Hacker News new | past | comments | ask | show | jobs | submit login

The "last mile" is a utility, and by its very nature will always result in monopolistic control.

It is impractical and irrational to wire up a home to multiple ISPs with their own fiber channels. Each channel could cost tens of thousands to install.

Instead the "last mile" should be a non-profit funded by ISPs who lease the channels. The non-profit is responsible for installing the "last mile," upgrading it, and repairing defects in it. The ISP is responsible for everything that happens from the exchange upon up (inc. pricing, support, peerage agreements, interconnects, etc).

This way you can still choose Comcast if you wish but may also have several local ISPs competing for your business. When you choose to change no engineer needs to come to your house, they just plug and unplug you at the exchange.




So an open-access network (https://en.wikipedia.org/wiki/Open-access_network).

This is done quite successfully in other countries around the world, like New Zealand where we mainly have one infrastructure provider (selected and monitored by the Government), who must offer fair and standard pricing to any ISP who wants to add value to the network.

As a result, we have many ISP options throughout the country and competition between providers is high.


What value do the ISPs actually add to the market?

It seems like a government granted middleman position to me, where I'd rather just purchase service from the government or the government's selected infrastructure manager.

Presumably there's already some base price for service that this infrastructure provider is charging everyone (a controller monopolist), what are you gaining on top of that by having a second middleman that is presumably also making a profit?


> What value do the ISPs actually add to the market?

They're the main point of contact for the end customers; as a typical customer, you only interact with Chorus (the public company that provides the fibre/copper in most of NZ) when they send technicians out to work on the connection upstream from your house.

So, the ISP handles billing, tech support, DNS, email, liaising with Chorus, etc. Different ISPs offer different service tiers, billing arrangements, options for buying/leasing the modem, etc.

Our residential electricity works similarly, which leads to retail companies offering both - for example I get one bill each month to pay for both fibre and electricity.


Given the government's track record with some big projects like healthcare.gov which is still a nightmare to use, I would not trust them to be an ISP. In fact, in the past, some states did (or still do) have their own ISP that they use at a lot of schools and public facilities. Ours was always slow with frequent outages.

I think a hybrid approach with the government managing and maintaining the physical lines and allowing isps to plug in and pay rent is the way to go. The ISP rent should be enough to cover the cost of physical line maintenance.


What are your specific complaints about healthcare.gov? I was out of the country when it initially rolled out, but have had no problems with it the past 2 years since I've moved back and needed to get it.


Washington State exchange is notoriously terrible in terms of downtime, availability, and customer service. I won't go into what I dealt with for six months using their service vs. my previous low-cost insurer that left the market due to regulations, but it was an absolute nightmare for me and my family. Small business owners got absolutely screwed in our state.


packet transit, installation, maintenance, customer service. Also most less technical ISP customers get a lot of value out of ISP managed equipment rental and don't want to buy and maintain a router-AP themselves.

Local loop unbundling for phone and DSL service was extremely successful in the US, there's not a lot of reason to think a government bid process would have done better than CLECs did.


The most immediate one to come to mind is that a private ISP can't indict me.


I hear San Francisco is about to attempt this? https://futurism.com/san-francisco-has-approved-a-plan-for-c... Where "about to" means soon-ish?


> The non-profit is responsible for installing the "last mile," upgrading it, and repairing defects in it.

What incentive does a non-profit have to invest the billions of dollars required to do these things? If you use public dollars to do it: what do you think happens when investment into upgrading from GPON to NGPON2 is competing for taxpayer dollars with roads and schools? I can't imagine that anyone in say San Francisco would rather have SFMTA running their internet than Comcast.


1) You get the government you deserve. If you're worried it would be mismanaged, become active.

2) I would absolutely rather have an unbiased and fair agency installing utilities than a rent seeking public company that is beholden to shareholder value.


You get the government your neighbors want. I'd vote in a heartbeat to raise water and sewer rates to upgrade our water systems so we could stop dumping untreated sewage into the Chesapeake Bay: https://www.baltimorebrew.com/2017/08/01/baltimore-released-.... But my neighbors won't do that. Do you think they'll be more forward looking when it comes to broadband? Or will all the retirees who dominate the voting decide that 50 mbps ought to be good enough for everybody?

Since my parents first got fiber 10 years ago, Verizon has spent a ton of money upgrading from BPON to GPON, and now is working on NGPON2. If a public utility were in charge, even if they were willing to raise the money to install fiber in the first place, there is no way in hell they'd have made those upgrades.

