WordsByNowak

A week ago I wrote about some figures put together by researchers using M-Labs, an online measurement tool started by Google, that showed Rogers to be the world’s worst throttler. In terms of percentage of internet connections slowed and sample test sizes, the Canadian cable provider proved to be the worst up until at least the first quarter of 2010. The researchers are working on releasing more up-to-date figures, but Rogers’ position is unlikely to improve much if at all, given the ongoing problems it has with throttling online video games.

That post attracted a considerable amount of attention. Not only was it the most read post I’ve had in the two-and-a-half years that I’ve been writing this blog, it was also picked up and made the top story on Huffington Post Canada over the weekend. It was also pointed to by numerous other blogs and podcasts.

My old colleagues at the CBC picked up the story as well and spoke with Milton Mueller, the Syracuse University professor who was the lead researcher behind the M-Labs findings. In no uncertain terms, Mueller - who appears to be quite tuned in to CRTC regulations - said Rogers is violating Canada’s net neutrality rules.

“Under the regulations that the CRTC promulgated for reasonable internet traffic management practices, I think 100 per cent, 24/7 throttling is not conformant,” he told the CBC. “So I think consumers would have a basis to complain and the CRTC would have a basis to act.”

A spokesperson for the company got back to me and countered, saying Rogers is in full compliance with CRTC regulations and pointed to a posted notification of its traffic management practices.

“While we don’t know exactly where this data comes from or the methodology used for this report, Rogers only manages upload traffic for P2P file sharing above 80 [kilobits per second],” the spokesperson said in an email. “From what we can tell, this report is measuring BitTorrent traffic only.”

I did my own tests using Glasnost, the same tool cited in Mueller’s findings, and it confirmed what Rogers has been saying all along. BitTorrent downloads were swimming along at between 8 and 11 megabits per second, but uploads were being slowed to around 80 kbps. (I also tried other kinds of traffic such as HTTP and Flash video, but got error messages saying my link couldn’t be tested - if anyone has any theories as to why, I’d love to hear them.)

Does this matter? It does for a few reasons. Firstly, I’m no technical wizard, but what I do know about BitTorrent is that if uploads are throttled, it’s just as good as throttling downloads too because peers often connect to the fastest uploaders first. This may not show up in Glasnost tests but can manifest itself as slow BitTorrent downloads in practice. Perhaps this is why Canadian ISPs are some of the stingiest around in providing decent upload speeds in the first place (Canada ranks 63rd in the world, according to Ookla, behind such internet heavyweights as Laos, Kenya and Belarus).

Secondly, as the CRTC has indirectly noted, throttling BitTorrent is one thing but extending it to other internet applications, such as World of Warcraft, is another. BitTorrent is unfortunately the black sheep of the internet because it is used by many to share movies and music, which is why regulators and other authorities tend to quietly tolerate ISPs beating up on it. However, BitTorrent does have legitimate uses and it is operated by a perfectly legitimate business, so crippling it is a violation of net neutrality principles since all perfectly reasonable uses of the internet should have the same rights and capabilities as others on the network.

Extending that same throttling to other perfectly legitimate uses such as gaming, whether it’s on purpose or accidental, only compounds the violation.

Thirdly and perhaps most importantly is the fact that Mueller pointed to. Around-the-clock throttling surely goes beyond the CRTC’s net neutrality rules, which require ISPs to use such measures as little as possible.

A few years back, I had a decent conversation with some Rogers engineers about throttling. I’ve extended an invitation to see if they’d like to revisit the topic and am waiting to hear back. As for the CRTC, how it proceeds on the issue may be influenced by its upcoming decision on usage-based billing, which is due any time now.