From the lowliest blogger to Jon Stewart, everybody is laughing at Sen. Ted Stevens and his remarks (1.2MB mp3) on net neutrality. The sound bite about the Internet being “a series of tubes” has come in for for the most ridicule.

I’ll grant that Stevens sounds pretty confused on the recording. But’s let’s give the guy a break. He was speaking off the cuff in a meeting, and he sounds a bit agitated. Have you ever listened to a recording of yourself speaking in an unscripted setting? For most people, it’s pretty depressing. We misspeak, drop words, repeat phrases, and mangle sentences all the time. Normally, listeners’ brains edit out the errors.

In this light, some of the ridicule of Stevens seems a bit unfair. He said the Internet is made up of “tubes”. Taken literally, that’s crazy. But experts talk about “pipes” all the time. Is the gap between “tubes” and “pipes” really so large? And when Stevens says that his staff sent him “an Internet” and it took several days to arrive, it sounds to me like he meant to say “an email” and just misspoke.

So let’s take Stevens seriously, and consider the possibility that somewhere in his head, or in the head of a staffer telling him what to say, there was a coherent argument that was supposed to come out of Stevens’ mouth but was garbled into what we heard. Let’s try to reconstruct that argument and see if it makes any sense.

In particular, let’s look at the much-quoted core of Stevens’ argument, as transcribed by Ryan Singel. Here is my cleaned-up restatement of that part of Stevens’ remarks:

NetFlix delivers movies by mail. What happens when they start delivering them by download? The Internet will get congested.

Last Friday morning, my staff sent me an email and it didn’t arrive until Tuesday. Why? Because the Internet was congested.

You want to help consumers? Consumers don’t benefit when the Net is congested. A few companies want to flood the Internet with traffic. Why shouldn’t ISPs be able to manage that traffic, so other traffic can get through? Your regulatory approach would make that impossible.

The Internet doesn’t have infinite capacity. It’s like a series of pipes. If you try to push too much traffic through the pipes, they’ll fill up and other traffic will be delayed.

The Department of Defense had to build their own network so their time-critical traffic wouldn’t get blocked by Internet congestion.

Maybe the companies that want to dump so much traffic on the Net should pay for the extra capacity. They shouldn’t just dump their traffic onto the same network links that all of us are paying for.

We don’t have regulation now, and the Net seems to be working reasonably well. Let’s leave it unregulated. Let’s wait to see if a problem really develops.

This is a rehash of two of the standard arguments of neutrality regulation opponents: let ISPs charge sites that send lots of traffic through their networks; and it’s not broke so don’t fix it. Nothing new here, but nothing scandalous either.

His examples, on the other hand, seem pretty weak. First, it’s hard to imagine that NetFlix would really use up so much bandwidth that they or their customers weren’t already paying for. If I buy an expensive broadband connection, and I want to use it to download a few gigabytes a month of movies, that seems fine. The traffic I slow down will mostly be my own.

Second, the slow email wouldn’t have been caused by general congestion on the Net. The cause must be either an inattentive person or downtime of a Senate server. My guess is that Stevens was searching his memory for examples of network delays, and this one popped up.

Third, the DoD has plenty of reasons other than congestion to have its own network. Secrecy, for example. And a need for redundancy in case of a denial-of-service attack on the Internet’s infrastructure. Congestion probably ranks pretty far down the list.

The bottom line? Stevens may have been trying to make a coherent argument. It’s not a great argument, and his examples were poorly chosen, but it’s far from the worst argument ever heard in the Senate.

Why then the shock and ridicule from the Internet public? Partly because the recording was a perfect seed for a Net ridicule meme. But partly, too, because people unfamiliar with everyday Washington expect a high level of debate in the Senate, and Stevens’ remarks, even if cleaned up, don’t nearly qualify. As Art Brodsky of Public Knowledge put it, “We didn’t [post the recording] to embarrass Sen. Stevens, but to give the public an inside view of what can go on at a markup. Just so you know.” Millions of netizens now know, and they’re alarmed.

One of the standard arguments one hears against network neutrality rules is that network providers need to provide Quality of Service (QoS) guarantees to certain kinds of traffic, such as video. If QoS is necessary, the argument goes, and if net neutrality rules would hamper QoS by requiring all traffic to be treated the same, then net neutrality rules must be harmful. Today, I want to unpack this argument and see how it holds up in light of computer science research and engineering experience.

First, I need to make clear that guaranteeing QoS for an application means more than just giving it lots of bandwidth or prioritizing its traffic above other applications. Those things might be helpful, but they’re not QoS (or at least not the kind I’m talking about today). What QoS mechanisms (try to) do is to make specific performance guarantees to an app over a short window of time.

An example may clarify this point. If you’re loading a web page, and your network connection hiccups so that you get no traffic for (say) half a second, you may notice a short pause but it won’t be a big deal. But if you’re having a voice conversation with somebody, a half-second gap will be very annoying. Web browsing needs decent bandwidth on average, but voice conversations needs better protection against short delays. That protection is QoS.

Careful readers will protest at this point that a good browsing experience depends on more than just average bandwidth. A half-second hiccup might not be a big problem, but a ten-minute pause would be too much, even if performance is really snappy afterward. The difference between voice conversations and browsing is one of degree – voice conversations want guarantees over fractions of seconds, and browsing wants them over fractions of minutes.

The reason we don’t need special QoS mechanisms for browsing is that the broadband Internet already provides performance that is almost always steady enough over the time intervals that matter for browsing.

Sometimes, too, there are simple tricks that can turn an app that cares about short delays into one that cares only about longer delays. For example, watching prerecorded audio or video streams doesn’t need QoS, because you can use buffering. If you’re watching a video, you can download every frame ten seconds before you’re going to watch it; then a hiccup of a few seconds won’t be a problem. This is why streaming audio and video work perfectly well today (when there is enough average bandwidth).

There are two other important cases where QoS isn’t needed. First, if an app needs higher average speed than the Net can provide, than QoS won’t help it – QoS makes the Net’s speed steadier but not faster. Second – and less obvious – if an app needs much less average speed than the Net can provide, then QoS might also be unnecessary. If speed doesn’t drop entirely to zero but fluctuates, with peaks and valleys, then even the valleys may be high enough to give the app what it needs. This is starting to happen for voice conversations – Skype and other VoIP systems seem to work pretty well without any special QoS support in the network.

We can’t say that QoS is never needed, but experience does teach that it’s easy, especially for non-experts, to overestimate the importance of QoS. That’s why I’m not convinced – though I could be, with more evidence – that QoS is a strong argument against net neutrality rules.