[dogma1138]

The EU telecom market is just as horrible, most countries have only 1-2 "true" ISP's with actual infrastructure to their subscribers and 2-3 cell providers the rest are ISP's that rent fibers/coax access and MVNOs.

Setting up infrastructure is extremely expensive especially in cities where allot of the European population lives. The US has a different problem where only 4% of the population lives in a major cities so ISP's have to lay allot of infrastructure and historically the more or less remain around their original starting zones and creep out very slowly.

Fiber/Coax infrastructure in the US should be done on a state and municipal levels and then offered to ISP's at a fixed price to recoup the costs via a tender based on the cost to the end consumer

[0xcde4c3db]

> The US has a different problem where only 4% of the population lives in a major cities

That requires a rather narrow definition of "major cities".

[dogma1138]

The largest 10 cities are home to 16M Americans, the 10 largest cities in the UK are host to 12M Brits. The 100 largest cities in the US are home to 59M Americans, the 100 largest cities in the UK are home to 34M Brits. The US is 5 times as populous but there is only 30% difference between the top 10 cities in the US vs UK and 50% in top 100. Most people in the US live in cities under <100K residents, most people in the UK live in cities with over 250K residents.

The rest of Europe isn't that much different. WW2 caused a major shift of population to major cities while in the US it didn't while the US population is still very urban they size of cities/towns is smaller in general.

Also i really hate when people compare the US to a specific EU country the US is bigger than Europe go out of major cities in Europe and you'll have shitty internet, there are towns and cities in Germany where the best you can get is a 10mbit DSL and it's not a unique case in Europe.

[jsjohnst]

Your numbers are way wrong, even if we discredit another commenter's point about city vs metropolitan areas. The top FIVE cities in the US total over 19 million people. If you instead use the top FIVE metropolitan areas you get just shy of 70 million. That's 20% of the US population living in or directly around the top five cities. Furthermore, there are over 300 cities in the US with populations over 100k. I'm too lazy to do the math to add it up, but I'd venture a guess than 50% of the US population lives in a metropolitan area of over 100k people.

(Edit: fixed 25 for 19, accidentally added a city twice)

[jsjohnst]

One last post to further emphasize my point, in the UK there are zero areas with a population of over 10k per square mile (highest[0] is ~15k per km², which is very roughly ~6k in miles²). In the US, there's >100 cities, per Wikipedia [1].

[0] https://en.m.wikipedia.org/wiki/List_of_English_districts_by... [1] https://en.m.wikipedia.org/wiki/List_of_United_States_cities...

[jessriedel]

Where are you guys getting these numbers? The top 5 cities in the US have 19.2 M residents, and the top 10 have 26 M.

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

This one is more sensitive to noise, but still pretty different than what you quote: The top 5 metro areas have 56.9 M

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

[jsjohnst]

Fixed my 25 figure, accidentally added one city twice I think.

For the ~70 million figure, see here:

https://en.m.wikipedia.org/wiki/List_of_metropolitan_areas_o...

[jessriedel]

I think last link is the wrong data to use. Despite the title of the article, it is using the largest possible "Combined statistical area" that contains the city, rather than anything commonly known as the metro area.

> A combined statistical area (CSA) is composed of adjacent metropolitan (MSA) and micropolitan statistical areas (µSA) in the United States and Puerto Rico that can demonstrate economic or social linkage. The OMB defines a CSA as consisting of various combinations of adjacent metropolitan and micropolitan areas with economic ties measured by commuting patterns.

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

For instance, the "Washington-Baltimore-Arlington, DC-MD-VA-WV-PA Combined Statistical Area" (which is not part of the top 5 cities) includes all of Hampshire County, West Virginia

https://en.wikipedia.org/wiki/Hampshire_County,_West_Virgini...

a Google image search confirms that this is a very rural area.

https://www.google.com/search?safe=off&site=&tbm=isch&source...

[chongli]

If you use metropolitan areas then you're just proving the original point. Metro areas in the US are HUGE! For example, the Greater Las Angeles area is ~34k square miles. The entire country of England is only ~50k square miles! Understanding this is key to understanding why the cable and telecom businesses in the US are so difficult to enter.

[jsjohnst]

Per this link [0], there's 10 million homes (out of 90 million with broadband) in the US with fiber to the premise, versus 250k in the UK. The number seems a bit suspect, but I generally trust Ars.

[0] http://arstechnica.com/information-technology/2016/05/how-th...

[jsjohnst]

Citation needed for the ~34k mi² estimate. Per Wikipedia it's <1/7 that number (and yet has 1/4 the population).

https://en.m.wikipedia.org/wiki/Los_Angeles_metropolitan_are...

[chongli]

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

[jsjohnst]

Also, the point I was refuting from the parent post is the claim that Europe is urban where as the US is not. That's simply not factual, as others have also pointed out.

[0xcde4c3db]

Many small/medium American "cities" are municipalities that are directly attached to (or even surrounded by) a larger urban area. It's basically a quirk of how local government is structured rather than anything that has to do with population or infrastructure density. For example, it's unlikely that many residents of Beverly Hills (<50K residents) think of themselves as living in a small town outside of Los Angeles.

[eli_gottlieb]

Similarly, the municipalities of Boston and Cambridge remain separated by the natural border of the Charles River, even if by nothing else at all. Cambridge and Somerville don't even have that much justification for existing as separate municipalities: you can accidentally walk across the municipal border and back again without noticing it in some places.

[russell]

I'm not sure where you got your numbers. The 10 largest cities in the US have a population of 26M (2015 estimates, Wikipedia). Generally population estimates are given for the metropolitan area. The 10 largest have a population of 73 million. There are 53 MSAs with a population over 1M with perhaps a total population of 150M. (I was too lazy to do all the addition.) The US has a large area, but it is primarily urban.

[Camillo]

> The largest 10 cities are home to 16M Americans, the 10 largest cities in the UK are host to 12M Brits. The 100 largest cities in the US are home to 59M Americans, the 100 largest cities in the UK are home to 34M Brits. The US is 5 times as populous but there is only 30% difference between the top 10 cities in the US vs UK and 50% in top 100.

This is a nonsensical statistic to use, irrespective of whether the numbers of right or wrong. Imagine taking the UK and copy-pasting it 10 times. Make the new deca-UK a single country, let's call it 10UK.

Obviously, 10UK has exactly the same proportion of people living in "major cities" as the original UK. Obviously, the infrastructure serving the 10 copies of each major city is equally affordable for a country with 10x of everything (or even more affordable, due to more economies of scale).

But the 10 largest cities of 10UK are the 10 copies of London, and thus the proportion of people living in the 10 largest cities of 10UK is equal to the proportion of people living in the single largest city in the UK. Clearly, this is the wrong number to use to estimate anything related to the problem at hand.

This is one of those rare situations where avoiding the wrong conclusion requires zero knowledge of the world; cognitive ability on its own is sufficient. In other words, you are demonstrably stupid.

[jsjohnst]

Responding with facts is enough, attacking the poster personally as you did in your last sentence isn't needed.