Highly Illogical Broadband Claims

The US jumped four places in Akamai’s most recent State of the Internet report, from 14th to 10th. Average connection speed increased from 17.2 Mbps to 18.7, an increase of 8.8%. The US increase was the second best in the world, behind little Korea’s 9.3% rise.  These measurements are taken from real life TCP streams, between end user machines and Akamai servers.

They don’t reflect actual broadband speeds advertised by ISPs because multiple streams are active at the same time. A more representative picture of overall broadband speed is the Akamai average peak connection speed (APCS), which hit 86.5 Mbps. This figure isn’t precisely representative of broadband speed because it’s helped by queueing upstream of ISP networks.

It’s always good to see broadband speeds going up. The cable company lobby, NCTA, celebrated the good figures with a blog post that drew a strange attack from net neutrality advocate Public Knowledge.

An Illogical Rate of Increase Claim

Average US Broadband Speed Improvements Over Two Year Spans. data source: Akamai State of the Internet reports

Public Knowledge claims that US broadband speed has increased at a faster rate since the 2015 Open Internet Order was imposed. The problems with this assertion are many:

Indeed, as the NCTA graph shows (based on the latest Akamai State of the Internet Report), the average speed of broadband connections has not only continued to rise since the FCC first adopted net neutrality rules in 2010, but the rate of increase has accelerated since the FCC adopted the Title II reclassification Order in February 2015.

In the first place, PK doesn’t quantify the rate of increase, they just say “look at the graph!” In fact, US broadband speeds have increased at a slower rate since the 2015 order was passed than they did in preceding two year intervals other than the drought that followed the 2010 Open Internet Order.

Over the two years since the order was passed, APCS has increased by 38%, the average of all annual averages of two-year increases. The overall average is depressed by dreadful figures in 2011, in the shadow of the 2010 Open Internet Order. But Open Internet Orders are not the root causes of broadband speed.

Regardless, in 2014, the year before the 2015 order was passed, the average two year increase was 45%, much better than what we’ve seen recently. The best years for two year rates of increase have been 2013 (50%) and 2014 (45%).  Coincidentally, these years spanned the court challenge to the 2010 Open Internet Order and the lawless period that followed the court’s rejection.

So Public Knowledge is peddling a fake fact.

An Illogical Explanation for the Fake Fact

Fake facts are useful when you need to support a fake argument. That argument, of course, would be the “Virtuous Cycle (or “Circle”, they call it by both names) Argument”. This abstraction proposes that broadband quality (or maybe investment, speed, or some other property) increases because of demand for (or maybe “by”, that’s unclear) applications and their users.

As networks become more magical, applications get even more wondrous. This pressures networks to become even more fantastic. But none of this can possibly happen unless networks are strongly regulated by the FCC, apparently.

The Virtuous Cycle/Circle is supposed to be an economic theory, but there’s no record of it in the economics literature. It has also been widely criticized by economists and others. At best, it’s a conjecture since there’s no empirical support for it; at worst, it’s simply a convenient post hoc rationalization.

What Really Explains Faster Broadband?

While it’s comforting for partisans to blame regulation for reduced rates of investment and slowing rates of improvement in network quality (or to credit regulation for increased investment and quality, as the case may be) there’s more to this game than mere regulation.

Fundamentally, network innovation, investment, and utility flow from technology and human needs. Regardless of the behavior of DC regulators, networks are going to keep improving.

People are going to keep on using applications too, and not simply because networks somehow force them to do so. In fact, the desire to use network applications comes from the appeal of applications in their own right. And this is made possible by better hardware and software in mobile devices.

Moore’s Law is the Logical Explanation

The real driver of the Internet ecosystem is chips and software. As processors in smartphones get better, programmers can create better applications. As processors and specialized networking chips get better, engineers can make networks faster, cheaper, and more reliable.

The virtuous cycle argument places the causes of innovation in the wrong place, in other words. The forces that make applications better also make networks better. And these forces are external to both networks and applications.

In effect, the FCC argues that man descends from monkeys, but the reality is that man and monkey share a common ancestor. The FCC cops a trick from diet fads that say eating carbohydrates makes us fat, when the reality is that eating too much makes us fat, regardless of what form the calories take.

When we understand that the completely unregulated markets for software and semiconductors drive improvements in applications and networks alike, we can better grasp the importance of  Augmented Reality.

We’re on the Cusp of Marvelous Developments

Apple’s Worldwide Developers Conference showcased developments in AR such as the ARKit, a set of developer tools for building better augmented reality applications. Pokémon Go showed us the way, but it’s stale now. That doesn’t mean AR is dead, it means we’re on the verge of the next step.

The new normal in computing is voice input instead of keyboard input. Instead of looking down at screens, displays will sit on our noses all the time. And instead of blasting sound all over the office, we’ll hear it from ear buds.

Instead wearing heart rate monitor straps, our watches will continually monitor heart rate and blood sugar. Our emotions will become inputs and outputs to the computing experience. Instead of sitting in our desks, computers will travel with us wherever we go.

These Developments are not Regulation Driven

The new paradigm for computing is happening, not just in the US but all over the world. It’s being made by entrepreneurs and innovators willing to invest and take risks, not by bureaucrats who get paid to say “no”.

The FCC can’t stop it, and by and large it can’t even alter the speed at which it comes about. That’s because innovation is global and the FCC is merely domestic.

What the FCC can do is help to keep large swathes of the American population from falling behind. And it can do this by saying yes to network deployment and innovation. A good first step in that process is to let go of the vacuous virtuous cycle of networks + apps innovation. That argument is illogical.

[Disclosure: I own Apple stock.]