My students have been reading what academics like to call a “seminal” work. It’s the 2004 paper by Tim Wu that put network neutrality on the map: Network Neutrality, Broadband Discrimination (uploaded here).
Eight years on, Canadians have lots of broadband discrimination and very little network neutrality. Yet hope springs eternal. And with the arrival of Jean-Pierre Blais as the new CRTC chair, the pro-Internet community may have an opportunity to change some of the misguided thinking that has undermined the Commission’s policymaking for years.
Taking the common out of carriage
One of the problems inherent in defending net neutrality is defining it. Let’s say for the moment that, as a necessary if not sufficient condition, net neutrality requires that broadband providers may not unreasonably discriminate in transmitting lawful network traffic, nor block lawful and non-harmful content, applications, services or devices. This concept is hardly newfangled or unique to the Internet. It lies at the heart of what a common carrier is expected to do for the privilege of sellling its services in a regulated or quasi-regulated market – like carrying coal in barges on English canals or people in taxicabs (the fare meter is part of the system that ensures everyone is treated on an equitable basis).
What is unusual if not unique to the non-discrimination principle on the public Internet is the telecom firms that used to be common carriers – firms like Bell – are no longer prohibited from owning, controlling or manipulating the content that travels across their networks. Thus, Bell not only provides carriage facilities for your content, like personal messages; it also owns large content providers such as CTV, whose content Bell distributes over the same network platform as your tweets.
That’s where life starts to get complicated – especially for a regulator that is at war with consumers in almost everything it does. (For some background, see my post from July 2011, Beyond UBB: the CRTC’s war on Canadian consumers.)
How is the ITMP framework broken? Let us count the ways
In the following paragraphs, I outline a number of arguments as to why the CRTC’s ITMP framework has little to do with net neutrality, and in fact has made important goals associated with an open Internet even more elusive than they were in years past. The point of this exercise might be a little Quixotic: to give the new chair and his staff some second thoughts about what the Commission hath wrought – and then to consider what changes in its priorities and procedures might help rebuild its reputation as a guardian of the public interest and an open Internet.
ITMP failing #1 – No neutrality. Canadian onliners were alleged to have come out victorious after the CRTC’s October 2009 “Review of the Internet traffic management practices of Internet service providers” – or “ITMPs” (Telecom Regulatory Policy CRTC 2009-657). At least, that’s what we were told at the time. The accompanying press release quoted the then chair, Konrad von Finckenstein, as saying that the framework “will foster an environment where ISPs, application providers and users have the utmost freedom to innovate.”
The record indicates that both the Commission and the Tory leadership believed the ITMP policy enshrines the principles of an open Internet, thanks to which Canadians are free to lead their online lives without undue constraints or discrimination imposed by their ISP. Here’s how the policy document construes the paramount issue of net neutrality (para 4):
At the core of the debate over “net neutrality” is whether innovation will continue to come from the edges of networks, without permission. Will there continue to be rapid and uncontrolled innovation in computer communications? Will citizens have full access to that innovation? The Commission earnestly hopes so. However, due to the limited capacity of their networks, carriers have legitimate interests in the management of these networks. Will they be able to develop networks that can bear the traffic generated by this innovation? Will they, in turn, be empowered to innovate?
Brave words. But while the Commission “hoped” for lots of innovation without permission, its language, assumptions and putative remedies were steering it in the opposite direction. Moreover, the above-cited paragraph is the only place in the whole document that actually mentions net neutrality. By contrast, the word “traffic” – a concern of suppliers, not buyers – appears 73 times, while a consumer-oriented expression like “affordable” is not mentioned even once.
The term “neutrality” does appear in one other spot: in the preamble, where the Commission lays out its guiding principles. But this reference concerns what the Commission calls “competitive neutrality,” which is about preventing the incumbents from treating their wholesale customers unfairly:
For retail services, ISPs may continue to employ ITMPs without prior Commission approval. The Commission will review such practices, assessing them against the framework, based upon concerns arising primarily through complaints by consumers.
For wholesale services there will be additional scrutiny. When an ISP employs more restrictive ITMPs for its wholesale services than for its retail services, it will require Commission approval to implement those practices. technical ITMPs applied to wholesale services must comply with the ITMP framework and must not have a significant and disproportionate impact on secondary ISP traffic.
And why is the regulator watching wholesale more closely than retail? Because it believes the broadband market in Canada is competitive enough to allow market forces to discipline retail pricing. So how are we doing on that score? Not so good, as we’ll see later in connection with the “adjustments” to Bell’s Internet pricing this past January.
A more explicit indictment of the original framework lies in the repair job the Commission had to do two years after the original policy was issued (Oct 2009). In September 2011, it issued a telecom information bulletin designed to achieve goals it should have taken care of from the get-go. The 2011 bulletin (CRTC 2011-609) is entitled “Internet traffic management practices – Guidelines for responding to complaints and enforcing framework compliance by Internet service providers.”
