Editore"s Note
Tilting at Windmills

Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

April 25, 2006
By: Kevin Drum

NET NEUTRALITY, TAKE 2....Let's take another crack at the net neutrality debate. Perhaps we can pop a collective gasket on this. Since I'm going to present some arguments both pro and con, though, please read the whole thing before leaving a trail of steaming gasket parts in comments.

This is also a long post, so I'm going to stick the rest of it below the fold.

First, here are some arguments against strict net neutrality and in favor of allowing tiered internet service:

  1. Video-on-Demand is a market I know a little bit about (at least, I did back when I worked for a startup VOD company), and the bandwidth and service issues that face commercial VOD rollouts are quite real. You and I may or may not care about VOD, but a lot of people do, and when telecom companies say that they need to make substantial investments to support large-scale VOD, they're right. When they further say that these investments will only be worthwhile if they can guarantee VOD service that works reliably and well, they're right again.

    The question, of course, is whether the only way to provide reliable VOD service over the internet is to offer a tiered service to video providers. I don't know the answer to that, but it's not transparently absurd to think that the answer might be yes.

  2. If they're freed from net neutrality rules, long haul carriers like AT&T and Verizon will have a big incentive to degrade the service of internet phone (VoIP) suppliers like Vonage, since Vonage is a direct competitor for both consumer and commercial telephone service. In fact, Vonage has faced discrimination in the past and suspects that it faces ongoing discrimination in several current cases. Thus, VoIP companies like Vonage have the most to lose from the demise of net neutrality.

    But here's the thing: Last year Vonage said it was satisfied (though not thrilled) with the net neutrality provisions in the Barton-Rush bill that's currently working its way through Congress. The bill has been modified since then, but as near as I can tell Vonage hasn't lobbied against it. When their CEO testified on the bill a few weeks ago, the only subject he brought up was 911 services. He didn't even mention net neutrality.

    So: if Vonage is satisfied, maybe the bill isn't all that bad?

  3. There are technical reasons to prefer a packet neutral internet architecture, and Henry Farrell outlines one of them here. The problem is that these arguments are very subtle, to the point where they become nearly religious in nature. I've been hanging around network geeks for a couple of decades now, and these kinds of religious wars are pretty familiar to me.

    Will the Barton-Rush bill doom the internet we know and love? Well, back in the early 90s I remember all the wailing and gnashing of teeth about how allowing commercial access to the internet would ruin everyone's favorite sandbox, but guess what? That experiment turned out pretty well. Bottom line: network purists are constantly arguing that the sky will fall if some proposal they dislike is adopted. I'd take them with a grain of salt.

  4. The key issue in the Barton-Rush bill is adjudication vs. rulemaking. I'm sure everyone else arguing about this issue is an expert in regulatory law, but I'm not and I can't immediately tell how big a deal this is.

    Basically, the argument is whether Congress should mandate some kind of net neutrality regime and task the FCC with making rules to implement it, or whether they should set out general principles, let things unfold, and allow the FCC to adjudicate complaints if and when they're submitted. Rules have the virtue of being proactive, but also have the potential to hammer something into place that will turn out not to make sense. Adjudication is more flexible, but it's also a lot slower. It allows telcos to stretch Barton-Rush's net neutrality principles far enough to (possibly) put competitors out of business, safe in the knowledge that it will take years for the FCC to tell them to cease and desist.

So there you have it: several good reasons to wonder if the Barton-Rush bill is really as bad as everyone is making it out to be. Now for the other side.

  1. Net neutrality is essentially a "common carrier" requirement, and history has shown that common carrier rules are good things. Take railroads, for example. In the past, before common carrier requirements were adopted, railroads discriminated with abandon, shutting out startups in return for kickbacks from big companies like Standard Oil. This was bad for startups and bad for the economy, and there's no special reason to think that human nature has become more altruistic in the past century. If phone companies aren't required to act as common carriers, there's every reason to think they'll eventually throttle innovation in return for lucrative kickbacks the same way that railroads did in the 19th century.

  2. Reed Hundt answers my plea to chime in on the net neutrality debate at TPMCafe, but essentially punts. However, he did give a speech at Freedom2Connect a couple of weeks ago in which he made a strong argument in favor of net neutrality and came out pretty clearly in favor of rulemaking vs. adjudication:

    We have the four [net neutrality principles in Barton-Rush], but they are a palsied, weak, shadow compared to rules. They are not in substance addressing the issues. They do not provide guides or signs, they do not provide a program, and theyre issued in the context of the House subcommittee saying no rules, do everything case by case.

    That means all decisions are about what happened 5-7 years ago. They mean any company can avoid an adjudication until its irrelevant. That does not impact capital planning, strategy or how networks are built. Its a rear view window look.

  3. More interesting, I thought, was his primary argument that we should simply make the whole problem go away by writing a check:

    The public ought to create a public thoroughfare to the Internet, by regulation, just like in France, Japan and Korea. It ought to be fast. Every year it ought to be faster....Were talking about basing our information economy on an investment that is maybe $20 billion, for Fiber to the Home. If you had that it would be the work of a moment to allocate it. Id write that check in a heartbeat.

    If we did this, he says, the whole issue disappears. When every home has cheap 100 Mbps access to the internet, services like VOD are a no-brainer. They can travel over a neutral internet with no special tiering at all. And $20 billion is a tiny price to guarantee equal access for all to the "public space...where democracy will be defined."

  4. Finally, here's probably the most convincing argument in favor of net neutrality: the telecom industry is against it. As near as I can tell, most telecom CEOs would sell their mothers into white slavery if they thought it would help them keep one of their competitors at bay for a year or five longer, and their record of bending, breaking, and twisting the rules in order to maintain their monopoly position without which none of this would really matter in the first place would fill a phone book. Frankly, you can't go too far wrong simply taking the opposite side of the telecom industry on every relevant issue.

Bottom line: given the potential for abuse with tiered internet access, ordinary prudence suggests that loosening up on strong net neutrality is probably a bad idea unless someone is keeping a pretty close eye on the consequences and with the contemporary Republican Party in charge of oversight, that's obviously not going to happen. Still, I think Reed Hundt alludes to the real issue here, which, as he says, has been swept under the carpet for too long. The fact is, as long as the long-haul backbone of the internet is in private hands, it's hard to justify not allowing its owners to pursue whatever pricing schemes they want. It doesn't mean we can't regulate them, just that it's always going to be a rear guard action.

