Hacker News Reader

Comment by nickik | Hacker News Reader

Maybe you are a troll, or maybe you just don't understand how networks work. The relevant factor is not the overall size, but rather the population density. Lower population density means more cable per person (on avg.), so it makes it MORE expensive.

Sweden is not very dense, large parts are uninhabited forests.

Sweden 24 United States 35 New Jersey 467.2

You are basically saying, "I live in the most densely populated state in the US, and my Internet is almost as good as the internet of the Swedish country side, U.S.A! U.S.A!"

People from the German Telekom like to us this argument as well, "Germany has much more population, we can't do what they do". That argument has been debunk and shown for what it is many times, by multiple experts. It's simple lobbying by a monopoly provider of copper to defend its monopoly against the competition heavy fiber.

View on HN

> Sweden is not very dense, large parts are uninhabited forests.

If they're actually uninhabited, then there is no need to build fiber out there, and you shouldn't count them in the density measure. What you're really interested in is the population-weighted density (i.e. the density of the places where people live).

Compare the list of U.S. urban areas to that of Sweden: https://en.wikipedia.org/wiki/List_of_urban_areas_in_Sweden; https://en.wikipedia.org/wiki/List_of_United_States_urban_ar.... The vast majority of the population of Sweden lives in urban areas with between 2,000 to 3,500 people per square kilometer. No urban area in the U.S. is as dense as Stockholm's (3,500). NYC is at the low end of that range (2,000), and is less dense than a bunch of Swedish cities with just ~50-100k people.

Major U.S. urban areas like Atlanta, Boston, Charlotte, and St. Louis are under 900 people per square kilometer. Major cities like Houston, Dallas, and Philly are just north of 1,000. That's 1/3 the density of Stockholm.

Anecdotal evidence. I live in Stockholm. I talk at a conference about once/month, nearly always outside Sweden. Nearly every industrial conference has problems with wifi. Never in Sweden. Anywhere do we have problems at hotels, venues, massive conferences. We have the bandwidth. Most other countries do not, and leaving Sweden is always like going back in time, tbh. Everybody i know has fibre to the home, irrespective if they live in an appartment or suburban house. The suburban house people complain that it costs them 1-2k Euro to pay for the fibre, but they pay for it. The people who don't have fibre to the home are mostly old people (lots of them!) and farmers (and lots of them have it anyways).

> The suburban house people complain that it costs them 1-2k Euro to pay for the fibre, but they pay for it.

I think that's the difference - the costs are upfront, not paid by capital/loan with the subscriber having to pay for it eventually.

Rationally, most people wouldn't want to pay that much into a US cable/net provider, considering their historical experience with them.

I would totally pay 1-2k upfront (assuming I owned the home or was planing to stay at least 5 years) if I "owned" the fiber and could choose my own ISP.

At that level of competition, it could easily drive the prices down into the $10-20 range for high bandwidth, which would probably break even with current cable internet prices now.

I pay nearly $100/mo for 300Mbps/20Mbps since I have about 9 active users in my home. I pay about $1.2k per year in internet, so $1-2k upfront for cheaper monthly pricing and more competitive features.

The carrier owns the fiber, but you still have to pay $2500 upfront plus a $100-500 monthly fee. Most of the time you can choose the ISP though. If you live in a co-operative you can own the fiber yourself and buy transit. Then there's the municipal fiber carriers that will own the fiber, but you only pay the prime cost resulting in lower upfront and monthly fees, and if you live on the countryside you can get a government grant, install the fiber yourself, then sell it for 1$ to one of the carriers usually resulting in low monthly fee.

Is this in Sweden? In the U.S., if a carrier wants to build fiber in a town or suburb, they have to build all the fiber first, then hope people sign up for $70-$100 per month with no money paid up front.

What you describe sounds like a great model. It'd be very difficult to do it in the U.S.

Yes Sweden. The Carriers only own the last mile, then they rent fiber from the backbone. Some carriers have backbone fiber, but most of it is owned by government funded Tier 1 Carrier "Telia" that makes sure all providers and Carriers get equal access.

Maybe that's why there is so much more fiber deployment in Sweden than in the U.S.? The fact that people are willing to pay the infrastructure costs up front? No carrier would be able to get a franchise under those terms in the U.S.

What do carriers in Sweden do with people who can't afford to pay the up front fee?

EDIT: Looking at Stokab, it looks like they don't require an up-front payment. But Stokab built its network over two decades, starting out in the core city and expanding based on market demand: https://ec.europa.eu/digital-single-market/sites/digital-age.... That model is illegal in almost every city in the U.S. Any carrier must agree to wire up a whole city on a very short timeline, even areas where there isn't sufficient market demand. Verizon was given less than 8 years to wire up all of New York City.