This makeover did nothing to get at the core public interest issues relating to how we can preserve an open Internet. Instead, it was an admission that the policy had no teeth; left complainants high and dry; and exposed the Commission’s absurd attempts to keep subscriber complaints a state secret – away from the prying eyes of the public this regulator is supposed to be protecting. Which brings us to another important body part…
ITMP failing #2 – No teeth. As events over the last couple of years have shown, the framework wasn’t – and isn’t – enforceable in any meaningful sense. How do we know? Because the incumbents ignored its provisions with impunity and are apparently still doing so.
Evidence of the Commission’s inept handling of complaints popped up as recently as last week. On June 28, after more than a year of arguments and filings regarding Rogers’ throttling of World of Warcraft, the Commission simply closed the case. It is apparently unconcerned that Rogers never owned responsibility for the problems it caused in applying DPI so recklessly on its network. And if Rogers fails to phase out throttling on its network by its self-imposed deadline - the end of 2012 – it’s difficult to imagine the Commission would drag Rogers into an enforcement proceeding. (I wrote a series of five posts last fall on this battle royal, beginning on September 21 with Game throttled? Complain about Rogers, but blame the CRTC.)
One party who is not amused by the Commission’s behavior is Jason Koblovsky. Jason, you will recall, was one of the lead complainants through this whole ordeal, along with his colleague Teresa Murphy at the Canadian Gamers Organization (CGO). Jason has written a thoughtful blog post about the latest turn of events (on OpenMedia). Despite their fighting this battle for so long, Jason points out the Commission didn’t even have the decency to notify the CGO directly about closing the file. They learned about it, like everyone else, from the “highly self congratulatory” June 28 notice. As Jason says:
“We had quite the fight to even get the CRTC to look into this, even after policy changes were made on ITMP complaints last year. To our knowledge our complaint is the only one that made it through to the enforcement division at the CRTC. We think the changes the CRTC has made are only surface changes and that ITMP policy still heavily weighs on the side of the Internet Providers rather than being balanced to ensure value for money.”
(I’m pleased to report Teresa has agreed to come to class this week and talk in person with my students about the WoW/Rogers affair – a case study that will provide some real-world context for our look at the alternate universe that is regulatory affairs.)
ITMP failing #3 – No transparency. The original 2009 policy provides that ISPs shall apply solutions to managing their traffic in a certain rank order. The first of these is to spend money: “Network investment is a fundamental tool for dealing with network congestion and should continue to be the primary solution that ISPs use; however, investment alone does not obviate the need for certain ITMPs.”
Whatever the potential benefits, the Commission has little more than moral suasion to encourage the incumbents to spend their shareholders’ money. It therefore turns to two other general options: technical ITMPs (like throttling) and economic ITMPs (like data caps). Here’s where the Commission’s choices show it to be depressingly naive about broadband adoption and the capabilities of mainstream end-users. The key choice here is that ISPs must put economic measures ahead of technical ones. The reason?…
“Economic practices are the most transparent ITMPs. They match consumer usage with willingness to pay, thus putting users in control and allowing market forces to work.”
As I’ve noted so many times, most of the students arriving in my 4th-year seminar have no idea what a gigabyte is – or bandwidth, or megabits per second, or how many megs in a gig, etc (that’s one reason they take the course). With that in mind, let’s check out the way Rogers explains in its Internet FAQ how its subs can go about calculating how much extra money they owe Rogers for using the Internet too much, i.e. exceeding their data cap, another concept no one outside the business has the slightest understanding of (this item is taken verbatim from the Rogers site):
How can I calculate my additional usage and charges?
The usage tool tracks your usage in megabytes. To convert your usage into gigabytes and calculate your additional charges, follow these steps:
Total Usage: 128,522 MB
Service Includes: 90,000 MB
1. Over usage in GB = (128,522 – 90,000) / 1,024 = 37.62 GB
2. Round down to the nearest gigabyte = 37 GB
3. Over-usage charge = 37 X $1.25 = $46.25
Quick: why does this FAQ use binary numbering (2^10) instead of the standard decimal numbering (10^3) for converting megabytes to gigabytes?… Join the club, neither do I.
Let’s also keep in mind that the Tory government’s 2006 direction to the CRTC on using market forces as a stand-in for regulation puts the incumbents in control, not users. Furthermore, willingness to pay, or to do anything as a consumer, presupposes a certain level of awareness and understanding – much like the idea of informed consent. I’ve been over this issue a million times with students and lots of other people. Bottom line: none of this connectivity stuff is “transparent” to ordinary end-users. If the CRTC ever did empirical research about consumer attitudes and behaviors (as opposed to its online “consultations”), it might have seen how wide of the mark this wishful thinking about consumers really is. Don’t take my word for it. In 2011, the FCC conducted a random-probability national survey of broadband homes and found that 80% of respondents had no idea whatsoever what access tier they were on, or how much bandwidth they were paying for in their own homes.
Thus, the very cornerstone of the ITMP framework – the belief that economic ITMPs actually favor consumer welfare – turns out to be based on assumptions that are not only untrue, but have been exploited by the incumbents to aid and abet price-gouging, all thanks to the Commission’s enthusiastic support for the deployment of data caps as a disincentive to using the unprecedented resources offered by the Internet.
We’ll take up that and a number of related issues in Part 2…