Conversely, if the federal government subsidized the whole thing at the cost of a few billion dollars a year, just as they did with the interstate highway system half a century ago, we could build an internet backbone that would be cheap, universal, public, and a huge boost to the American economy. The feds don't even have to own it, they just have to pay for it.

I don't know if that's the answer, but it's too bad it's not at least up for discussion. If Japan can do it, why can't we?

Kevin Drum 9:35 PM Permalink | Trackbacks | Comments (66)

Bookmark and Share
 
Comments

Can we start off by not calling each other idiots until we say something idiotic?

Because if confession of ignorance of highly technical stuff makes you an idiot, well, who's not in trouble?

Posted by: frankly0 on April 25, 2006 at 9:51 PM | PERMALINK

I was just thinking that I am a bit safter than most people at least, a little, because my ISP is a municipal wireless company. It's basically the local government so it has to adhere to certain state regs etc.

Posted by: MNPundit on April 25, 2006 at 9:53 PM | PERMALINK

Bottom line: network purists are constantly arguing that the sky will fall if some proposal they dislike is adopted

unlike purists of any other kind. political purists are especially immune to this kind of thinking.

Posted by: cleek on April 25, 2006 at 9:54 PM | PERMALINK

Net neutrality is essentially a "common carrier" requirement, and history has shown that common carrier rules are good things.

But the net neutrality law is not really net neutral because it regulates businesses and tells businesses how they should behave. It does not allow the free market to determine whether or not customers want ISPs to charge companies like Google and Yahoo for stealing their bandwith. The best policy is to allow customers to decide how they want their ISPs to act through the free market.

As near as I can tell, most telecom CEOs would sell their mothers into white slavery if they thought it would help them keep one of their competitors at bay for a year or five longer,

In the free market if that happens, customers would stop using the telecom's broadband and so they will close as a business. So we don't have to worry about that.

More interesting, I thought, was his primary argument that we should simply make the whole problem go away by writing a check:

This would be horrible because it would be stealing the internet from the telecoms who created the internet in the first place. Since telecoms created the internet, they deserver to keep it instead of having a communist government takeover of it.

The question, of course, is whether the only way to provide reliable VOD service over the internet is to offer a tiered service to video providers. I don't know the answer to that, but it's not transparently absurd to think that the answer might be yes.

I agree with you. The answer is yes.

Posted by: Al on April 25, 2006 at 9:56 PM | PERMALINK

Telcos Take It on the Chin at Judiciary Hearing Good of them to upstage the pathetic cast of the House commerce committee.

Posted by: John on April 25, 2006 at 9:56 PM | PERMALINK

Conversely, if the federal government subsidized the whole thing at the cost of a few billion dollars a year, just as they did with the interstate highway system half a century ago...

I'm all for the feds subsidizing text and even images, but spending billions on TV bandwidth? That is likely to promote far more entertainment than substance. I'm less sure (but still skeptical) about VOIP.

An information superhighway is one thing. A video superhighway is another.

Nice presentation, btw.

Posted by: Measure for Measure on April 25, 2006 at 9:57 PM | PERMALINK

"Conversely, if the federal government subsidized the whole thing [Fiber to the Home] at the cost of a few billion dollars a year, just as they did with the interstate highway system half a century ago, we could build an internet backbone that would be cheap, universal, public, and a huge boost to the American economy."

That's precisely the analogy I use everytime I discuss this issue.

***

Didn't the Dems recently propose to have universal access to broadband (defined as something ludicrous like 1.5 Mbps) in the next few years? How shortsighted -- let's make a policy for 100 Mbps in the same timeframe!

Posted by: Mike D on April 25, 2006 at 9:58 PM | PERMALINK

Al, honey. it's getting old. really. old, stale and boring.

Posted by: cleek on April 25, 2006 at 9:59 PM | PERMALINK

lHow shortsighted -- let's make a policy for 100 Mbps in the same timeframe!

by the time we got 100Mbps rolled-out, we'll be able to buy 10Tbps connections at Best Buy

Posted by: cleek on April 25, 2006 at 10:00 PM | PERMALINK

...and a huge boost to the American economy.

More like a huge mixed bag. I'd suspect that the telecoms would be hurt by VOIP, most businesses would enjoy lower phone bills and the movie and music biz would be creamed by P2P.

Posted by: Measure for Measure on April 25, 2006 at 10:05 PM | PERMALINK

Al It does not allow the free market to determine whether or not customers want ISPs to charge companies like Google and Yahoo for stealing their bandwith.

But Google and Yahoo aren't stealing bandwith. Bandwith is paid for by users. Do you really think that when Verizon et al set their rates for internet subscribers, they don't factor in all the expenses involved in delivering it, plus a nice profit?

What the telecoms are trying to get away with is like this: suppose you ran a business, and your product was delivered by FedEx, with your customers paying FedEx for it. Now suppose FedEx came to you one day and said, "You are making a nice profit off our delivery service. Besides what your customers pay, I also want you to pay us for it, or else your deliveries are going to be a lot slower, if they make it there at all."

Posted by: Les Brunswick on April 25, 2006 at 10:13 PM | PERMALINK

What cleek said, Al. You're not even trying. Ever heard of the term 'monopoly'?

Posted by: tom on April 25, 2006 at 10:16 PM | PERMALINK

I'm sorry, you lost me when you told me I would have to look beneath your folds.

As for video on demand? If I choose COX (I recently chose away from COX to DSL) I can choose a 6M downlink line for $30 per month, or a 9M guaranteed low latency line for $50 per month.

I imagine YouTube pays for bigger pipes than does say Chris Day. YouTube probably sits closer to the network hubs than Chris does too.

So it seems that the high bandwidth applications are still being paid for but in a non-discriminatory manner.

Uh, when do your free cell minutes start? Does that affect your calling behavior?

Posted by: jerry on April 25, 2006 at 10:25 PM | PERMALINK

As near as I can tell, most telecom CEOs would sell their mothers into white slavery if they thought it would help them keep one of their competitors at bay for a year or five longer

Um, "white slavery"?