You get an offer to pay it off in like three or more years. But if you can't you don't get any fiber. But you can often install afterwards if there is fiber in the area.

Stokab is a municipal and non-profit carrier and almost all cities have one of those, competing with the private carriers. The municipal often requires 70% + coverage, witch is very hard to get in a small non tech village with old people. Stokab is a bit different because they have about 1.5 million people living in their area witch is dense compared to the rest of Sweden.

The private carriers don't even bother if they can't get a few hundred subscribers in the area. So if you live in a small village, but not small enough to get a government grant it can be though.

When you start digging up the streets you must allow competitors to also join, so when one carrier has decided to start digging, you often get offers from other Carriers too, with a slightly lower price. So there's usually many carriers in the same area.

Most of the monthly fee you pay does not go into paying off the initial investment. For example, Verizon wireline's EBITDA margin is 25%. So out of your $100 bill, $75 goes to maintenance, customer service, etc. and $25 is left before paying interest in debt used to build the network, paying taxes, or accounting for the depreciating value of equipment marching toward obsolescence.

They might pay, if they got to chose their provider freely like they can in Sweden, where they build open access networks where any ISP can provide service.

I bet you could build it in to the sale price of reselling a home. IE this improvement must be made to make it distinct in the market.

I would happily pay 100-200k if I could capitalize it on a 30y 0-3% mortgage to bring fiber from a carrier neutral facility (say, the Westin Building in Seattle) to a farm. It's about $10k/km for aerial fiber and $50k/km for buried fiber in most places (far higher in a city, but the distances are shorter.)

There's probably a reasonable market doing this at $10-20k/house. There's clearly a market for multi-family dwellings like condos/apartment buildings (webpass, condointernet, etc. have shown), but maybe the next step will be neighborhoods pre-wired with fiber to some kind of switchbox with your choice of carrier coming in.

In the US we typically capitalize utility projects over long time periods, e.g. 30 years. When you pay your water bill the loan is added into the base payment before usage. We built electricity to virtually every home in the US, we could build fiber too.

It's notable that the phone network was built by a government sanctioned monopoly that was permitted to be quite profitable for most of its existence. Pull AT&T annual reports from the 1950s and 1960s and compare them to the financials of Verizon's wireline division today.

> No urban area in the U.S. is as dense as Stockholm's (3,500). NYC is at the low end of that range (2,000), and is less dense than a bunch of Swedish cities with just ~50-100k people.

Wat. NYC is over 10k/km^2. Manhattan is over 25k/km^2.

The "NYC" urban area encompasses NYC + Westchester + Long Island + Newark + parts of CT. Half of the population that lives/works in/around New York lives outside of the city proper.

NYC is not 13,318 sq mi which is what your talking about aka an area ~200 miles wide and bigger than the state of Maryland. By comparison DC is 61 square miles total your figure is like lumping DC, Baltimore, and Richmond VA into one metro area.

https://en.wikipedia.org/wiki/New_York_metropolitan_area

DC and Baltimore are typically lumped in as one metro area.

https://en.wikipedia.org/wiki/Baltimore%E2%80%93Washington_m...

Not really: "It is composed primarily of two major metropolitan statistical areas, the Washington–Arlington–Alexandria, DC–VA–MD–WV MSA and the Baltimore-Columbia-Towson, MD MSA. In addition, six other smaller urban areas not contiguous to the main urban area but having strong commuting ties with the main area are also included in the metropolitan area."

I mean some people talk about the Northeast megalopolis. https://en.wikipedia.org/wiki/Northeast_megalopolis Boston, New York City, Philadelphia, Baltimore, Washington DC, and some times Richmond. But, that does not mean Boston and DC are somehow the same city.

>"The "NYC" urban area encompasses NYC + Westchester + Long Island + Newark + parts of CT."

Not it does not, that is the "Tri-State" area. I have never heard anyone use the term "NYC Urban area." What you are saying makes it sounds as if those places are suburbs of New York City and none of them are. New York City is 5 boroughs.

Also I think you are confusing Manhattan and New York City. Half the daily population of Manhattan are commuters:

http://cityroom.blogs.nytimes.com/2013/06/03/commuters-nearl...

Okay but then you can't compare that to Stockholm proper. It should be exceedingly obvious to anyone who has even Google image searched the two cities that NYC is much denser.

