Just What Do They Mean by 'Net Neutrality'?
On December 1, FCC Commission Chair Julius Genachowski made a speech announcing proposed rules through which the FCC proposed to establish “net neutrality.” There is much controversy about the rules and about “net neutrality” in general. The frustrating part: when you examine the arguments closely, it’s clear that no two commentators appear to be talking about the same thing.
So what does “net neutrality” mean?
To answer that question, first of all we need to understand a little bit about the Internet and how it works. (More technical readers can skip this section, or just amuse yourselves observing the simplifications and omissions I’m making to keep this reasonably concise.)
The Internet arose from ARPAnet, and ARPAnet arose as the answer to a very interesting question: how can you build a reliable communications network when your switching stations may, at any moment, turn into a very large, glowing hole in the ground? Traditional telephone networks -- known as circuit-switching networks -- wouldn’t do it: they depend on having telephone switches that make fixed connections for the duration of a call.
You can think about a circuit-switched network by thinking of the old switchboard telephone operators. When you made a call, your local operator takes a wire from your phone, and plugs it into the socket to send the call to another phone. You’re now connected, and the call works as long as the operator doesn’t unplug you again. When you made a long-distance call -- and frighteningly, I still remember when long-distance calls were made this way -- you got in touch with the local operator, who then used a separate line to talk to a long-distance operator, who talked to one or more other operators; when they had all talked it over, various plugs went into the appropriate sockets. The result was that there was literally a wire connection going from your phone to the person at the other end.
Of course, if something happened to any of the connections, from someone clumsily kicking out a plug to an atomic bomb going off in the neighborhood, you were cut off.
The ARPAnet was based on another idea, called packet-switching. Packet-switching works more like sending mail: your message is split up into separate packets, and each packet is sent on independently to its destination. Think of these packets like postcards: you want to send your new NaNoWriMo novel to your best friend, who will read it and praise it, making you feel ever so much better. So you split it up, putting the text onto postcards, and mail each postcard to your friend.
Now, the post office takes each card, sorts it, and sends it to a mail distribution center, which sends it to another center near your friend; from there it goes to the mail man letter carrier, who carries it to your friend’s house and drops it in the mud mailbox. Over time, you will get the whole novel, and if something happens to the distribution center in the middle, the Post Office simply re-routes the following postcards through another location. Other than some details, like what happens to a lost postcard (number them, and send a postcard back asking for copies of the missing pieces), this is basically how the whole Internet works.
In the ARPAnet, and later the Internet, all your data is broken into packets (postcards) with a destination address and a sender address. If someone blows up the Omaha switching center, messages are simply routed, say, to Fargo in its place. As far as the network is concerned, all it cares about are the addresses.
Which leads us to net neutrality (you thought I’d forgotten). The basic assumption about the Internet is that it doesn’t matter what the content parts of the packet are, just as a postman doesn’t care if you’ve written a letter in English, Swahili, Klingon, or a special code of your own invention. They just deliver to the address given.
But now, what if the Post Office had a special deal with Hallmark, so that if you were sending a Hallmark card, it got special treatment -- or what if the Post Office decided that it was carrying too many Hallmark cards, so it set a limit that it would only let a letter carrier carry 100 cards a day? Obviously, this would mean that Hallmark was very much dependent on staying in the good graces of the Post Office, and should the Post Office decide to go into the card business themselves, Hallmark could be in real trouble.
Around 2004, some of the big Internet service providers like Comcast decided to do just that: worried about people using file sharing or sending lots of video, they started looking at the contents of packets to see what was being sent, and throttling how much data of specific types a customer could receive. This led to the first push for net neutrality, of a sort that we might call net neutrality of the first kind, or “content neutrality.” Along with that, people demanded that they be able to connect their own equipment to their networks, and that third parties like new equipment companies had to be allowed to build new Internet equipment themselves.
The push for net neutrality, however, was quickly picked up by other people for other political purposes, starting with the idea that “net neutrality” meant that everyone ought to have equal access to Internet service, whether they live in the borough of Manhattan, or Manhattan, Kansas, or in a cabin twelve miles by road from the nearest human habitation. This quickly picked up other ideas: that “net neutrality” meant different ethnic groups have equal access -- which would mean the government looking not just at the content of the messages, but the race of the person on the wire; or that different viewpoints ought to have equal access to the Internet -- so, potentially, Fox News would be limited based on how much bandwidth the Huffington Post consumed, and even that “hate speech” and “lies” could be regulated.
This is what we might call net neutrality of the second kind: not content-neutral, but instead, content “fairness.”
Stated that baldly, a lot of people would object (and rightly, I believe). The original Obama administration proposals were much more intrusive than what we know of the new proposal -- “what we know” because, oddly, the proposed rules haven’t been made public. Based on Genachowski’s speech, the new rules sound reasonably benign -- which, to a cynic like me, immediately leads to a second question: why aren’t the new rules being published?
Add to that other recent proposals, like the Obama administration’s proposal for a law authorizing fairly intrusive regulation, up to the legal right to completely shut down the Internet in an “emergency.” Or the recent BBC interview in which another FCC commissioner, Michael J. Copps, suggests that the broadcast media should have to meet a more stringent “public values test” for license renewal -- and makes it clear that “fairness” and “local content” are parts of that test as he sees it. One might object that this is only for the broadcast media, except for the number of other politicians who have proposed the FCC should also regulate the cable networks and Internet.
This, then, is the question I think everyone must ask when discussing net neutrality: do we mean the content-neutral, don’t read-the-postcards net neutrality of the first kind? Or the “fair access” net neutrality of the second kind?