Posted by: bc on April 25, 2006 at 10:42 PM | PERMALINK

As a lawyer who works with agencies that have both rulemaking and adjudicatory powers (but in a different field), rulemaking tends to make more sense. Adjudication is subject to regulatory capture (timber industry) while large scale rulemaking gives the general public a seat at the table. Rules can also be written in a way as to allow for adjudicated procedures in fact-specific cases.

btw, any computer geek knows that the backbone of the internet was created by DARPA and american universities.

Posted by: Francis on April 25, 2006 at 10:47 PM | PERMALINK

"Conversely, if the federal government subsidized the whole thing at the cost of a few billion dollars a year, just as they did with the interstate highway system half a century ago, we could build an internet backbone that would be cheap, universal, public, and a huge boost to the American economy."

The Common Good strikes again!

Posted by: anrig on April 25, 2006 at 10:48 PM | PERMALINK

Okay, so I'm an idiot. I read Reed's post, and did not garner the idea that the federal government should be the backbone provider. I think that's the solution. It's by far the cheapest answer. If it stays open, the internet will contribute as much to the american economy as the highway system has.

Posted by: JayAckroyd on April 25, 2006 at 10:49 PM | PERMALINK

I want to talk briefly about Reed Hundt's argument that we should just spend the peanuts that it would take to build a public common carrier data network. I'm in the process of writing a book that covers, among other things, the hundred-year failure of the federal government to build a common carrier electricity transmission network, resulting in a cumulative cost due to inefficiencies and regional monopoly practices by private electric utilities that is probably in the range of low trillions of dollars.

The story there is basically the same as the story here, and has been replayed, over and over, with the notable exception of the Roosevelt administration during the Depression, since Jackson was president. The federal government doesn't do infrastructure. They give away stuff in hopes that somebody does it for them. In the case of railroads, it was land. In the case of nuclear power, it was all of the R&D and prototyping, free insurance, huge subsidies, and the entire front-end and back-end infrastructure (built, again, by subsidized private monopolies).

Clearly, there are times when a market solution works best, and times when a public infrastructure solution works best. My point is that this country has never, ever made the right choice when there is a clear advantage to building a public national infrastructure, and even though it's a great idea, it will not happen now. Ever. Even if it cost the federal governement nothing, but deprived private companies of a chance to make money. Even if the companies that make money are effectively monopolies, at enormous cost to the overall economy.

For that reason, while a clear policy mandate for common carrier access, backed by real consequences for monopolies that try to game that mandate, is far from the best choice, it's the best choice given the hand that we've been dealt. The worst choice is creating a new niche for private, unregulated , monopoly control of critical public infrastructure, which is exactly what the big telecom and cable companies are gunning for in this legislation.

Posted by: Kevin Bell on April 25, 2006 at 10:49 PM | PERMALINK

Please, stealing BW from the poor old telecoms. Sorry to beat a Billy Goat Gruff eater, but this is pathetic.

One day, Google went shopping for an OC12 or whatever. You don't think the the telcoms didn't fall all over themselves to offer great deals and service to get that contract. Were they being robbed at that point?

At the other end, they were pumping all the great special introductory offers to the individual consumers. Hmmm, sounds like another case of the telcos being held hostage.

Ultimately, the telcos are somewhat limited in their ability to differentiate themselves from each other by capital investment or resources. Instead the best place for them to shine and stick it to their competitors is by providing the best services. Unfortunately, this runs counter to most of their basic tenets as they prefer to optimize profits & shareholder earnings by triming that which creates the best services ... a well trained and staffed operations and support team. You know, the people that build the networks, troubleshoot the networks and appraise the customer regarding their networks status.

Posted by: tanstaafl on April 25, 2006 at 10:52 PM | PERMALINK

Yes, please on Kevin's last point. Right now there is so little incentive for any kind of broadband to trek out to the countryside.

Posted by: Tilli (Mojave Desert) on April 25, 2006 at 11:17 PM | PERMALINK

common carrier "net neutrality" should be preserved on the internet. Telecom concerns about bandwidth and "making sure high-bandwidth applications work" are ridiculous. If they provide the bandwidth, those who want to use it will pay for it. Period. THAT is allowing the market (users) to set the rules and make things work. Abandoning net neutrality is letting the telecoms set the rules, not the market.

Posted by: Adam Piontek on April 25, 2006 at 11:26 PM | PERMALINK

Now that we spent a trillion on Iraq, everything looks cheap. Can't we just go back to shooting lawyers?

Posted by: Matt on April 25, 2006 at 11:27 PM | PERMALINK

Google, YouTube, etc. pay huge bills every month to these bandwidth providers. Right now, if you could take away the bandwidth costs, YouTube could make money, and probably dozens of other services.

Without net neutrality, their ability to be profitable would go down even farther, especially if they try to compete with any services provided by an ISP. The real problem is that the telco will sit between the end user and the content provider and disallow services if the telco feels like it.

Until the option for a lot of different service providers exist (and no place I've ever been has offered more than 2 providers realistically, with service at roughly the same price), this is just carte blanche for the telcos to act as a monopoly.

Also you could probably do some VOD right now with peer to peer sharing, it's the politics of that market that are a bigger barrier imho.

Posted by: Quillian on April 25, 2006 at 11:47 PM | PERMALINK

Were talking about basing our information economy on an investment that is maybe $20 billion, for Fiber to the Home.

Given 100 million homes, that's only $200 per home. I just don't believe that is a realistic number. Heck, you pay a $100 service charge just to get the phone company to activate a phone line that already exists.

Posted by: John on April 25, 2006 at 11:53 PM | PERMALINK

So Basically they want people to but the VOD system that you see across the country in Hotel Rooms?

Cheesy Sleazy. Sounds, to me, like a new angle on whats nothing more than 1's and 0's...Sheesh.
Lets Call it SupraModern Spintronic Holographic Electron Beam Magnetic Control System and Hollyweird Enricher.
Rehashed Electronics under new marketing. Streaming Internet Video for your TV!!
[[Oh Wait My TV already Streams Video Signals!! YAY Oh YAY.]]

Gasket Blown. Later.

Posted by: Mach Tuck on April 25, 2006 at 11:54 PM | PERMALINK

Kevin:

You need to do another post to explain this. This is not about charging end-consumers more for bandwidth - that happens how. It's about teleco's having the right to charge web providers premiums to guarantee faster delivery.