I didn't compare it to Stockholm proper. I compared the NYC urban area (basically, NYC and it's suburbs and satellite cities) to Stockholm's urban area. Stockholm proper is less dense than NYC proper, but that is by itself not that useful a number. More people live outside NYC in the urban area as live inside NYC. The opposite is true for Stockholm.

What you really are interested in is linear density.

The only metric that matters when building a network is linear route miles of cable per subscriber needed to provide service.

There is a single metric you can use to approximate this for any country: people per mile of road. Every (on-grid) dwelling and business is on a street or a road, so if you put a fiber cable on every road or street, then you will pass every inhabitant in the country.

Sweden has 10 people per mile of road. The US has 30 people per mile of road.

Sweden simply does better, despite having to deploy more infrastructure to provide service.

But neither the U.S. nor Sweden puts fiber on every road and street. Only 60% of Swedish households have fiber, and you can reach that figure without serving households outside of relatively dense towns.

Also, it looks like you're counting Sweden's vast unpaved road network, which makes up about 75% of its roads. I bet very few of the 60% of households with fiber live on an unpaved road. If you compare paved roads only, the U.S. and Sweden have the same linear density. And even then, it's much less expensive to connect two dense urban areas at either end of a 100 mile road than to serve sparse suburbs and exurbs along a 100 mile stretch of road.

> But neither the U.S. nor Sweden puts fiber on every road and street.

Firstly, Sweden pretty much intends to do so. The official goal is to bring at least 100 Mbps service to at least 90% of the population by 2020. Many municipalities/states have programs to build out fiber to everybody.

Sweden also already has a cable running to almost every premise, so in practice every road and every street is already passed. In 2011 xDSL covered 98,5% of the population and only 1100 households or businesses had no service or less than 1 Mbps service.

Secondly, it's an estimate, but it's a far better estimate than population density or any other single metric presented so far. It correlates with the linear miles of cable needed to pass all premises, which is the deciding factor in determining costs. It is also readily available and it measures compratively how hard it would be wire up a nation.

By this measure Sweden does far better, even having a need for more infrastructure per inhabitant to reach the same end result, i.e. broadband service for everyone.

You could cover 90% of the population of Sweden while only covering a small fraction of the roads. You could probably cover 98.5% of the population without covering much of the road network.

> You could cover 90% of the population of Sweden while only covering a small fraction of the roads.

Is this just a guess or do you have anything to back this up with?

> You could probably cover 98.5% of the population without covering much of the road network.

Here's a map of xDSL coverage: http://www.europarl.europa.eu/document/activities/cont/20110...

Let's see how you propose to cover that land area without also covering much of the road network.

Look at the HSPA map on page 5. Look at how little of the country you need to cover to hit 99.6% of the population.

That's an interesting observation. I wonder if that 99.6% figure is a mistake or if they use a different coverage definition for wireless, because the numbers don't add up.

Looking at the map, the northernmost county has barely any HSPA coverage. Nevertheless that county has a population of a quarter of a million inhabitants. To get 99.6% coverage, you'd still have to cover over 200 000 of those and everybody else in the country. That's clearly impossible as 100 000 inhabitants in that county live outside the ten largest "metro" areas.

There's just no way of not covering a lot of roads if if you aim for coverage in the high nineties with such a dispersed population.

> If you compare paved roads only, the U.S. and Sweden have the same linear density

That's a fair point.

> And even then, it's much less expensive to connect two dense urban areas at either end of a 100 mile road than to serve sparse suburbs and exurbs along a 100 mile stretch of road.

Not necessarily. Build costs in dense urban areas are high, but there are more subscribers. Suburbs and exurbs are cheaper to build per mile, but have less customers. It averages out fairly well.

> Sweden simply does better, despite having to deploy more infrastructure to provide service.

While this is probably true in an absolute sense, Sweden is also a much smaller, more homogeneous country than the U.S.; I don't think governments scale very well, so a small affluent country can likely be more nimble and innovative than a large affluent country (think startups vs enterprises).

I've seen this said many times, but I've never understood why people think this argument makes sense. If anything it's the opposite: the larger the country the larger the economies of scale. If there are diseconomies of scale then you can organise it at the state level or municipal level. Being a larger country simply gives you the option to organise it at a larger scale.

> If anything it's the opposite: the larger the country the larger the economies of scale.

Democracies aren't businesses. Getting some kind of consensus from a huge, diverse populace is hard. I think you end up with diseconomies of scale.

> If there are diseconomies of scale then you can organise it at the state level or municipal level

This requires the federal government to now maintain 50 distinct sets of policy, or delegate to regional bureaucrats. Both of these are diseconomies of scale.

