Lessons learned by IT in 2009 #1: 'Net neutrality' is a myth

Gaining the most leverage

The campaign-ability of net neutrality was most effectively realized in 2008 by a ranking member of then-candidate Barack Obama's transition team named Julius Genachowski, now the chairman of the FCC. In the FCC's notice of proposed rulemaking for net neutrality last October, Chairman Genachowski essentially absorbed Google's definition of the Internet as something designed from the beginning to enable free and open communication over the Web. Credited with the invention of the whole thing were: the US government for TCP/IP, Tim Berners-Lee for HTTP, and Marc Andreessen for Mosaic.

"Another more technical aspect of Internet openness has had the effect of empowering entrepreneurs to innovate without needing to seek permission. TCP/IP reflects a so-called 'end-to-end' system design, in which the routers in the middle of the network are not optimized toward the handling of any particular application, while network endpoints (the user's computer or other communicating device) are expected to perform the functions necessary to support specific networked applications," reads the NPRM (PDF available here). "The practical implication of this design philosophy has been that a software developer can develop new networked applications by writing programs only for end-user computers, without needing to modify the far more specialized programs running on network routing equipment."

History shows that the Internet wasn't so much a product of design as one of happenstance: The net happened. It is the very fact that applications have been written for the Internet without direct concern for how the network can sustain their traffic, that has fueled the entire net neutrality debate this past year. While the issue was born from the debate over national licensing, it transformed into a kind of privacy issue: If Comcast could throttle down your Internet pipeline after booting up a BitTorrent client, something, somewhere, must know that you're using BitTorrent. Which means you must be downloading something really big...and should Comcast know about that?

What's forgotten during that whole debate is the probability that Comcast may prefer not to know about that. If it did have "awareness," if you will, of your online activity, it could become legally responsible for it, and thus a target of lawsuits from rights holders who believe BitTorrent is the harbinger of IP theft.

Casting a skeptical eye on the whole affair from its outset has been FCC Commissioner Robert McDowell. In his partial dissent from the proposed rulemaking last October, Comm. McDowell cited the advanced engineering of Cisco's routers as evidence that the Internet is not, and never was, a dumb pipeline.

"Is the Commission suggesting today that the government draw a bright line of distinction between networks and applications in an effort to justify regulation in this space?" McDowell asked. "If so, should not the Commission refine its view because networks and applications are converging faster than regulators can measure? Otherwise, would the Commission not be favoring one market player over another absent evidence of an abuse of market power? For example, Cisco builds Internet routers that contain over 28.1 million lines of code. How are we to ascertain whether each line of code offers a pure operating system function or some other application that adds value? Should that be the Commission's role? Can we make such determinations efficiently? Do we even have the statutory authority to do any of this?"

McDowell's partial dissent is one of the few statements of reasoned opposition to currently proposed legislation that turns the entire issue of net neutrality -- appropriately -- on its head. Specifically, if the law were to decide that all applications were created equal, wouldn't the prohibition of any type of discrimination by routers and engineering equipment result in a kind of Darwinian virtual battleground, where applications of lesser bandwidth (those that are less equal than others) get pushed out?

Or to put it more directly: Would net neutrality legislation and regulation, as it is currently presented, bring about the very situation it was intended to prevent?

"During the course of this debate, many have confused the important difference between discriminatory conduct and anticompetitive conduct," McDowell goes on. "But the reality is that the Internet can function only if engineers are allowed to discriminate among different types of traffic. The word 'discriminate' carries with it negative connotations, but to network engineers it means 'network management.' Discriminatory conduct, in the network management context, does not necessarily mean anticompetitive conduct. For example, to enjoy online video downloads without interruption or distortion, consumers expect video bits to be given priority over other bits, such as e-mail bits. Such conduct is discriminatory, but not necessarily anticompetitive. If discriminatory conduct were to become anticompetitive conduct, then could it not be addressed in the context of competition and antitrust laws?"

How not to close a sale

In the era before net neutrality regulation was first considered, there were no truly bandwidth-heavy applications -- even HTTP can be rather lightweight. And there were no hot-button applications provoking knee-jerk responses from their practitioners -- FTP, for example, wasn't considered a kind of virtual red-light district the way many paint BitTorrent today.

But as many legitimate BitTorrent users will point out, there's much more to the Internet than the Web. In order for ISPs to continue making investments in building out broadband to do things way beyond the Web, from their vantage point, they would prefer some incentives, a little payback.

Voice-over-IP is a main area of contention. It's a bandwidth-intensive Internet application that's viewed as inherently legitimate, unlike BitTorrent which is often given a negative connotation. ISPs such as Comcast could conceivably afford to build out broadband if they could compete more effectively against Verizon and AT&T for telephone service. For that to happen, VoIP needs to not only become more cost-effective, but more reliably and explicitly managed -- a methodology which, when cast in a more populist light, looks like preferential treatment.

McDowell's scrutiny brings with it the danger that the debate over net neutrality will thoroughly decompose into a reasoned, thoughtful, intellectual discussion over network management practices and engineering principles -- in short, something that won't get hits on Google News. For that reason, most other net neutrality skeptics and opponents have instead resorted to clamoring for the oft-fumbled football of populism: a way to compel citizens to stand on one side or the other, on issues they don't necessarily have to understand.

So net neutrality opponents are attempting to leverage the First Amendment as being threatened by Genachowski's proposed rulemaking. Earlier this month, Kyle McSlarrow, who leads the NCTA -- one of the cable industry's leading advocacy groups -- suggested that by disallowing any kind of special management of VoIP traffic, the US government may not only be threatening the rights of NCTA member companies such as Comcast (assuming the right to offer services falls under "free speech"), but the free speech rights of those who would prefer to communicate with one another using VoIP rather than landline or wireless.
"By its plain terms and history, the First Amendment is a limitation on government power, not an empowerment of government," McSlarrow said.

Then last week, in response to criticism that net neutrality is not a civil rights issue in the way that...say, civil rights was a civil rights issue in the 1960s, the SaveTheInternet.com Coalition suggested that telecommunications companies such as Verizon and AT&T were making investments -- hedge bets, of a sort -- in "national civil rights groups and in communities of color," in hopes of encouraging "the civil rights community" to stand with them in opposition to net neutrality.

"In the wake of a conservative agenda to shrink government and decrease public investment in our communities, there is a long history of investment by telecommunications companies in national civil rights groups and in communities of color. So when companies like Verizon or AT&T frame an issue, they are trusted and believed...In a society marked by media systems that maintain and echo the interests of the dominant group," wrote contributor Malkia Cyril, "it's sound business and political strategy for telecommunications companies to invest in subordinated communities."

The smokescreens from both campaigns effectively mask an old and, to borrow a phrase, inconvenient truth: Any decision on the regulation of Internet traffic, including to disallow some or all regulation of it, constitutes control of that traffic. If it's decided that VoIP and BitTorrent and HTTP are "all created equal" -- to misappropriate those words used more eloquently to refer to people, not apps -- then conceivably the least efficient and most bandwidth-intensive of those apps will always take precedent over the simplest and least complex, perhaps even HTTP itself. And if it's decided instead that certain applications do deserve preferential treatment, then the boundlessness of today's playing field of popular debate indicates that the clamor over which application deserves that preference, will be a bloody free-for-all.

In either instance, applications and their creators and users will compete for bandwidth, space, time, and money. All we're dickering about now are the rules of the fight: Marcus of Queensbury, or cage-match. In no scenario is there an outcome of blissful, functional utopia. There is no net neutrality, due in large part to the fact that the concept wasn't created by engineers, but by politicians.