So - ads get downloaded real fast - and Kevin's blog gets downloaded real slow.

Posted by: pebird on April 26, 2006 at 12:09 AM | PERMALINK

Point 3 of the arguments against net neutrality is a statement that experts are in favor of net neutrality combined with a statement that you don't believe them. How on earth does that count as an argument in favor?

Also, you might want to consider not using the phrase "white slavery" in the future. Why can't they just sell their mothers into slavery (or sex slavery, if that's what you mean)?

Posted by: KCinDC on April 26, 2006 at 12:18 AM | PERMALINK

Here is one possible scenario I can imagine where some version of tierd might make sense.

Verizon runs fiber to my house that can carry 1G. However my cheap contract with Verizon caps this at 1M.

At the Disney site, I pay $5 to watch the HD version of their latest blockbuster. Disney contacts Verizon, pays them $1 to up the bandwidth of my connection for the stream from Disney and maybe tweaks things so there will be no glitches in the stream.

As far as I'm concerned, I'm only dealing with Disney and if I have problems it is Disney's fault. It is Disneys responsibility to make sure Verizon delivers it correctly.

However I sure that most of the above could be done in a neutral manner.

Posted by: MonkeyBoy on April 26, 2006 at 12:32 AM | PERMALINK

For a really good history of the Telcom industry, mergers and more, for the last 20+ years, see:

http://www.niemanwatchdog.org/index.cfm?fuseaction=ask_this.view&askthisid=00196

Given their history, there is little reason to trust that the Telcom companies will do anything for the benefit of consumers.

Posted by: Mark on April 26, 2006 at 12:35 AM | PERMALINK

Al repeats the idea that Google et al are "stealing bandwith". How? Are they not paying an ISP to or outlet to connect to the internet? Is TimeWarner stealing bandwith from SBC when I use TWC to connect to a shop in another part of the country? Or for that matter another part of the world? Where does this alleged theft come in?

And it gets worse if you add in the potential effect on political speech. Say the owners of XYZ Telcom decides that it will show its support for the Republican party by deciding to shut down traffic between sites such as Washington Monthly and it's readers. I'm still paying my ISP and presumably so is WM, but I cannot access this site. This would be entirely legal under this proposed legislation, and would not be considered an "in kind" contribution to the Republican party as you are not a political party, only a news and commentary website.

Sorry but this all sounds too much like "you wouldn't want some of your packets to go 'missing' now wouldya? Could happen if youse don't pay up."

Posted by: clyde on April 26, 2006 at 12:49 AM | PERMALINK

If in doubt about the capacities of fiber optics, or it's potential uses, this article published in the Salt Lake Tribune in 2003 with regards to a local intitiative explains a lot, es[ecially with regards to the technology involved. $20 billion is not really that large of an investment for something that will last as long a fiber will.

Posted by: mattH on April 26, 2006 at 12:53 AM | PERMALINK

I have to agree with Kevin that so much of this net neutrality argument is a war of technical religions.

What so many people miss is that the whole direction of regulation was decided by the 1994 elections. Once the Democrats lost the house, the die was cast for the 1996 Telecom bill which very much favored the RBOCs. The reason for it is simple. The RBOCs owned the customer connection plain and simple. Long Distance was an artificial business, created by ATT to charge high rates for calls to subsidize local service. The 1996 Telecom act exposed it, by creating a path for the RBOCs to go into that business. The long distance carriers couldn't afford to build out a competitive local structure.

VOIP providers like Vonage are also artifical businesses that operate on someone else's infrastructure. There are virtually no barriers of entry to the business, and I will be quite surprised to see them stay a separate entity.

So how does the regulatory environment work with ensuring the survival of artificial businesses? It doesn't. The FCC is powerless against the relentless legal attacks from the common carriers.

As to the fantasies about packet inspection and routing based on content, its a bit extreme, since it would involve enormous overhead to do more than cursory routing based on address, or basic packet information. Besides, there is still a lot of dark fiber in the ground from the massive 1990s build out. The possibility exists that if the incumbents get too greedy, that someone could start buying a lot of bandwidth from the next tier of carriers.

The other entities that own a connection to the customer are the cable companies. These folks are still mostly true monopolists, and are by far the biggest price gougers on the internet.

People in the industry have heard talk about tiered internet backbone structures for close to a decade now. It hasn't happened yet.

Posted by: RickG on April 26, 2006 at 1:00 AM | PERMALINK

Lets's try an analogy:

Let's say you own a nice, well-kept, little Internet store.

Because you have worked hard and played by the rules to build up your shop over the years, you enjoy a steady stream of traffic and have managed to accumulate a small but growing number of happy, regular customers. You employ a nice, young man named Vito to update inventory, and business is generally good.

Then one day, a big, burly man in a white suit and fedora strolls confidently into your store. He walks up to the counter and introduces himself.

"My name is Don Telco. How's business?"

"Ok I guess," you shrug.

"You know this is my neighborhood," the Don declares. "You should show me some respect. You should let me wet my beak a little."

"What do you want?" you ask warily.

"Give me $200 a month for your own protection," the Don replies, "and I'll forget the insult."

"But I don't need your protection," you counter, "this is a nice neighborhood".

Don Telco chuckles. "You will paisan. You will."

more...

Posted by: Night Owl on April 26, 2006 at 1:07 AM | PERMALINK

Let's put all the roadways into a few private hands, too. Won't THAT feel great!!

Posted by: ferd on April 26, 2006 at 1:09 AM | PERMALINK
Video-on-Demand is a market I know a little bit about (at least, I did back when I worked for a startup VOD company), and the bandwidth and service issues that face commercial VOD rollouts are quite real. You and I may or may not care about VOD, but a lot of people do, and when telecom companies say that they need to make substantial investments to support large-scale VOD, they're right. When they further say that these investments will only be worthwhile if they can guarantee VOD service that works reliably and well, they're right again.

The question, of course, is whether the only way to provide reliable VOD service over the internet is to offer a tiered service to video providers. I don't know the answer to that, but it's not transparently absurd to think that the answer might be yes.

The big problem with this line of argument is that it assumes that the only way to deliver VOD service is over the internet.

Which, of course, is transparently false. Why does the current functionality and openness of the internet need to be sacrificed so that telcos can use it as a platform for a service already available through different systems?

