Former FCC Chairman Tom Wheeler has broken his silence on the FCC’s desire to re-impose his Title II regulatory framework on Internet Service Providers. This is dog-bites-man news apart from the fact that his arguments refute themselves.

In keeping with the tradition he laid down in his 2015 Protecting and Promoting the Open Internet Order, he relies on the argument that the FCC needs the authority to ensure that ISP conduct is “just and reasonable” at all times. Never mind that this Communications Act language is generally applied to rates rather than general conduct.

This echoes Wheeler’s 2015 order that mentioned just and reasonable rates 13 times despite promising us that price caps are not going to happen. Certainly capping rates and fees for pole attachments and network interconnection has some impact on user fees.

Dubious history

Always the historian, Wheeler cites examples from the deep past to justify Title II, starting with the 1860 Pacific Telegraph Act. That piece of history doesn’t really make his case, as it contains a passage that grants free use of the network to certain government agencies in the first use of Zero Rating I’ve seen: Provided, That the use of the line be given, at any time, free of cost, to the Coast Survey, the Smithsonian Institution, and the National Observatory, for scientific purposes.

For some reason, free access to mobile networks for telemedicine visits with VA hospitals doesn’t get this treatment by the 2015 Open Internet rules. Wheeler’s cite of the PTA also fails to acknowledge the fact that the law prioritized some traffic. He says:

Not only did the law fund the transcontinental telegraph, but it also mandated that the messages on that network be “impartially transmitted in order of their reception.”

…but that fragment is followed by an exception.

The rest of the sentence

The exception is instructive: …excepting that the dispatches of the government shall have priority. Given that the government reserves priority to itself as the funder of the network, this looks an awful lot like paid prioritization, doesn’t it?

Wheeler didn’t read the Telegraph Act very carefully, but even if he had the relevance of an 1860 law about a single-purpose network to today’s multi-purpose Internet frankly escapes me. This is scraping the bottom of the barrel.

Fiery inferno

The chairman also makes the dishonest move of demonizing Verizon for throttling the Santa Clara County fire department’s smart phone account when it exceeded its data limit.

The problem was quickly “corrected” without Title II by a free bump-up to a higher service level than the one the district actually purchased. (The firefighters should have been using FirstNet anyhow.)

Title II doesn’t ban service levels, nor does it bar government agencies from making ordering mistakes.

The infamous BitTorrent affair

Wheeler concludes his history tour with a blatant misrepresentation of the 2008 Comcast/BitTorrent affair in which the ISP ham-fistedly dealt with complaints that it was throttling Vonage by throttling a tool used by media pirates instead. The correct way to resolve the problem was built-in to an equipment upgrade Comcast had ordered.

The equipment was unfortunately delayed for six months by last minute changes to the DOCSIS 3 standard. This was explained to FCC staff at the time, but Wheeler presumes Comcast was protecting its cable TV business:

In 2008, the FCC found Comcast was throttling the video streaming service BitTorrent, presumably because it was competitive with their core cable TV business.

But hey, we can’t really expect a former FCC chairman to have any knowledge of the firms he not only used to regulate but for whom he worked for three decades or so. Much later, ISPs realized that moments of congestion in their networks had more to do with buffer bloat than anything else. Buffer bloat is a wide-spread Internet problem independent of the video business.

Just and reasonable sounds great!

I mention these three misrepresentations to highlight the issue with a vague and airy just and reasonable standard for ISP conduct. If we want ISPs to behave in the fashion can’t we expect the same from their regulators?

In the real world there’s a huge gap between abstract principles and concrete practices. In part, this is bridged by subjective judgements of impartial and well-informed regulators. As Wheeler shows by his mistreatment of fact and law in the three instances cited, such regulators can be hard to find.

Hence, the Administrative Procedure Act constrains the ability of regulators to act whimsically, arbitrarily, irresponsibly, maliciously, or in excess of their authority. When FCC Chairman Kevin Martin dropped the hammer on Comcast over BitTorrent, he was reversed by the court because his FCC had neglected to formulate rules prior to their enforcement, an act of pure whimsy.

Fair Play

We don’t just want regulated firms to be just and reasonable, we want them to be treated fairly by their government overlords. If nothing else, fair and impartial treatment by the FCC sets a good example for affected firms.

A steady hand at the tiller also calms markets, enabling capital-intensive industries like broadband to make long-term plans. On that note, Wheeler’s investment analysis is a real doozy.

He attempts to show that Title II didn’t inhibit investment in broadband by citing an advocacy claim that capital spending by broadband providers didn’t immediately increase in 2018, the year in which the Restoring Internet Freedom Order repealed his order. This claim is about as serious as calling hallucinating ChatGPT “the oracle”, as Wheeler actually does.

Beyond hallucination

The history of net neutrality and its FCC enforcement doesn’t inspire confidence in regulators. Chairmen Martin and Genachowski acted unlawfully, Wheeler arbitrarily, and Pai hurriedly. The status quo, allowing California to make up its own rules, is unacceptable.

A return to heavy-handed, arbitrary regulation with no grounding in engineering or economics will be even worse. FCC enforcement decisions will end up in the courts where the entire enterprise is likely to be vacated by the Supreme Court on Major Questions grounds.

Congress needs to step up by passing a law declaring that Internet Service is interstate commerce, wherever and however it is provided. This should be straightforward even if detailed Open Internet rules are not.

The FCC should then do what it can to support competition among facilities-based ISPs, including making mid-band spectrum licenses available for Fixed Wireless Access. This will shift the calculus away from politically-motivated regulators to markets, which have proved capable of disciplining producers effectively and consistently when properly functioning.

A little history

For what it’s worth, net neutrality didn’t begin with BitTorrent, as Wheeler claims, it actually began in a 1997 blog post, The Rise of the Stupid Network, written by disgruntled AT&T employee David Isenberg, a man with no professional experience with Internet engineering. It was angry but not terrible.

But it soon morphed into the very bad Dawn of the Stupid Network before being reduced to sheer fantasy in Larry Lessig’s Code, to wit: Like a daydreaming postal worker, the network simply moves the data and leaves its interpretation to the applications…

Lessig passed the idea on to his student Tim Wu, the man who gave net neutrality its name. Wu subsequently vacated the net neutrality scene, passing the mantle to Barbara van Schewick, the current director of Lessig’s Google-funded Stanford Center on Internet and Society.

Original net neutrality musings were more like attempts by non-technical people to understand how the Internet was different from all previous networks than prescriptions for regulation. The idea that this vision of a revolution in networking would later justify telephone network regulations is the very height of irony. For example, End-to-End Arguments in System Design author David Reed rejects Title II while continuing to support an open Internet.

A path forward

The current Open Internet proceeding gets one thing right: The Internet, having gobbled up all of its predecessors, has become the single most important network in the world. Contrary to the desires of regulators to use this fact to elevate their stature, this means it is scarcely in need of radically old-school regulatory treatment.

The Internet has succeeded. Everybody with a shiny new communications technology wants to use it to fill in gaps in Internet coverage or to improve the Internet’s power and efficiency. This goes for AI as well as broadband.

One of the first large-scale uses of AI was predictive maintenance for computers and networks, way back in the 1990s. Consequently, network engineering professionals are not impressed by talking heads waving the ChatGPT wand over their life’s work as Wheeler does.

The dead end

Engineers like to joke about putting themselves out of business by creating great new things that don’t require specialist improvements. We should cultivate that ethos in regulators.

What would the FCC look like if it saw its job as creating markets that discipline themselves rather than running around like day-dreaming NBA referees whistling fouls? I suspect it would be more like Pai’s FCC than Wheeler’s. That would be good for the Internet and great for all of us.

Unfortunately, FCC chair Jessica Rosenworcel is now signaling that she won’t vacate state net neutrality laws. Pre-empting such laws would have been the only tangible benefit of this year’s proceeding. I have to wonder if the FCC majority takes its own rhetoric seriously.