5 ms·

Density is not something that can be described adequately in a single number. It is entirely possible to have high residential densities in cities with remarkab

Density is not something that can be described adequately in a single number. It is entirely possible to have high residential densities in cities with remarkably low average densities. Moscow is a great example of this...their plazas and parks drive down average density way beyond the density that the average person experiences. Everybody talks about the last mile being the most expensive part of infrastructure, but if the last meter is 90% of the cost of the last mile (I don't know what the exact proportion is, but it is definitely close), then 100 houses in 10 sq km census block is going to be in a cost category that is orders of magnitude higher than 100 apartments in a tower with a 9.9 sq km park around it.

dublinben 13y ago
Density either prevents better infrastructure in the US, or it doesn't. My use of statistics was merely to show that density has no relation to the existence of fiber internet infrastructure. The US cannot be too sprawling for fiber if areas with lower density in other countries have better internet. Major cities cannot also be too dense when cities with higher densities abroad also have better internet. This is a commonly repeated argument in favor of the status-quo in the US, and I'm just trying to refute it whenever possible.
- saosebastiao 13y ago
  Your use of statistics is meaningless because average density is meaningless. It suffers from the same resolution problem as the coastline paradox [1]. I'm an example of this: If you measure density by the block, I live in a density of 120,000 persons/km2, but if you measure density by the city, I live in a density of 1720 persons/km2, and if you measure density by the MSA, I live in a density of 168 persons/km2. I have gigabit, and the fact that I live in an MSA with 168 persons/km2 is in no way a refutation of the fact that density matters for infrastructure costs. [1] http://en.wikipedia.org/wiki/Coastline_paradox http://en.wikipedia.org/wiki/Coastline_paradox
  - brenschluss 13y ago
    Saying average density is meaningless is like saying, because of the coastline paradox, even a small island's coastline is the same length as a large continent's coastline -- they're both infinite!
    - saosebastiao 13y ago
      Correct. However, if you can define a level of resolution to measure the coastline/density, then then it can be compared. Average density/coastline is meaningless without context.
      - brenschluss 13y ago
        Great point.
reversi_online 13y ago
That still doesn't explain why there's no gigabit FTTH in NYC, with its high average density. It sure as heck doesn't explain why there's no gigabit FTTH in Manhattan proper--by your reckoning the borough should have a residential density even higher than its overall average would suggest by virtue of Central Park taking up so much of the upper half.
- saosebastiao 13y ago
  High density doesn't eliminate millions of other reasons that a given city doesn't have it. It just provides a floor for cost competitiveness, and a major input into ROI. Even in Kansas City, google has stated they won't roll out FTTH to every neighborhood, even in the long term...and in the short term, they have focused on the neighborhoods with the fastest ROI. I'll let you guess whether the high density or the low density gets built first.
  - reversi_online 13y ago
    Then why did you bring up density in the first place? Unless you're saying that urban and suburban America each lack FTTH for their own wholly unrelated reasons, an explanation that seems rather lacking in parsimony. [1] http://en.wikipedia.org/wiki/Occam%27s_razor http://en.wikipedia.org/wiki/Occam%27s_razor
    - saosebastiao 13y ago
      Because a huge and significant chunk of America lives at a density that is prohibitive for small competitors to provide FTTH. The only way to get it is through municipal funding, because the large broadband providers certainly aren't going to provide it.

dublinben 13y ago

Density either prevents better infrastructure in the US, or it doesn't. My use of statistics was merely to show that density has no relation to the existence of fiber internet infrastructure. The US cannot be too sprawling for fiber if areas with lower density in other countries have better internet. Major cities cannot also be too dense when cities with higher densities abroad also have better internet. This is a commonly repeated argument in favor of the status-quo in the US, and I'm just trying to refute it whenever possible.

saosebastiao 13y ago
Your use of statistics is meaningless because average density is meaningless. It suffers from the same resolution problem as the coastline paradox [1]. I'm an example of this: If you measure density by the block, I live in a density of 120,000 persons/km2, but if you measure density by the city, I live in a density of 1720 persons/km2, and if you measure density by the MSA, I live in a density of 168 persons/km2. I have gigabit, and the fact that I live in an MSA with 168 persons/km2 is in no way a refutation of the fact that density matters for infrastructure costs. [1] http://en.wikipedia.org/wiki/Coastline_paradox http://en.wikipedia.org/wiki/Coastline_paradox
- brenschluss 13y ago
  Saying average density is meaningless is like saying, because of the coastline paradox, even a small island's coastline is the same length as a large continent's coastline -- they're both infinite!
  - saosebastiao 13y ago
    Correct. However, if you can define a level of resolution to measure the coastline/density, then then it can be compared. Average density/coastline is meaningless without context.
    - brenschluss 13y ago
      Great point.

reversi_online 13y ago

That still doesn't explain why there's no gigabit FTTH in NYC, with its high average density. It sure as heck doesn't explain why there's no gigabit FTTH in Manhattan proper--by your reckoning the borough should have a residential density even higher than its overall average would suggest by virtue of Central Park taking up so much of the upper half.

saosebastiao 13y ago
High density doesn't eliminate millions of other reasons that a given city doesn't have it. It just provides a floor for cost competitiveness, and a major input into ROI. Even in Kansas City, google has stated they won't roll out FTTH to every neighborhood, even in the long term...and in the short term, they have focused on the neighborhoods with the fastest ROI. I'll let you guess whether the high density or the low density gets built first.
- reversi_online 13y ago
  Then why did you bring up density in the first place? Unless you're saying that urban and suburban America each lack FTTH for their own wholly unrelated reasons, an explanation that seems rather lacking in parsimony. [1] http://en.wikipedia.org/wiki/Occam%27s_razor http://en.wikipedia.org/wiki/Occam%27s_razor
  - saosebastiao 13y ago
    Because a huge and significant chunk of America lives at a density that is prohibitive for small competitors to provide FTTH. The only way to get it is through municipal funding, because the large broadband providers certainly aren't going to provide it.