If they're freed from net neutrality rules, long haul carriers like AT&T and Verizon will have a big incentive to degrade the service of internet phone (VoIP) suppliers like Vonage, since Vonage is a direct competitor for both consumer and commercial telephone service. In fact, Vonage has faced discrimination in the past and suspects that it faces ongoing discrimination in several current cases. Thus, VoIP companies like Vonage have the most to lose from the demise of net neutrality.

But here's the thing: Last year Vonage said it was satisfied (though not thrilled) with the net neutrality provisions in the Barton-Rush bill that's currently working its way through Congress. The bill has been modified since then, but as near as I can tell Vonage hasn't lobbied against it. When their CEO testified on the bill a few weeks ago, the only subject he brought up was 911 services. He didn't even mention net neutrality.

So: if Vonage is satisfied, maybe the bill isn't all that bad?

Its seems to be a reasonable indication that isn't bad for Vonage (assuming that Vonage's leadership is competent and that their public testimony matches their private lobbying, which are probably reasonable, but by no means certain, assumptions.) I don't see why that is an indication in the slightest that the bill isn't bad for end users, or for content/service providers that are not Vonage.

There are technical reasons to prefer a packet neutral internet architecture, and Henry Farrell outlines one of them here. The problem is that these arguments are very subtle, to the point where they become nearly religious in nature.

That makes no sense. "This argument is technical and subtle so that I would need to learn more to evaluate it" does not equate to "this argument is religious", though apparently you share the trait of being unable to discern the rather enormous differences between those two with the ID nutballs claiming that the arguments for evolution are at best on par with those for various forms of creationism and should not be taught without them.

Technical arguments you are not equipped to evaluate are not "religious".

Will the Barton-Rush bill doom the internet we know and love? Well, back in the early 90s I remember all the wailing and gnashing of teeth about how allowing commercial access to the internet would ruin everyone's favorite sandbox, but guess what? That experiment turned out pretty well. Bottom line: network purists are constantly arguing that the sky will fall if some proposal they dislike is adopted. I'd take them with a grain of salt.

This argument is ludicrous. First, it misrepresents the arguments, which were far more specific than were presented here, and basically predicted exactly the kinds of pollution of intrusive advertisement, spam, and competing commercial standards that have, in fact, occurred. Generalizing them to merely "ruining the sandbox" so that you can pretend they were generally wrong is either ignorant or dishonest.

Second, and more importantly, "Some people that I will label as 'network purists' made an argument that one change would 'destroy the internet', and were wrong. Other people, that I also label as 'network purists', now make an argument that a different change will 'destroy the internet'. The recent argument should be dismissed because the first one was incorrect."

This leverages the fallacies of equivocation (the 'network purists', while perhaps overlapping, are not the same group), and a form of ad hominem (the 'network purists' were wrong once, and thus their current argument is especially suspect.)

The key issue in the Barton-Rush bill is adjudication vs. rulemaking. I'm sure everyone else arguing about this issue is an expert in regulatory law, but I'm not and I can't immediately tell how big a deal this is.

Basically, the argument is whether Congress should mandate some kind of net neutrality regime and task the FCC with making rules to implement it, or whether they should set out general principles, let things unfold, and allow the FCC to adjudicate complaints if and when they're submitted. Rules have the virtue of being proactive, but also have the potential to hammer something into place that will turn out not to make sense. Adjudication is more flexible, but it's also a lot slower. It allows telcos to stretch Barton-Rush's net neutrality principles far enough to (possibly) put competitors out of business, safe in the knowledge that it will take years for the FCC to tell them to cease and desist.

Wait, this is an argument for Barton-Rush? Because, as I read it, it is a pretty strong argument for why one should be suspicious of Barton-Rush.

Net neutrality is essentially a "common carrier" requirement, and history has shown that common carrier rules are good things. Take railroads, for example. In the past, before common carrier requirements were adopted, railroads discriminated with abandon, shutting out startups in return for kickbacks from big companies like Standard Oil. This was bad for startups and bad for the economy, and there's no special reason to think that human nature has become more altruistic in the past century. If phone companies aren't required to act as common carriers, there's every reason to think they'll eventually throttle innovation in return for lucrative kickbacks the same way that railroads did in the 19th century.

Perhaps more relevantly to the players here, with the rise of VoIP, without net neutrality, the "common carrier" regulation of phone service will be defeated.

If we did this, he says, the whole issue disappears. When every home has cheap 100 Mbps access to the internet, services like VOD are a no-brainer. They can travel over a neutral internet with no special tiering at all. And $20 billion is a tiny price to guarantee equal access for all to the "public space...where democracy will be defined."

A public universal high-speed infrastructure -- like the public highway system, but for data -- would be a bargain at several multiples of that price, and a real platform for innovation.

Finally, here's probably the most convincing argument in favor of net neutrality: the telecom industry is against it.

Wow, lots of tribalism here: its an argument against net neutrality that 'net purists' want it, and an "the most convincing argument" for net neutrality that telcos oppose it. Clearly, Kevin, clearly getting the right group identity is high on your agenda ('Net purists' bad! Must reflexively oppose! Oops, but telcos worse! Must really reflexively oppose!), but, really, couldn't you at least pretend to be more focussed on substance?

Conversely, if the federal government subsidized the whole thing at the cost of a few billion dollars a year, just as they did with the interstate highway system half a century ago, we could build an internet backbone that would be cheap, universal, public, and a huge boost to the American economy. The feds don't even have to own it, they just have to pay for it.

Well, if the feds are going to pay for it, it probably shouldn't be privately owned, though the highway system might be a good model (the feds subsidize the infrastructure, particularly new high-capacity backbone connections, but state and local government pays two, controls most of the local topology, and actually ends up owning and being responsible for most of the infrastructure.)

Posted by: cmdicely on April 26, 2006 at 1:12 AM | PERMALINK

Al repeats the idea that Google et al are "stealing bandwith". How? Are they not paying an ISP to or outlet to connect to the internet?

The idea that they are "stealing bandwidth" by not paying for every hop their data makes between whoever they are paying and the customer (of course, that seems odd, as you'd assume that the different backbone providers are charging each other for bandwidth just like they charge lower-on-the-food-chain users, so this sounds like trying to get paid twice for carrying the same data -- or, more likely, looking for an excuse to filter data based on source to destroy competition.)

