BitTorrent CEO Eric Klinker offers what in essence is an op-ed at GigaOM in which he takes issue with the way carriers present the rationale for bandwidth caps, which put a limit on the amount of data a broadband subscriber can use.
The carriers, Klinker suggests, position bandwidth caps as ways to limit "bandwidth hogs." He suggests that this assessment is wrong. The crux of Klinker's argument is that networks-like highways -- are designed to handle peak loads. Capacity, traffic lanes or bandwidth, is added as those peak loads increase. Metering use and punishing those who use more than their share won't cut peak traffic simply because nobody (including the bandwidth hogs) would slow down their use at peak hours. Therefore, metering would do nothing to truly address the problem.
The vehicular traffic analogy makes it clear: A limit on driving-one above which an extra charge is incurred-wouldn't lead people to pull off the roads at rush hour because they presumably are going to work. If anything, they would cut out a pleasure drive at an off-peak hour to stay below the limit. It's the same for the Internet: Folks are more likely to cut their usage during off-hours. But these are the times at which the network isn't stressed anyway.
If the concern was congestion, one might expect more of a focus on an actual congestion throttling system like Comcast's, where heavy users find their traffic slowed a bit during periods of actual network congestion and then restored to full speed when congestion disappears. Monthly data caps don't actually discourage people from using the Internet during the busiest times, though they may lower some total monthly usage.
Nobody is arguing that caps would make much of a difference to all but the biggest users today. In the long run, however, it may be that carriers such as AT&T -- and, as Light Reading suggests-cable operators (who will use the cover provided by the telco move to follow suit) are gearing up to use metering as a way to maintain profitability as the video expansion continues.
That explosion is now led by Netflix, which the NPD Group recently said streams 61 percent of "digital movie units" during the first two months of the year. Other providers will catch up and a lot more bandwidth will be needed.
The bottom line is that the telecommunications industry has a lot of experience in network capacity planning. In the telco world, it is known as "contention." The idea is that networks are not designed to handle inordinately heavy loads that rarely occur, but are engineered to operate solidly during normal conditions. The concept is no different in the video world. The point is that a lot of very smart people have spent decades thinking about this topic.
Whether or not the current model is fair to carriers is a fair question. Whatever the answer, however, it is undeniable that this is a politically difficult question, and carriers may be tempted to create a bandwidth hog red herring. It would be troubling if Klinker and Anderson are correct in their suggestion that the carriers are trying to justify metering based on a reasonable-sounding, but misleading, rationale.