The problem with Network Neutrality is that it's very easy to see the development of a monopoly/oligopoly that will have near complete control of the world's most valuable man-made resource.

When it comes to the corporate behavior, only the most cynical predictions have been at all reliable. Those sanguine about the result of absolute freedom-of-action by internet providers clearly have little sense of history.

God forbid that internet access become a relatively unprofitable commodity rather than an utterly complicated, multi-tiered service in which profit is extracted through the creation of an artificial shortage.

I don't understand why the Internet is not government controlled. Would you leave road maintenance to private companies? How about air traffic, or sewers or water? The Internet is as important, if not more important (in a Constitutional way) than these services, but we leave it in the hands of self-interested parties.

The difference between these and the internet is right of way and externalities. That is, the government is involved in roads and sewers because (1) you can't have a road network in most places without the ability to forcibly purchase small strips of land from lots of people and (2) until recently, there was no efficient way to charge usage fees in a reasonable way. The internet doesn't really have that problem: charging is quite simple and the flexibility of routing means we're not nearly so beholden to any one set of property owners.

Now, I think there might be a case made for making last mile internet either municipally run or a heavily regulated business in the same way that water/power/sewer service in cities is, but that's a very different argument.

Ideally, the last mile fiber would be divorced from the IP connectivity to the rest of the world, so that a residence would pay the same money to use that last mile fiber whether they're running 1 megabit or 1 gigabit. If you then had real competition for options from your central office to the rest of the world, you'd see performance for a given price increase dramatically.

This is basically how things are done in Sweden, at least outside the major cities. A municipality builds and runs fiber to every home, residents choose from a number of different ISPs and each ISP pays a flat rate to the municipality for each active subscriber they have. ISPs can then offer any type of service within reasonable limits. This means that I can get for example 100 Mbps service and VoIP for ~25 dollars a month, or 10 Mbps for 15.

I think that you're exactly wrong on this. It makes sense for government to build roads because, for one thing, it's physically infeasible to have multiple sets of road-builders competing with each other. There is finite land.

On the other hand, it's quite feasible to have many competing internet providers. I'd say we have too few, actually, and that government makes it worse. The airways should be more open and whatever infrastructure the public subsidizes (telephone wires, etc) should be equally available to all ISPs.

More competition is the key here, not a government monopoly. All we need the government for is ground rules to make the game fair.

It certainly is more so in some countries like China and North Korea and look how that's going. Government's aren't always benevolent, they have their own agendas too that may not match some people's expectations of online freedom for one thing, and it's not like government's don't also increasingly have business interests in part influencing them. There are a lot of lobbyists these days...

Actually if it becomes unprofitable then there will be little incentive to invest in it and it will languish for years. I'm all for net neutrality but I'm also for corporate profitability. If not for the high margins FIOS and the like bring in, then there wouldn't be a decent internet option in my neck of the woods.

How many people have FIOS? Where I'm sitting our costs haven't changed in 5 years, and our bandwidth has held steady.

Where I was sitting last year, costs haven't changed in 8 years and bandwidth has held steady. I'd say those eight years are the norm, and FIOS is something of an exception.

By "relatively unprofitable" I meant "just profitable enough". When the a product becomes commoditized the market for it becomes (much more) efficient, and thus only minimally profitable, because the principal goal of the actors within it is cost reduction.

If internet access becomes a commodity it is _great_ for consumers (c.f. computer memory, public water, electricity, undifferentiated agricultural products) but horrible for the providers/producers.

Thus, somewhat ironically, the only incentives corporate lobbyists have to advocate for free-er markets is in order to maintain market inefficiencies or to reduce their supplier's margins.

I thought Network Neutrality was about restricting what businesses could do? In any other nation I would absolutely say the government should just own the internet. In the US I'm a bit more cautious but I agree that companies can't be given free reign.

Akamai was arguably paying Comcast for carrying their transit traffic. But they never should have been paying for traffic with a final destination in Comcast's network (peering traffic).

As a Tier 1 ISP, Level3 does not require Comcast to carry any transit traffic for it. If Level3 has to pay to deliver traffic that Comcast's customers themselves have requested, Comcast is clearly charging Level3 (and by extension, Netflix) for access to the customers on their network. If Comcast broke off its peering agreement with Level3, all that traffic would instead be going through a different transit provider. Comcast would then likely have to pay to receive that traffic.

Comcast is in a position to charge for access to their customers because they have a monopoly on access to those customers. Those customers may not even have other choices for broadband internet access, because infrastructure has such high barriers to entry.

Because the consumer broadband market has high barriers to entry,[1] there is little competition.[2] That lack of competition in infrastructure is now distorting the market for internet applications, possibly the closest thing to perfectly competitive market that has ever existed,[3] by raising both the barrier to entry and transaction costs.

Yes, regulation would distort the market. In a market that is tending towards monopoly or oligopoly, this is a good thing. Net neutrality is an ill-defined idea, but I think we can agree that we want the market for internet applications to remain as free of a market as possible.

[1] This is not the case under regulatory regimes that require line-sharing, where the consumer broadband market is decoupled from the infrastructure market. Line sharing is not required in the US.

[2] Infrastructure markets are natural monopolies <http://www.linfo.org/natural_monopoly.html>; due to economies of scale.

[3] A definition of perfect competition: http://en.wikipedia.org/wiki/Perfect_competition

I agree that either Akamai and Level 3 should both pay (for all traffic flowing into Comcast; whether it's "CDN" traffic or "regular" traffic makes no difference) or neither should pay. Since they're both saving Comcast money, I'm leaning towards neither. It's unfortunate for everyone else that Akamai apparently set this precedent.

The problem is that these complicated business relationships are affecting our access to necessary services. Would you tolerate your sewer being cut off because two private parties didn't want eachother's shit going through their pipes?

If this were just about a bunch of TV shows it would be different, but this is about the fundemental communications medium of our society. How can we so tightly regulate utilities and over-the-air television, but be so lax with something just as important?