The implications of the end of net neutrality

Isn't that what the new administration promises? Deregulation and ending net neutrality. I don't like monopolies either, but it's not like Google/Apple/Facebook/Amazon don't have them. They have all the patents in the world to bludgeon telecom operators to death if those operators tried to start their own competing social networks or search engines.

Your view of the world doesn't explain why Google doesn't build fiber to cities like Baltimore and LA that have been publicly disparaging their incumbent providers and begging someone to come build fiber and compete. Or why Google won't wire up Silicon Valley. Is it at all credible to suggest Comcast or whatever has more pull than Google in Google's own backyard?

I tend to agree with your overall point that regulated monopolies are the way to go for internet service. But that has its own set of challenges. The AT&T monopoly worked pretty damn well to tell the truth. They built what was at the time a gold-plated telephone network. It was also expensive for consumers, and the government was willing to let AT&T be profitable enough to make building worth its while. That's very different than how "utilities" are regulated today, where they deal with crumbling infrastructure because no politician wants to authorize raising water or sewer rates.

> remote/waldoed surgery needs superb data delivery guarantees

It isn't a question of bandwidth or latency, the problem is that life-critical systems need hardware-level redundancy that mass market networks just don't have. Some drunk crashes into the pole at the end of your street and your internet goes down in the middle of surgery because there is only one path.

This is why the hospitals that do this sort of thing have redundant dedicated physical connections on both ends. There is no other reliable way to do it. But network neutrality has nothing to do with that sort of dedicated physical point to point connection.

> providers of ad-heavy contents should be allowed to pay for the data their users consume

Nobody actually wants to do that. If ad providers actually cared about customer bandwidth charges they would spend more than the obviously-zero-if-not-somehow-negative amount of time they currently spend on optimizing their ads for resource consumption.

The entire thing is a sham. Nobody wants to subsidize data. The ISPs raise data rates but zero-rate their own offerings to push customers away from their competitors, then go back to their competitors with their hand out and demand money to reestablish the level playing field. Or worse, charge a toll to get on the high side of the unleveled playing field.

If this wasn't the case then there would be no need to involve the ISP in the transaction. If advertisers wanted to give away their profits to users then they could just pay the users directly. Some have tried this and given up, because it's ridiculous. Users would much rather have more/better journalism or services or whatever thing the ads are supporting than some tiny subsidy which is nothing more than a pittance at the level of an individual user.

The only reason ISPs are involved is that they can charge $PROHIBITIVE_PRICE for data to end customers but $MONOPOLY_PRICE to content producers and it's only by getting this "discount" that it makes any sense for the content producer to engage in the transaction. The only result compared to the situation where this is prohibited is a harmful transfer of resources from high-competition content producers to low-competition ISPs.

The cost of providing last mile internet service is dominated by fixed costs, meaning costs that are the same regardless of whether you serve 10 or 20 customers on a particular street, and regardless of how much data those customers transfer. These costs are not small; creating the capacity to serve a million households will cost north of a billion dollars.

There are some policies that can knock that cost down some. The best one is to require conduit to be installed in the road anywhere it doesn't exist when road service is done. We should enact this policy. It will save us a lot of money. But it's hard to get passed because the cost is in current-year tax dollars to install conduit and the benefit is in much lower long-term costs for power distribution and telecommunications.

And it "only" lowers costs, it doesn't change the competitive dynamic. The nature of the problem is this. If you have one telecommunications company then you have one piece of fiber in the ground, and when some idiot with a backhoe cuts the fiber, they need to send a truck out to fix it. If you have three competitors then there are three pieces of fiber, all next to each other, owned by different companies, all get cut at once, so you need three trucks. If you have ten competitors you need ten trucks. Most of the costs of operating last mile infrastructure share this horrible scalability.

Meanwhile if you have ten competitors then each competitor has 1/10th as many customers, so the cost per customer is ten times as high. This is why last mile telecommunications is called a natural monopoly -- having a single provider will give you half the cost of having even two.

So that is the source of the problem, how do we fix it? The answer is to isolate the natural monopoly to the fullest extent possible. Let it be a monopoly but let it be nothing else. Some entity (the government, a nonprofit, Comcast) owns the fiber in the ground but they do nothing else. They don't offer retail internet service, they don't offer TV service, their only job is to maintain the physical infrastructure and lease wholesale capacity to third party providers at uniform rates.

Then you have plenty of competition, because anyone can lease capacity and offer service, no one has any undue advantage over the others because the rates are uniform and published, and the monopoly provider can't give any advantage to its own over-the-top offerings because it doesn't have any.

If the boxes collocated at an ISP broke today, the data would flow to users from distant data centers. All that would happen is the ISP and the content provider might end up paying a little more in network costs.

The edge servers in the ISP network would only have the same non-network-neutrality effects if the ISP offered a string-and-soup-can connection to the internet that could not support the service.

The fear is - as it's always been - the ISP's ability to make a remote service effectively unavailable to their customers.

1) If you are against Net Neutrality, then you believe Comcast will be a good and objective manager of the internet.

2) If you believe nearly everything Comcast touches it fucks up and cannot be trusted to manage anything, let alone something as open and free as the internet, then you are for Net Neutrality.

Yes and no. The direct effect is on US consumers, but because of how much of the money made from internet content and online services is from US consumers, it likely has a very significant indirect effect on global competition.

A last mile telecommunications network to your house? Comcast is known to immediately increase speeds and slash prices whenever anyone does, but only in the specific region the competitor offers service, thereby ensuring that the new competitor is unprofitable and that all future competitors get the message not to do that.

So presumably indefinitely.

Ending net neutrality simply gives control of the internet to companies with currently established hardware monopolies.

Net neutrality is what's keeping the internet a free market currently.