These kinds of problems are hard to solve within a large enterprise with a hierarchical power structure; the United States is a much larger, more diverse entity with a distributed/consensus-driven power structure.

> Being a larger country simply gives you the option to organise it at a larger scale.

The argument that it is merely an option assumes that you are able to rationally choose the best option from the ones available. Not the case when we're talking about government. And the larger the government, the more political power is at stake, the less rational the decision making becomes.

The Op also pointed out that not all of Sweden is served by this municipal ownership arrangement. It may very well be the case that the Op has his or her own bias and Sweden is in fact not widely services this way.

Sweden does "x" better argument is getting tired. Sweden has a lot of negative outcomes as well from socializing so much. Our problem isn't that fact that internet isn't socialized its that its monopolized by 3-4 companies. The incentive to innovate at rapid rate is not there but rather keep the status quo. That being said Google and Verizon are two fiber companies that are leading innovation and forcing other companies to keep up or get left behind by leaving a vacuum of people that want that service. Thats how/why capitalism works. Over time fiber options will rise and we will be in a better position to have some options of our provider instead of being stuck with a state run service that could become crappier over time.

In the US, it's locally monopolized with a single provider.

Sometimes you can get DSL, but since Verizon employees get fired for fixing outages now, that will be gone soon.

Interesting, I had to go searching, but here's the info on the firing of Verizon employees:

http://arstechnica.com/information-technology/2016/10/verizo...

Meh, this is just union posturing to try to save obsolete jobs.

"The wireless home phone service, VoiceLink, is not a proper replacement for copper phone lines because it doesn’t work with security alarms, fax machines, medical devices such as pacemakers that require telephone monitoring, and other services, the union said."

Really? Verizon should be maintaining and repairing obsolete infrastructure so people can keep using fax machines? GMAFB.

I live in a smaller city, not in Stockholm (the capital) or anything close to it. I still have access to 1gbit/s if I want to. :)

There isn't that much fiber in the forests, but anywhere there is a city there is. Sweden has a goal to give 90% of the population fiber before 2020 even if that is probably not going to happen.

Jesus Christ, Stockholm is more densely populated than NYC? No wonder getting an apartment around here is so absurdly difficult.

They're comparing to NYC, not Manhattan. NYC includes 5 boroughs (and some of them are _large_), and while Manhattan and inner parts of Brooklyn, Bronx, and Queens are dense, most of Staten Island and outer parts of Bronx and Queens are relatively sparse.

You can use the same trick to come to the surprising conclusion that LA is denser than NYC. Population-weighed density is definitely a better way to measure density if you want numbers that reflect intuition about how dense a place "feels" rather than merely reflecting what arbitrary political boundaries were chosen

It's much worse than that, NYC is 789 km^2 of land, the metropolitan area is 8,936: https://censusreporter.org/profiles/31000US35620-new-york-ne...

Looks like you've awoken to the syndrome.

Based on the writing style, it would seem you're the troll.

I'm not going to get into a debate in the comment thread, but I will respond to you. You should try not to be so reactionary. I am neither cheerleading for the U.S.A. nor condemning Sweden. What I am saying is that the U.S. spans a continent. There are areas of awesome internet and areas of horrible internet and everything in between. If you think large areas of Sweden are sparsely inhabited have a drive through Utah or New Mexico or Arizona. If you look at Europe I assume you'll also find areas of awesome internet and areas of horrible internet, and that would at least be a somewhat reasonable comparison to make with the U.S.

I don't think it has all that much to do with population density. I think it has everything to do with Sweden not having AT&T, Comcast, and TWC to establish the ground rules for network deployment with their lobbyists.

You can't really treat an entire region like it has a uniform population density, either. Area is not the relevant unit of measure, but total linear distance of the edges of the spanning tree.

Maybe a more predictive measure for networking costs would be median distance to nearest neighbor?

Yes, but that's hard to calculate. What you really want to calculate is linear density. My best approximation for this is people per mile of road. This is easy to calculate and this data is readily available.

Sweden has 10 people per mile of road. The US has 30 people per mile of road.

Sweden simply does better, despite having to deploy more infrastructure to provide service.

Sweden used to have a state telecom monopoly which then became the incumbent. You are right that regulatory capture has played a big part in the development of the telecommunications sector. Sweden had a strong incumbent, but that did not stop the municipalities and power companies from building out competing fiber networks, nor stop the regulator from acting in the best interest of the people.

So, the big difference was that the Swedes had both political support for building fiber networks and a more can do attitude towards getting things done, without excluding large parts of the community.

Reply on news.ycombinator.com