Earlier in December, I wrote a blog entry entitled “Net Neutrality and other Myths,” in which I discussed some of the issues with net neutrality. I have since changed the title of that article to “Net Neutrality and other Misnomers” because net neutrality is not so much fiction as it is not a clear name for the issue (though alliterative and catchy).
Net neutrality is not a “myth” as much as a misnomer because from a technical perspective, all networks must discriminate (e.g., set priority levels) to some extent in order to function when there is insufficient bandwidth for all network requests to be handled at the same time. Therefore, no network can be completely “neutral” about the traffic it carries and still function for its users. This is the same reason that the postal service offers different rates for different speeds of delivery of a package. You pay more for priority shipping than standard shipping, otherwise the postal service would take longer for everyone who shipped through it (and might not even work as well as it does today for the billions of packages it handles each year). Just like the USPS, computer networks have to set priority for some traffic in order to work properly.
The net neutrality debate is not so much about this technical problem as it is about the concern that service and content providers may become aligned to prioritize the delivery of specific content at the expense of other content providers, driving some off a network such as Verizon. This, I agree, is a bad thing. I don’t think Comcast should have the right to contract with a content provider like NBC, and then de-prioritize content from TNT or CBS. This is like USPS determining priority of your package based on whether it is a legal opinion letter to your client or junk mail to switch back to Comcast. I don’t like the idea of the postal service opening my letter to you to figure out how fast it should deliver the package (or deliver it at all, for that matter). Instead, I prefer to pay them more to make the letter, irrespective of its content, get to its intended recipient faster.
This is, in fact, the state of internet access today. Service providers charge more for faster (or symmetrical) internet access as compared to their “standard” service offering. If you want symmetrical gigabit internet access, you can get it, but it will cost far more than the $55 per month I pay for my Fios internet service at home. Net neutrality would effectively require that I should not have to subscribe to one service provider’s service in order to get faster access to a certain kind of content.
A concern expressed by one reader of my blog was that without net neutrality, the major service providers will have no incentive to increase the bandwidth available to subscribers. This concern, however, flies in the face of the actual increase in bandwidth since the 1990s for residential subscribers of internet service, absent any clear “net neutrality” principles enforced by the FCC.
Back in 1993, my 486 came equipped with a 2400 baud modem, which would allow me to send and receive data at the whopping speed of 2.4 kilobits per second. When I first signed on with America Online for dial up service in 2000, the best modem speeds were 56 kilobits per second. Therefore, in seven years, internet access speeds had increased by a factor of 23. In 2003, I was among the first subscribers in my neighborhood for aDSL, with a top speed of 768 kilobits per second (though, upload speeds were slower – 128 kilobits). In that period of three years, my internet access had increased by a factor of 13. In 2009, I became a Fios customer, which provides me with about 15000 kilobits per second of download speeds (and 5000 kilobits upload). In that period of six years, my available bandwidth increased by a factor of 19. From 1993 to 2009, overall internet speed available to me in Maryland has increased at a steady rate. However, the cost per kilobit of bandwidth is where the real story is with regards to improvement:
|Bandwidth (kbs)||Annual Cost (in 2008 dollars)||Unit Cost per Kbs|
Back in 1993, bandwidth was expensive per kilobit. Today, the amount of bandwidth available to me costs almost nothing per kilobit. (I grant that my upload speeds are slower with DSL and Fios – Fios’ upload is a mere 5000 kilobits per second, which costs about $0.12 per kilobit – still a dramatic drop in cost from 1993). And all of this occurred without a firm “net neutrality” policy in place.
Now, the writer pointed out that the phone companies that control most internet access for consumers have had no incentive to help out VOIP, which is essentially a competing phone service that utilizes one’s internet connection to send and receive phone calls. Ironically, a purist net neutrality policy would actually hinder improvements to VOIP. This is because, for VOIP service to actually improve, VOIP packets themselves would need to be given priority across the internet so that a greater percentage of these packets would reach their destination. Strict net neutrality insists on no differentiation in content for the purposes of priority.
Of course, Verizon and the other major phone carriers have not wanted to encourage VOIP, at least when it first came out, because VOIP was direct competition to phone service, especially long distance service (where the majority of savings on cost were for VOIP users). Today, Vonage and Skype both offer competing local and long distance services that ride on a user’s internet access. In fact, an article (click here) indicates that both service providers have expanded their international calling plans at flat rates to between forty and sixty countries. In addition, cell phone plans with flat rates for local and long distance plans are becoming increasingly popular.
Would net neutrality have improved VOIP service sooner? Perhaps. But if you wanted a phone call on VOIP to be as reliable as your POTS line at home in 2003, the FCC instead would have needed to require internet service providers to prioritize VOIP packets over other kinds of traffic being carried on their networks. This would not really be “neutrality.” But, I think the point is that the large telecommunications companies did not appreciate the competition for one of their core products, and were not going to do anything to support the technology or the potential vendors that might sell them.
I don’t think it is clear, however, that a government mandate to support a technology like VOIP would have led to more reliable VOIP services for consumers. Back in the 1990s, the government did intervene to try and increase competition in the local phone service market, by requiring that the major carriers like Verizon allow smaller competitive local exchange carriers (CLEC) to have access to Verizon equipment in order to deliver competitive local phone service. Initially, a lot of new vendors came into the market and started providing service, such as WinStar, WorldCom, and others. At the end of the dot-com bubble in 2001, many of these CLECs closed their doors because, even with government intervention, the market could not support that many local exchange carriers. And CLECs got a bad rap in the market because many turned out to be vapor-vendors that sold more air than actual services to customers. In some cases, these vendors had a competing product that delivered poorer or less reliable service than Verizon for local phone service. And at least one of these CLECs, WorldCom, found its CEO indicted for cooking the books and fraudulently overbilling customers.
Today, Verizon and Comcast have expanded their reach to producing content available to internet users. Comcast recently purchased NBC. Verizon’s FIOS service includes on demand services that provide, in some cases, exclusive access to content. The principles of net neutrality would require that Comcast not make access to NBC content (like Law & Order) faster on its network than content from competing content providers. And I fully agree with this principle.
But absent such a policy, I’m not convinced that network speeds to the internet will drastically slow down because the major players of internet access also control some of the major content producers for the internet. To the contrary, I agree with Ray Kurzweil that we are really in the knee of the curve with regards to the amount of bandwidth available to consumers. Even more dramatic speed increases in CPU, memory, computing power and bandwidth are coming and will likely arrive in my lifetime, if not sooner. See Ray Kurzweil, The Singularity is Near (Viking, 2005). And I do not believe that any one entity has sufficient control over this growth to have a halting influence over it. The increasing complexity and sophistication of technology is driven by more than any single telecommunications carrier, no matter its size or assets. To stop this progression at this point would require that a dinosaur-killer scale comet hit the earth, or more likely, every teenager on the planet stopped using their cell phones, twitter and facebook.
Furthermore, the internet itself, without net neutrality, has fostered an explosion of user-generated content. The growth here is to such an extreme that there has been a significant decentralization of the entertainment industry. See The Long Tail by Chris Anderson. The internet itself, by allowing more people to self-publish, has moved us further away from the centralized publishing of content by the few. And the fact is that all of us that want to self-publish are willing to pay for internet access sufficient to do so (either by sitting at Starbucks each day on their WiFi connection, or by paying $55 a month for FIOS at home).
In sum, net neutrality is a misnomer, telecommunications companies should not be able to discriminate against access to content they do not own, and even if they are doing this, the world and the technology in it is moving on without them.