On August 1, the FCC issued a ruling that Comcast had endangered the “vibrant and open nature of the Internet” by slowing down customer use of BitTorrent and other peer-to-peer applications, and that it had misrepresented its practices to both its customers and the Commission.
The finding would appear decisive, but the Commission’s ruling has something for everybody. “Neutrality” proponents will claim that it justifies their vision of an “unimpeded” end-to-end Internet. Providers will note that the finding doesn’t rule out traffic management during times of congestions, just the clumsy way Comcast went about it. Consumer advocates will take pleasure in the Commission’s championship of transparency. Others will take heart in the Commission’s statement that blocking odious content such as child pornography or pirated intellectual property is still “consistent with Federal policy.” The ruling enunciates clear principles, even if peppered with interesting contradictions, and makes clear that the FCC can do whatever it needs to do without new legislation. So the Commission’s ruling does a great deal for a great many, but it still doesn’t do one thing.
The ruling doesn’t solve the fundamental problem confronting the Internet – what to do when it gets too crowded.
Let’s first take a moment to look at the background of the FCC’s ruling. Several independent groups, including the Associated Press, discovered that Comcast was slowing traffic that used peer-to-peer applications, principally BitTorrent, but others as well. At first Comcast denied it, but then admitted it was true, explaining that it did so only to manage periods of peak congestion. It then changed its story again, admitting that it regularly slows this kind of traffic regardless of the level of congestion.
As ever, it’s the cover-up, not the crime. I don’t understand why Comcast was slowing traffic when there weren’t congested conditions, although I can imagine doing so prevents congestion before the fact. And slowing down one type of application – peer-to-peer – sounds discriminatory, but you can see the argument for it. It’s like saying that you can’t water your lawn when there’s a drought and the reservoir is low – it may be arbitrary (some folks would rather give up bathing than stop watering the lawn), but at least it solves the problem.
Comcast’s putting the burden of limiting congestion by going after peer-to-peer networks has that quality. One reason they drew such fire for doing so is that when you say “peer to peer networks,” people think – piracy. But blaming peer-to-per networks for piracy is like blaming the ocean for pirates – being on the ocean doesn’t make you a pirate, but that’s where all the pirates work. Piracy isn’t Comcast’s concern, nor should it be – that would be like asking rhe airlines to search their passengers for contraband. File swapping increases the demand for bandwidth, and bandwidth is what Comcast sells.
What Comcast doesn’t sell much of over high-speed Internet connections is content, which leads to one of the oddest aspects of the Commission’s finding. It says that Comcast’s motive for reining it BitTorrent-type applications is that those applications “provide Internet users with the opportunity to view high-quality video that they might otherwise watch (and pay for) on cable television.” Yes, Comcast does sell cable-TV service, but the order doesn’t contain proof that Comcast was intervening due to BitTorrent’s competition with Comcast TV services. And would you give up cable TV for what you can get through the Internet? Or more accurately – have you?
No, peer-to-per networks and BitTorrent applications are the likely target for limitation during congested periods because they’re big, incredibly big. P2P does create congestion problems. The bandwidth used by YouTube, a multibillion dollar subsidiary of Google, itself the eleventh largest corporation in America, is greater than the bandwidth of the entire Internet eight years ago. If you want to cut back on bandwidth use, there’s an easy target. And, at first blush, it’s not a bad presumption that reining back peer-to-peer has lower social costs than, say, e-commerce, or school or government or health networks.
Is that unfair, or arbitrary? Well, that takes us back to the question I posed earlier -- what do we do when the Internet is crowded? And that’s what I’ll address later today.
#2
So what do we do when the Internet is crowded? Anything that has a fixed level of capacity is vulnerable to too much demand. When there are too many cars on the road, there’s traffic and people sit and idle their engines. So Comcast’s solution was arbitrary (and inexpertly administered), but ultimately no more so than telling people they can’t water their lawn during a drought.
Was Comcast’s move the best solution? No, not by a long shot. The best solution is to let users and websites determine what speed and reliability are worth to them. Different websites will value speed differently, and for different reasons. A health clinic with real-time monitors or an on-line stock trader will think every increment of speed and reliability is worth it. A new competitor to Google might value speed highly to compete with the incumbent by being faster. So some sites are simply going to be willing to buy the right to priority for their site’s traffic when the Internet gets crowded. It would be like having HOV lanes that let some traffic pass more quickly.
The people who propose “net neutrality” oppose this vision of a competitive Internet. They argue that the Internet would lose its “open and vibrant” nature if this were allowed. But it’s hard to see how. For example, some people send their mail by priority overnight, some send it by second day delivery, and some with a regular stamp, but all the mail gets through. And some businesses have 24/7 live customer service lines, some only during office hours, and some respond to e-mail the next day – they decide what that level of service is worth and act accordingly … and so, too, do customers.
