Pass the Popcorn! CRTC Offers Great Opportunity To Watch Someone Else Play With Critical Infrastructure.

According to this official news release, the Canadian Radio-Television and Telecommunications Commission (CRTC) denied a request from the Canadian Association of Internet Providers (CAIP) to stop Bell Canada from throttling without notice the traffic of rival ISPs leasing access to Bell Canada’s network. Instead, CRTC punted to a general inquiry on traffic shaping.

According to Michael Geist, expert on all things telecom and Candian and general super-smart guy, this is not the last word from the CRTC on the question. But since — according to the public notice — the first public hearing on the subject is scheduled for July 9, 2009, Canadian ISPs can look forward to a considerable period of time when they live at the mercy of their largest rival.

This does not depress me, as I do not live in Canada. Rather, I am excited at the prospect of some other country (for a change) deciding to make offerings to the Gods of the Marketplace and play games with its critical infrastructure while I get to watch. Until now, Canada has generally been outranking us in the international rankings on penetration, although it ranks less well on affodability and only so-so on speed (as compared to countries with real broadband). Those who see such things as relevant (and not everybody does, the situation is complex and the data messy, hard to come by, and subject to multiple interpretations) generally regard this as a consequence of bad policy choices by the FCC (again, not everyone agrees, the data — to the extent we even have data — is very messy and complex). In particular, a lot of us think that the decision to eliminate mandatory wholesale access and rely on “intermodal” competition was a phenomenally bad idea.

Now we may get a chance over the next few years to test this hypothesis, and at someone else’s expense! Go Canada!

More below . . . .

Broadband Reports does a pretty good job laying out the issues. Briefly, some years back, when the FCC eliminate requirements that telcos lease broadband facilities to competitors and generally deregulate broadband access on a theory that this would spur investment and encourage “intermodal” competition, the CRTC went the other way. It required Bell Canada to make DSL lines available at wholesale rates to rival competitors, like we did here in the U.S. until 2005.

Last year, Bell Canada began throttling user traffic (or, if one prefers the industry euphamism “shaping”) on its residential DSL service. It also started doing the same for its wholesale customers — but without telling them. Unsurprisingly, ISPs were annoyed to discover that their chief rival had taken such liberties and pissed off their customers for them. They complained to the Canadian Radio-Television Commission, Canada’s equivalent of the FCC. Bell Canada claimed that it was necessary to throttle everyone in order to prevent them from overwhelming the network. In other words, even though the competing ISPs were willing to pay for more bandwidth, they couldn’t get it. And, rather than let the competing ISPs offer different kinds of network management schemes, Bell Canada required them to adopt the same throttling scheme as Bell Canada — and under Bell Canada’s control. When pressed by CRTC to provide evidence of congestion, Bell Canada submitted data which suggested that it did not face an immediate crisis of bandwidth absent throttling. CRTC, however, felt differently.

Reading through the CRTC decision is rather enjoyable, given that I am not Canadian. It is rather like reading the Comcast/BitTorrent Order from some alternate universe where Robert McDowell was Chairman of the FCC. CRTC accepts Bell Canada’s data without question and finds their choice of network management reasonable because, after all, Bell Canada should run the network and government should not interfere in something as delightfully free market as the internet. CRTC also rejects the idea that this can possibly be anticompetitive if Bell Canada throttles the traffic of its rivals in the same manner it throttles its own. CRTC also dismisses in a fairly cavalier fashion the fact that Bell Canada implemented this network management scheme without notice, and that it may violate the privacy rights of users.

Needless to say, those who opposed the result we got here in the U.S. are delighted with the CRTC for “getting it right.” As for myself, I’m fairly delighted as well. While this hardly denotes a true test given the “observer effect” (i.e., Bell Canada knows it is being watched), I am looking forward to seeing how this plays out over the next few years (assuming CRTC moves as slowly as the FCC does on a general notice of inquiry). For one thing, we will get to observe whether there is an “exaflood” threatening us all or not. Those who disbelieve the exaflood note that only those with a financial interest in the exaflood claim it exists. Those who think network congestion is a real issue argue that traffic data is private and needs to be kept proprietary. I personally think that this amounts to “trust us, there is an exaflood” from people I am not inclined to trust. But that’s just me.

So how will Canada’s ISP industry evolve? Will Bell Canada take this as license to play further games, the way the Bells here in the U.S. did once they knew the FCC would not enforce the line sharing requirements (until they were eliminated)? Will the Canadian independent ISP industry gradually whither away while Canadians end up with reduced access to internet services and/or content? Or will we in the U.S. suffer under bandwidth caps that choke our access, or end up drowning our ISPs in an exaflood of data they are powerless to manage because of the awful intrusion of government bureaucrats into network management. Will the fact that Bell Canada’s video product is not throttled make a difference for the development of Internet delivered video as a potential competitor to Bell Canada? Or will the U.S. never evolve a video competitor to Comcast and Time Warner because the would-be video providers cannot buy sufficient quality of service for reliable delivery?

I expect there will be no clear answers for awhile. Reality is always messy, and I expect data to be conflicting for the first few years. As I have noted before in the broadband debate (and elsewhere) life and public policy are all about trade offs. The trick (IMO) is neither blind worship of the free market or adoration of government, but to pick the right tools for the job you want accomplished. Still, its rare that someone else decides to leap off the free market deep end in such a radical, pro-incumbent way ahead of us. I’m glad Canada elected to be the guinea pig this time.

Stay tuned . . . .


  1. The post of mine from CircleID and my blog where I discuss the discrepancy between data from public peering points and private ones does *not* argue that the data is proprietary and needs to stay that way.

    Rather, I argue that the speculations about Internet traffic volume drawn exclusively from public sources are misleading. We know that a great deal of data moves between CDNs and ISPs through private peering arrangements, and we know that Odlyzko doesn’t measure this traffic.

    Consequently, his claim that traffic volume is growing more slowly than in the past is 180 degrees from the truth. In all likelihood, traffic is growing so rapidly that it’s become necessary for CDNs to bypass public peering. Odlyzko is playing this passive-aggressive game where he makes no effort to collect any data on private peering and then makes wild conjectures on the data he does have. He knows better.

    The only way we can know how much traffic is moving across the Internet is to collect data at the end points. I would not be at all upset for a government mandate to come along requiring ISP reporting of this data to the FCC and subsequent release to the public in anonymized form. I’d like to know total traffic, distribution by user and protocol type, and loss and delay characteristics. Having such data would make everyone’s job a lot easier and lead to better policy.

    Intuitively, I believe that there is so much video on the Internet now that total traffic has to be growing at the highest rate we’ve seen since the mid-90s, but I can’t support that with data. The current state of affairs is so bad, I can’t refute it with data either, and that’s no fun.

    I think you’d agree with me that faith-based Internet policy is a loser: “give me the facts, ma’am, and just the facts;” I’ll take it from there.

  2. I absolutely agree. I find the lack of data extremely disturbing.

    Where we diverge is what to do in our lack of data. My feeling is that I need to act on the data I have, knowing it is incomplete and counting on the fact that if I push hard enough, those with the data will provide it.

    Not my preferred mode of operation. But, as the financial crisis dramatically demonstrates, refusing to act because it might make things worse carries consequences of its own.

  3. I certainly agree that sound policy shouldn’t be crisis-driven, but all too often it seems to take a meltdown to get anyone’s attention. Maybe it will get a little better now that the election is over.

  4. you two should co-chair the FCC, really, i mean it!

Comments are closed