While Google is very pro-net-neutrality, they are also working, apparently, on building their own pipes so that, if it comes down to that, they can own virtually all the hops from their servers to the customer, and get their service through even if the telcos win, without paying protection money.
Of course, not everyone is Google.

Posted by: cmdicely on April 26, 2006 at 1:20 AM | PERMALINK

The town next to ours has its own cable TV company and it's own power plant and grid, And they get great service at much lower rates. But I don't envy them one itty bit 'cause I LOVE to write that big check to Verizon. "Watch me draw some more zeeroes with my markin' pen, Shirl!!!" LOVE IT!

Posted by: ferd on April 26, 2006 at 1:26 AM | PERMALINK

According to Google maps, I am 3.1 miles from the building where Mosaic was written, and there is no DSL service available from my telco(yes, we are in town). Color me unimpressed by the telcos.

And if I'm not mistaken, they have already billed us a couple of times for advanced services that were never delivered.

Also, did Keven just make an argument for structural separation? Wow, now's he's really among the radicals (like me). Good luck trying to get the wireline duopoly to go along with being just another service provider among the many options available on the fiber to your house. They'll bury congress in monetary free speech to avoid that.

Posted by: dglynn on April 26, 2006 at 1:36 AM | PERMALINK

More or less repeating my post from the last thread, I encourage Kevin (and everyone else) to read Farhad Manjoo's article on this subject in Salon, The Corporate Toll on the Internet. Please pay special attention to pages three and four.

Excerpt:

*********

The notion that the Internet shouldn't perform special functions on network data is known as the "end-to-end principle." The idea, first outlined by computer scientists Jerome Saltzer, David Clark, and David Reed in 1984, is widely seen as a key to the network's success. It is precisely because the Internet doesn't have any intelligence of its own that it's been so useful for so many different kinds of things; the network works consistently and evenly for everyone, and, therefore, everyone is free to add their own brand of intelligence to it.

...

Gary Bachula, vice president for external affairs of Internet2, a nonprofit project by universities and corporations to build an extremely fast and large network, argues that managing online traffic just doesn't work very well. At the February Senate hearing, he testified that when Internet2 began setting up its large network, called Abilene, "our engineers started with the assumption that we should find technical ways of prioritizing certain kinds of bits, such as streaming video, or video conferencing, in order to assure that they arrive without delay. As it developed, though, all of our research and practical experience supported the conclusion that it was far more cost effective to simply provide more bandwidth. With enough bandwidth in the network, there is no congestion and video bits do not need preferential treatment."

Today, Bachula continued, "our Abilene network does not give preferential treatment to anyone's bits, but our users routinely experiment with streaming HDTV, hold thousands of high-quality two-way videoconferences simultaneously, and transfer huge files of scientific data around the globe without loss of packets."

Not only is adding intelligence to a network not very useful, Bachula pointed out, it's not very cheap. A system that splits data into various lanes of traffic requires expensive equipment, both within the network and at people's homes. Right now, broadband companies are spending a great deal on things like set-top boxes, phone routers and other equipment for their advanced services. "Simple is cheaper," Bachula said. "Complex is costly" -- a cost that may well be passed on to customers.

*********

There you have it. The reason the net works as well as it does is because it's architecture is simple. Making a "smart network" to deliver prioritized content is complex, leads to additional problems and is more costly than just providing more bandwidth.

Simple is cheaper and it works better. Keep the net neutral.

Posted by: The Ox on April 26, 2006 at 3:22 AM | PERMALINK

A T & T's development of a prioritized/tiered network would provide technology that it could readily sell to the People's Republic of China. The telcos, the totalitarians, and the National Security Agency would all benefit from a collaborative effort.

Posted by: Lloyd on April 26, 2006 at 7:40 AM | PERMALINK

I'm with cmdicely: Kevin, there's nothing subtle about the kind of argument Henry makes. Of course, I'm biased, being a professor who teaches networking classes, but it's a very real and fairly straightforward issue, and one that we've seen real examples of over the last half-dozen years with the rise of things like peer-to-peer and streaming media.

You describe it as a "religious war", but do you know any network geeks who take the opposite side? Is there any division of opinion in the networking community about how congestion control really works on the Internet today and about the effects of defectors? I believe there's enough published literature on models and simulations to say we know the basics of how it works.

Posted by: Tom Hudson on April 26, 2006 at 7:41 AM | PERMALINK

I don't understand the VOD argument, and I too have some experience in this industry. The biggest bandwidth issue for a large streaming operation is not the existence of bottlenecks between the server and the client (though that is a problem,) but the cost of high-volume bandwidth that is multihomed for reliability. A tiered service is only going to exacerbate that cost problem.

Posted by: mj on April 26, 2006 at 8:34 AM | PERMALINK

Kevin, on point one, you should read this Robert Cringley column on the subject. You can do video on demand without a tiered internet if you use p2p techniques. To wit: http://www.pbs.org/cringely/pulpit/pulpit20060302.html

Posted by: Tom Haviland on April 26, 2006 at 8:53 AM | PERMALINK


KEVIN DRUM: here are some arguments against strict net neutrality and in favor of allowing tiered internet service:
We already have tiered service in the form of bandwidth levels, often identified with the typical market labels: standard, silver, gold. What's more, not all internet content is free. Pay sites are a form of tiered service. If providers want to profit from this, let them create content.
KEVIN: if Vonage is satisfied, maybe the bill isn't all that bad?
You tell us Vonage being happy with the bill makes it good and then later say that ATT being happy makes it bad. The Vonage Corporation is not some kind of citizen advocate looking out for the public's best interest. That's the government's job. For all we know, Vonage has already been paid off to keep quiet or contracted exclusivity agreements with telcos. Of course, the congressional members supporting this bill may also already have been promised lucrative deals.
KEVIN: Well, back in the early 90s I remember all the wailing and gnashing of teeth about how allowing commercial access to the internet would ruin everyone's favorite sandbox, but guess what? That experiment turned out pretty well.
There are millions of people whose computers are near useless or glacially slow even with high-speed connections because they are ridden with various viral marketing applications, web bugs, pop-ups, spambots, and auto-downloaded unwanted programs and shortcuts. Homepages are hijacked, screensavers are installed sans permission, and system drivers are corrupted to the point where favorite programs no longer function. This is how that experiment turned out. Hours of wasted time is the frustrating result. Those with the barest minimum of tech know-how are called upon repeatedly by family and friends to serve as an unpaid guru fighting to sift out the crap corporations hide in our sandboxes.