Why should the Internet be any different? Neutrality proponents point to the Internet’s epochal character and its potential to bring together all of humanity in a new and profound way. No one argues that. But the singularity of the Internet doesn’t tell us what to do to use it in a way that does the most good for the greatest number. And when it gets congested, the best answer probably isn’t the one “neutrality” proponents favor – everybody being slow together -- hey, at least it’s “neutral.” Besides, that’s not what happens now anyway. The big sites – Google, EBay, and so on – already have big speed advantages due to more and better equipment (including caching) that smaller guys can’t get. That’s why they’re the money behind “neutrality” – because “neutrality” preserves their advantages.
The FCC ruling has to be seen in this context. There is pressure on the FCC now to allow the principle that some websites can buy faster or more reliable service from Internet providers, even though such a principle would allow far greater competition and innovation on the Internet in the long run. And now, in this decision, the FCC suggests that it may want new rules regarding management of congestion. We don’t know that yet – the opinion, on the one hand, tells us that the problem isn’t rules per se, but that companies such as Comcast must apply rules in a non-discriminatory and transparent manner. But on the other hand, the FCC advisory suggests that there’s no way to do that, since techniques that would allow rules to be enforced (such as Deep Packet Inspection) would not be “acceptable.”
So where does the FCC leave us? The Internet does not have limitless capacity, so there are going to be times when it will get congested. One option is to let the market sort it out – a competitive Internet. “Neutrality” advocates don’t like that. Another is to allow broadband providers the ability to implement rules (when needed, and transparently, to be sure) to manage those periods. But “neutrality” advocates don’t like that solution either.
That leaves the third option – letting everyone tough it out and share the misery. It’s the option that gives us highway traffic that doesn’t move, airplanes that sit on tarmacs, and brown-outs during the summer heat. To some advocates, this solution may be in keeping with how they construe the egalitarian character of the Internet. But for users, it’s a blueprint for an Internet that cannot sink but will not swim.
In September 2008, Comcast responded to the FCC by announcing it was testing a program that would slow down those customers using the most bandwidth, regardless of the purposes or programs they were employing, when the system neared congestion. While this approach may be more to the FCC’s liking, it may not be different in practice from Comcast’s discredited approach.
Cannot Sink But Will Not Swim
On August 1, the FCC issued a ruling that Comcast had endangered the “vibrant and open nature of the Internet” by slowing down customer use of BitTorrent and other peer-to-peer applications, and that it had misrepresented its practices to both its customers and the Commission.
The finding would appear decisive, but the Commission’s ruling has something for everybody. “Neutrality” proponents will claim that it justifies their vision of an “unimpeded” end-to-end Internet. Providers will note that the finding doesn’t rule out traffic management during times of congestions, just the clumsy way Comcast went about it. Consumer advocates will take pleasure in the Commission’s championship of transparency. Others will take heart in the Commission’s statement that blocking odious content such as child pornography or pirated intellectual property is still “consistent with Federal policy.” The ruling enunciates clear principles, even if peppered with interesting contradictions, and makes clear that the FCC can do whatever it needs to do without new legislation. So the Commission’s ruling does a great deal for a great many, but it still doesn’t do one thing.
The ruling doesn’t solve the fundamental problem confronting the Internet – what to do when it gets too crowded.
Let’s first take a moment to look at the background of the FCC’s ruling. Several independent groups, including the Associated Press, discovered that Comcast was slowing traffic that used peer-to-peer applications, principally BitTorrent, but others as well. At first Comcast denied it, but then admitted it was true, explaining that it did so only to manage periods of peak congestion. It then changed its story again, admitting that it regularly slows this kind of traffic regardless of the level of congestion.
As ever, it’s the cover-up, not the crime. I don’t understand why Comcast was slowing traffic when there weren’t congested conditions, although I can imagine doing so prevents congestion before the fact. And slowing down one type of application – peer-to-peer – sounds discriminatory, but you can see the argument for it. It’s like saying that you can’t water your lawn when there’s a drought and the reservoir is low – it may be arbitrary (some folks would rather give up bathing than stop watering the lawn), but at least it solves the problem.
Comcast’s putting the burden of limiting congestion by going after peer-to-peer networks has that quality. One reason they drew such fire for doing so is that when you say “peer to peer networks,” people think – piracy. But blaming peer-to-per networks for piracy is like blaming the ocean for pirates – being on the ocean doesn’t make you a pirate, but that’s where all the pirates work. Piracy isn’t Comcast’s concern, nor should it be – that would be like asking rhe airlines to search their passengers for contraband. File swapping increases the demand for bandwidth, and bandwidth is what Comcast sells.