You can see this in practice. Here in Maryland, the only upgrade in transit service I can think of in my lifetime is running the Penn line from Baltimore on weekends. More typically, the public authorities run the transit systems on the edge of collapse. Simply maintaining existing service levels in the face of under-maintained infrastructure is considered a victory.


That has been tried in a few places, and I'm not sure that it has succeeded anywhere. Municipal fiber was laid where I used to live in Provo, UT and it had huge funding problems. ISPs could compete, but all the value is in the last mile not in the gateway.

Eventually Google bought it out, but the city had to increase taxes to help pay for the current bond, and was looking at more bonds to pay for it before Google stepped in.


It has definitely succeeded. For example, New Zealand uses this model, and we're actually in the middle of rolling out fiber to almost every house in the country.

Okay, so technically in New Zealand the last-mile infrastructure is owned by a utility company, not a non-profit. However, the utility companies are selected by the government and tightly regulated, so they act more like non-profits than typical companies. Also note that they aren't allowed to actually sell internet access to consumers; they only maintain the infrastructure.


When I lived there, I had 200mbps up and 200 down for $70 a month and it was great. This was before Google came to town. But since I didn't own property, I guess I was out of the loop on property taxes.


I had it as well at 100Mbs before Google fiber. The service itself worked great, but it had to be subsidized by taxpayers throughout its lifetime - it never broke even after the subscribers fees.


This is how it has worked for 15-20 years in France, the UK, and probably most of Europe.


> The "last mile" is a utility, and by its very nature will always result in monopolistic control.

This is THE central point. Everyone saying that they would be OK with the latest FCC regs if there were only more competition need to consider this. In what bizarro world would anyone actually want ISPs competing for the last mile?


The one where 5g is faster than my comcast connection.


> In what bizarro world would anyone actually want ISPs competing for the last mile

You can easily have up to 1 gbit symmetric internet connection in Russia for $10 because of that competition.

It's not rocket science, guys, US ISPs suck because your regulation killed all the competitors, and your solution is what, more regulation?


I have options of cable, DSL, and fiber. With each new ISP offering service to my house my rates have gotten cheaper and bandwidth faster.


If you last mile is minimally-maintained aging copper far from the central office, and the competition is fiber...


How does this compare to municipal electricity companies? Do you think the same model could be applied?


Coming in 2019, electric companies monitor your usage profile to tell when you turn on different appliances and charge you a higher rate for running your TV or charging your electric car. But don't worry, the free market will fix it.


In California you do get different rates if you get an electric car. You also get different rates (higher) if the State thinks you don't "need" air conditioning.


This is already a thing in several states and the general consensus is that it saves citizens money.


Is that because electric rates are heavily regulated as utilities, or because the free market let the only electric supplier do whatever they wanted to with your rate structure?


Isn't that already the case? You don't have two electrical lines running into your house.


That's the argument. The power utilities are tightly regulated.


You've just described Local Loop Unbundling (https://en.wikipedia.org/wiki/Local-loop_unbundling), this has been the way things have worked for 15-20 years in France, the UK, and probably most of Europe.


I think that would be a good value proposition for a city that does not already prefer one ISP over another. Interesting idea to spin something like this up.


Many states have laws that forbid this, along with any attempt at local control of the issue.


> Instead the "last mile" should be a non-profit funded by ISPs who lease the channels.

That's an interesting idea, kinda like Interac for telecom. From what I hear, Interac has a bit of an issue with being fairly closed to new peering agreements, so I wonder how that could be solved for the telco space, maybe constitute the maintenance corporation to be open to bids from new ISPs.


Instead the "last mile" should be a non-profit funded by ISPs who lease the channels. The non-profit is responsible for installing the "last mile," upgrading it, and repairing defects in it.

Yes, that's called "government".

We do that with roads and mail, we can do the same with communications.


Even if you avoid the last mile cluster F*, you still need to rent physical space to ISPs for setting up their network equipment. With multiple ISPs, its not clear who should "own" the space. I doubt individual homeowners want to deal with that headache.


How much money are we talking for a mile of fiber? $6K for materials and $12K for labor? Anything else? Could you do a neighborhood for $18K?


cars cost "tens of thousands" yet personal automobile ownership has been feasible and has a huge positive impact on the economy. If that 10-90K ballpark cost estimate is accurate then accessing the information superhighway could be a similar situation.


Right, but not being able to afford a car severely limits people's' access to jobs, care, entertainment, groceries, etc. And this would be worse; access to the Internet is starting to be assumed here in the US. If it's going to be it should be treated as a right.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: