no longer be a commercial mobile service because subscribers could not connect to "all" endpoints of the network, such as IP-enabled TVs, washing machines and thermostats and other smart devices "that are unable to communicate. The whole mobile broadband argument is about whether mobile devices can connect to phone numbers!
If you thought it was foolish, just try to find out what worldwide this part of words about VoIP means:
The spread of VoIP and the prevalence of usage is perpendicular to the commission's point about the relationship between mobile broadband and VoIP, whether many users or few are used and whether they are pre-installed or acquired on an ad hoc basis, the question of whether VoIP functionality is part of the service in question – mobile broadband service – or Other services such as mobile broadband give users access.
I just want to reiterate that the basic question here is "should mobile broadband providers be able to block and choke internet traffic" and somehow the court has become sidetracked in nonsensical arguments about the availability of VoIP services such as Connecting to the phone system determines the answer.
Oh, but we're not done – let's go ahead and enter into a linguistic debate on the definitions of "applications" and "services."
Neither party identifies (and we have not found) either a set of regulatory definitions purported to draw lines between "applications" and "services," or a set of widely accepted linguistic practices that draw such a line or generally govern when the ability of apps that are usable with a service should be included in the "capabilities" of the service.
I would like to see AT&T send this language to a customer annoyed that CNN is streaming for free on their data plan, but Fox News does not.
The court then tries to make this upside down clearer by making a false call that absolutely no people would ever have in reality.
If someone tells a friend, "I just got a great new tablet with mobile broadband," it would hardly be a solecism for the friend to answer, "Great – does your service reach you from the landline?" the tablet owner, "Not now – but it can if I set up a Google Voice number," but it just shows the linguistic ambiguity.
I've read this rejected the spec script for a Verizon commercial over and over and it just gets more fun and sad at once, every time.
Monopoly: They're Awesome
Going forward, the Court addresses the fact that the US broadband market is suffering from a lack of competition and concludes that it actually is enough competition even for people who have only one choice of broadband provider. Seriously!
However, we are satisfied with the other reasons for the Commission to believe that competition exists in the broadband market. in empirical research that supports the claim that the presence of two lead suppliers is enough to ensure that there is meaningful competition. Consumers in areas with fewer than two providers can also reap the benefits of competition; A provider in this area "will tend to treat customers who don't have a competitive choice as if they do" because competitive pressure elsewhere "often has congestion effects on a given company."
This is just astonishing, obviously not true – if it was, everyone would love the prices and services they receive from the competing ISP industry in America. Instead, Americans pay more for lower speeds than most other countries. I would like to see this judge go on to tell millions of people begging for better broadband in the countryside that they are indeed "reaping the benefits of competition."
But don't worry about that lack of competition anyway, the court says: If Internet service providers do bad things, they will shape up because of the damage to their reputation .
In addition, these suppliers can meet high operating and reputation costs by behaving poorly in competitive areas. Based on these sensible findings and our highly deferential standard of review, it was not arbitrary for the Commission to conclude that fixed broadband providers have competitive pressure.
I just want to be clear that the court says AT&T and Verizon and Comcast and Spectrum are caring, smart companies who will do the right thing because they are so concerned about their reputation. It's a great idea, but here in the real world, everyone hates their ISPs more than ever. Verizon severed the connections of firefighters during a fire, which may be the most obvious way to damage your reputation. And AT&T wastes busyness with HBO through layoffs and talent withdrawals, even as activist investors point out that executives are bad at their jobs.
These companies are pets with virtually no competition, and they routinely act that way because we are stuck with them.
I will not get anyone to talk through the very technical legal arguments about whether the FCC can override state laws, except to say that the court was not very impressed with the FCC's argument that it has the implied authority to do so, and was pretty dumb about it.
If Congress wanted Title I to westernize the commission with some kind of dormant -Commerce clausal power to negate the statutory (and sovereign) powers of states just by washing their hands of their own regulatory authority, Congress could have said so.
All public opinion is broken by begging either the states or Congress to write a law, which frankly is what is going to happen. (And, in fact, the House has passed the popular Save the Internet Act, but the Senate won't take it up, while states across the country have adopted their own rules of online neutrality.)
Broadband Internet regulation has been the subject of protracted litigation , with broadband providers exposed and then released from regular carrier regulation over the past decade. We refuse to turn the switch of regulation of joint controllers once again under these circumstances.
So that's the most important, unsigned meaning. But I just want to point out that the two signed extras are equally bonkers. For example, Judge Stephen Williams wrote an opinion that partially agrees and disapproves in part, but the only thing you really need to know about it is that it starts with a quote from Macbeth :
And be these juggling the enemies believed no more,
That palter with us in a double sense;
Who holds the promise word to our ear,
And breaks it to our hope.
Then Macbeth says and finds that the witch's assurances were pure object and that his life collapses around him. Enactors of the Order of 2018, though certainly not Macbeths, can still feel a certain kinship, and were told that they acted lawfully to reject the heavy-handed title of Title II for the Internet, but that each of the 50 states stands free to impose just that.
And Judge Patricia Millet wrote an opinion listed as contemporaneous, since she agrees that the court is bound by the Brand X decision, but all other parts of it are a staggering reprimand of it terrible majority opinion. But it is also written as dramatically as possible.
The Commission's decision to stick to DNS and cache as an acid test for its regulatory classification "cannot bear very much reality." Today, the typical broadband offer is very similar to its Brand X version. The enclosed garden has been demolished and its fields sown with salt.
This quote in the first sentence is footnote to T.S. Eliot, don't worry, this whole thing remains as extra as ever. But I think Judge Millet has one thing right: Antonin Scalia had this nailed in 2005, and we have paid for the bad brand from the brand ever since.
Not only is the walled garden in ruin, but the roles of DNS and caching have changed dramatically since Brand X was determined. And they have done so in ways that strongly favor the classification of broadband as a telecommunications service, as Justice Scalia had originally advocated.
The more I read this decision, the more it became clear that legal machinations and discarded decisions based on a single bad precedent are getting further and further away from the realities of internet access for ordinary people. This decision may be appealed, and Brand X may overturn, but it is well past time for net neutrality to leave the world of endless court challenges and legal trifles, and just become the law.
And at least it seems that this ridiculous decision opens the door for states like California to do just that.