Posted by: jayarbee on April 26, 2006 at 8:59 AM | PERMALINK

When I was in high school, back in the early 80s, I had a classmate who had transferred in from Norway with his family. I recall his parents' amazement at the fact that they ordered telephone service in the morning, and it was installed and working that very afternoon. Apparently this was in great contrast to things back in Norway, where it took weeks for an order to be fulfilled.

America used to have had a telecommunications system that was a justifiable source of national pride, and that was easily the most advanced in the world.

Not any more.

While I think we should resist exaggerating, it has to be said that among rich countries, the telecommunications system of the United States is at best middle of the pack. The mobile telephony products and services in the US are decidely inferior to what's available in the Pacific Rim, and throughout much of Europe. Broadband service is expensive in the states, and slow, and broadband penetration rates are lower than in many economies.

So, it looks to my eyes like what we're discussing is a call for a subsidy from that part of the digital economy that is innovative and successful (Google, Ebay, etc.) to the part that is failing us (the telcos). Or am I missing something?

Posted by: P.B. Almeida on April 26, 2006 at 9:22 AM | PERMALINK

And another thing — if this is all about "Video On Demand" — er, well, I'm quite happy to wait until we actually have increased bandwidth to deal with it. I don't want the costly kludge of a "smart" network that favors some bits (the ones that make double money for the telcos) over others (the pesky bits that only make them money once).

When we have enough bandwidth to sustain VOD, then we'll have internet VOD. In the meantime, it's not like we have a right to VOD on the internet, or some sort of pressing need for it...

Posted by: Adam Piontek on April 26, 2006 at 9:27 AM | PERMALINK

> I was just thinking that I am a bit safter
> than most people at least, a little, because
> my ISP is a municipal wireless company. It's
> basically the local government so it has to
> adhere to certain state regs etc

You are aware that the telcos have various state and federal bills pending to outlaw municipal ownership of wireless and indeed any Internet/data services? In fact I think they already got one passed in my state.

Which I think answers the question of what would happen if a non-neutral Internet were allowed.

Cranky

Posted by: Cranky Observer on April 26, 2006 at 9:34 AM | PERMALINK

I find it interesting that Kevin starts with the telcos' position: video on demand is the answer.

But oddly enough, I never heard what the question was. Who exactly is clamouring for video on demand? Something wrong with Blockbuster, or the DVD machines at McDonalds (now $1/night)? I know some real movie junkies, and I don't know any of them who are anxious to start paying what amounts to viewing taxes to telcos.

I realize that the "content" industry rip-off rent collectiong specialists are all hot to get this running, but again I don't know any "consumers" (funny how we have been converted from Citizens to Consumers) who are anxious to pay more money to the guys in NYC, LA, and Nashville who are still charging them $17.95 for a CD.

Cranky

Posted by: Cranky Observer on April 26, 2006 at 9:38 AM | PERMALINK

"In the free market....customers would go to another service..."

It's not a free market and most customers have no other service to go to. I've been writing about this on my own blog for months and I've found the ultimate weapons against free market cultists, the FCC's own statistics, which Vinton Cerf quoted to the relevant senate committee.

Currently 47% of U.S. subscribers have fewer than two broadband providers to choose from; of that number, 19% have no broadband provider at all. This accounts for cable and DSL providers. Alternative broadband providers have 1.3% of the market, a *decrease* from two years previous. Most of the remaining 53% of subscribers have exactly two providers in their area -- usually one is DSL and the other is cable.

"Consumers will just choose a different provider" does not apply if there are no other providers to choose from.

Posted by: Constance Reader on April 26, 2006 at 10:13 AM | PERMALINK

Among the ironies here is that, even if you insist on delivering Video on Demand over the internet, only the stupidest, least market-viable implementations of VOD actually require the kind of tiered service structure the telcos are pretending to talk about. (I say "pretending" because even with such a tiered system, that simpleminded version of VOD would fail on a regular basis whenever there was enough demand.)

Any even slightly smarter VOD system with a few bucks worth of storage on the customer end can do fine with broadband as it is now. But that would mean telcos sharing their profits.

Posted by: paul on April 26, 2006 at 10:19 AM | PERMALINK

Wait, what's this Google stealing bandwidth. I'm using Google and I'm paying through the teeth for a decently fast DSL connection to go out on the Internet to secure services I need. I'm paying for the bandwith to get my Googles.

I've noticed that every time the telecoms get more power, I pay more and get less.

Net neutrality for me.

Posted by: Old Professor on April 26, 2006 at 10:24 AM | PERMALINK

The public infrastructure investment concept could be a winner in upcoming election cycles ... whether the funds are provided directly, via tax rebates or credits of some kind.
Broadband to every home.
SF is about to get universal wifi, and it is universally popular
Broadband is cheap democracy

Posted by: Jim on April 26, 2006 at 10:38 AM | PERMALINK

No such thing as a free lunch. Just make sure no one has a monopoly. Capitalism is self regulating.

Posted by: Bill on April 26, 2006 at 11:01 AM | PERMALINK

Well all lose!
Coverage

Posted by: Service on April 26, 2006 at 11:07 AM | PERMALINK

This post doesn't address the real net neutrality problem. Packet type neutrality is not what people need to worry about (e.g. favoring VOD over, say, FTP because VOD needs real-time quality of service while file transfer doesn't). The problem is origination neutrality, where backbone providers favor or disfavor the company which provides a certain service.

Posted by: cafl on April 26, 2006 at 11:30 AM | PERMALINK
I was just thinking that I am a bit safter than most people at least, a little, because my ISP is a municipal wireless company.

Your ISP doesn't really matter all that much if the telcos that own the pipes between your content provider and your ISP are dropping the packets containing the content you want.

Posted by: cmdicely on April 26, 2006 at 11:35 AM | PERMALINK

Perhaps more relevantly to the players here, with the rise of VoIP, without net neutrality, the "common carrier" regulation of phone service will be defeated.