What Comcast doesn’t sell much of over high-speed Internet connections is content, which leads to one of the oddest aspects of the Commission’s finding. It says that Comcast’s motive for reining it BitTorrent-type applications is that those applications “provide Internet users with the opportunity to view high-quality video that they might otherwise watch (and pay for) on cable television.” Yes, Comcast does sell cable-TV service, but the order doesn’t contain proof that Comcast was intervening due to BitTorrent’s competition with Comcast TV services. And would you give up cable TV for what you can get through the Internet? Or more accurately – have you?
No, peer-to-per networks and BitTorrent applications are the likely target for limitation during congested periods because they’re big, incredibly big. P2P does create congestion problems. The bandwidth used by YouTube, a multibillion dollar subsidiary of Google, itself the eleventh largest corporation in America, is greater than the bandwidth of the entire Internet eight years ago. If you want to cut back on bandwidth use, there’s an easy target. And, at first blush, it’s not a bad presumption that reining back peer-to-peer has lower social costs than, say, e-commerce, or school or government or health networks.
Is that unfair, or arbitrary? Well, that takes us back to the question I posed earlier -- what do we do when the Internet is crowded? And that’s what I’ll address later today.
#2
So what do we do when the Internet is crowded? Anything that has a fixed level of capacity is vulnerable to too much demand. When there are too many cars on the road, there’s traffic and people sit and idle their engines. So Comcast’s solution was arbitrary (and inexpertly administered), but ultimately no more so than telling people they can’t water their lawn during a drought.
Was Comcast’s move the best solution? No, not by a long shot. The best solution is to let users and websites determine what speed and reliability are worth to them. Different websites will value speed differently, and for different reasons. A health clinic with real-time monitors or an on-line stock trader will think every increment of speed and reliability is worth it. A new competitor to Google might value speed highly to compete with the incumbent by being faster. So some sites are simply going to be willing to buy the right to priority for their site’s traffic when the Internet gets crowded. It would be like having HOV lanes that let some traffic pass more quickly.
The people who propose “net neutrality” oppose this vision of a competitive Internet. They argue that the Internet would lose its “open and vibrant” nature if this were allowed. But it’s hard to see how. For example, some people send their mail by priority overnight, some send it by second day delivery, and some with a regular stamp, but all the mail gets through. And some businesses have 24/7 live customer service lines, some only during office hours, and some respond to e-mail the next day – they decide what that level of service is worth and act accordingly … and so, too, do customers.
Why should the Internet be any different? Neutrality proponents point to the Internet’s epochal character and its potential to bring together all of humanity in a new and profound way. No one argues that. But the singularity of the Internet doesn’t tell us what to do to use it in a way that does the most good for the greatest number. And when it gets congested, the best answer probably isn’t the one “neutrality” proponents favor – everybody being slow together -- hey, at least it’s “neutral.” Besides, that’s not what happens now anyway. The big sites – Google, EBay, and so on – already have big speed advantages due to more and better equipment (including caching) that smaller guys can’t get. That’s why they’re the money behind “neutrality” – because “neutrality” preserves their advantages.
The FCC ruling has to be seen in this context. There is pressure on the FCC now to allow the principle that some websites can buy faster or more reliable service from Internet providers, even though such a principle would allow far greater competition and innovation on the Internet in the long run. And now, in this decision, the FCC suggests that it may want new rules regarding management of congestion. We don’t know that yet – the opinion, on the one hand, tells us that the problem isn’t rules per se, but that companies such as Comcast must apply rules in a non-discriminatory and transparent manner. But on the other hand, the FCC advisory suggests that there’s no way to do that, since techniques that would allow rules to be enforced (such as Deep Packet Inspection) would not be “acceptable.”
So where does the FCC leave us? The Internet does not have limitless capacity, so there are going to be times when it will get congested. One option is to let the market sort it out – a competitive Internet. “Neutrality” advocates don’t like that. Another is to allow broadband providers the ability to implement rules (when needed, and transparently, to be sure) to manage those periods. But “neutrality” advocates don’t like that solution either.
That leaves the third option – letting everyone tough it out and share the misery. It’s the option that gives us highway traffic that doesn’t move, airplanes that sit on tarmacs, and brown-outs during the summer heat. To some advocates, this solution may be in keeping with how they construe the egalitarian character of the Internet. But for users, it’s a blueprint for an Internet that cannot sink but will not swim.
In September 2008, Comcast responded to the FCC by announcing it was testing a program that would slow down those customers using the most bandwidth, regardless of the purposes or programs they were employing, when the system neared congestion. While this approach may be more to the FCC’s liking, it may not be different in practice from Comcast’s discredited approach.