This is something I haven't heard brought up before, but it is important. If VoIP catches on, and carriers are allowed to discriminate based on the source of the connection, we could (probably would) have a situation where you would get a good phone connection only if you were calling someone with the same provider you use. Right now, you take for granted that you will have the same level of service, no matter who you call. If you like that, you should be in favor of net neutrality.

Posted by: sc on April 26, 2006 at 12:12 PM | PERMALINK

SC: Carriers have already tried to discriminate against subscribers who use Vonage's VoIP service by blocking their ports. It's the Madison River Communications case, and to the FCC's credit, they slapped Madison down for it.

Posted by: Constance Reader on April 26, 2006 at 12:20 PM | PERMALINK

Suppose that some clever entrepreneur were to build a super-duper Wi-Fi set, costing perhaps a thousand dollars, with an electronically steerable phased-array antennae. Let us say that such a set, when mounted on a rooftop or a small tower, might attain an angular resolution of two or three degrees, and a range of a couple of miles under urban conditions. Given this long range, the system would work initially with an extremely low density of users, so they could be the kind of people who put solar cells on their roofs. These super-duper Wi-Fi sets can share their long-distance connections with ordinary cheap Wi-Fi sets. Hooking all these Wi-Fi sets together into a volunteer mesh network would sidestep nearly all legal restrictions. There would be no government subsidy, simply a large number of individuals exercising their individual rights. In short, it would work very much the same way open-source software works. In places like Southern California, the mesh might run continuously for a hundred miles or more. Google would find it very easy to connect up with such networks, on a non-exclusive basis, but so, of course, would all the other internet companies and telecoms providers. The telephone companies might very well wind up in a much worse position than if they had stuck to net neutrality.

http://en.wikipedia.org/wiki/Wi-Fi
http://www.firstmilesolutions.com/
http://www.wifiworldrecord.com/
http://en.wikipedia.org/wiki/Wireless_mesh_network
http://pdos.csail.mit.edu/roofnet/doku.php
http://www.skypilot.com/technology/advancedantenna.htm

Posted by: Andrew D. Todd on April 26, 2006 at 12:39 PM | PERMALINK

Nice post. However...

There are two technical issues here that are being mixed. E.g.:

  • VoIP depends on guaranteed low latency and guaranteed but low bandwidth; that is, bidirectional interactive realtime.
  • VOD needs high bandwidth guaranteed over longer periods but doesn't care as much about latency; that is, unidirectional streaming, non-realtime for archive content or near-realtime for live content.
  • Typical web browsing has no dependence on latency or bandwidth, although the quality of the experience obviously improves as latency decreases and bandwidth increases.
  • Other applications, such as videoconferencing, will require both low latency and high bandwidth, but those aren't widespread yet.
As far as carrier's arguments that they need additional tolls to make the infrastructure investments required, I'm extremely sceptical. The carriers are presenting a false dilemma, and there is no reason to fall for it.

The appropriate question to ask is: Is there a way to ensure net neutrality while still providing QoS guarantees, and a means of paying providers (whether source or destination) for those guarantees?

The answer is "yes". It is feasible to have the consumer, directly or indirectly, choose and pay for a given QoS, and have the consumer pay for it.

However, to be effective, that also requires that there be no competitive chokepoints; that consumers and providers have a competitive choice, especially among "last mile" providers. Unfortunately, the number of options appears to be decreasing.

Posted by: has407 on April 26, 2006 at 1:32 PM | PERMALINK

I was just thinking that I am a bit safter than most people at least, a little, because my ISP is a municipal wireless company. It's basically the local government so it has to adhere to certain state regs etc.

Posted by: MNPundit

That's fine as long as the sites you visit on the web use the same ISP, but otherwise you will have the same problem everybody else has.

Um, "white slavery"?

Posted by: bc

"White slavery" typically means selling a woman into prostitution.

Posted by: anandine on April 26, 2006 at 1:43 PM | PERMALINK

If Japan can do it, why can't we?

The answer can be found by looking at the difference in the structure of the value-add carriers/ISPs vs. the common-carriers or RBOCs. The Japanese structure made the RBOCs relatively weaker; the US structure made them relatively stronger.

There are also only two major RBOCs in Japan (NTT East and NTT West), both owned by NTT Holdings, which simplified things; and has allowed NTT Holdings to tradeoff declining RBOC revenue with other value-add services from other holding companies (e.g., NTT Communications). The cultural differences that allowed that to happen in Japan, and not happen in the US, are also significant.

The tension between cable providers and the RBOCs in the US also provided an excuse for the US RBOCs to initially get out from under the leash. And we heard the same "we need to be able to offer value-add services to make the infrastructure investment needed and stay competitive". The .com bust subsequently had a magnifying effect in the US.

In short, to fix the problem, you need to go back and make the RBOCs what they should have remained: a regulated monopoly who owns wires and little more; same for the cable companies. (When and if technology eliminates the competitive chokepoints--in particular the "last mile"-- then, and only then, do you remove those constraints.) Unfortunately at the moment we have a screwed up mish-mash that is the worst of both worlds.

Posted by: has407 on April 26, 2006 at 3:04 PM | PERMALINK

I should also mention that the timing of deregulation and the breakup of NTT in Japan was also significant vs. the US. That happened much later in Japan, and at the that time there was significant visibility into, and thought devoted to, the impact of the Internet. (I'm not as familiar with France's trajectory, but I would bet it's similar. They also had previous experience with an Internet-lite-like capability, if anyone remembers "Minitel" of decades past.)

Posted by: has407 on April 26, 2006 at 3:17 PM | PERMALINK

A few billion spent by the USA gov't to benefit (potentially) all citizens for decades to come ... hmm, let's see .. versus one month, or maybe even two months of the Iraqi occupation and suppression of dissent / "insurgents" fees.

Gee, I wonder what I would vote for ?

Posted by: Jon H. on April 26, 2006 at 11:59 PM | PERMALINK

SC, if it were economically sensible for companies to only allow phonecalls to those who have the same provider, we would see a similar pattern in other industries. But it isn't, so we don't (and it isn't allowed anyway). Look, if Internet providers wanted to blackout their users completely, they could do so. Of course, their customers will quickly leave. You wouldn't need the government to protect them; they would do it on their own.

Posted by: lessgov on April 27, 2006 at 12:28 AM | PERMALINK




 

 

Read Jonathan Rowe remembrance and articles
Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

Advertise in WM



buy from Amazon and
support the Monthly