3 ms·

The problem with old fiber is that active networks are even more of a dead-end than copper. Active fiber networks leave you with whatever tech you buried and pu

The problem with old fiber is that active networks are even more of a dead-end than copper. Active fiber networks leave you with whatever tech you buried and put up on every corner; upgrading is basically as expensive as laying new line. It seems to me like in the 80s and 90s active networks were favored (see e.g. OPAL) - which would be far more useless today than copper, as copper's capabilities expanded hugely over time as more sophisticated modulation techniques became possible.

radicaldreamer 5y ago
Copper is still trash though compared to passive fiber... and expensive to run. Loads of electricity and amps required to use it and a lot of fine tuning and interference management and maintenance. A ton of costly, endless work when you can bypass all of that by laying some passive fiber...
quink 5y ago
> upgrading is basically as expensive as laying new line That assertion is highly dependent on too many factors and even recent developments to be of much use. That general assertion needs to be backed up with everything around the soil type, the copper diameter, labour costs, protocols and type of fibre, etc., etc. And if you’re going down the route of having VDSL ISAMs in the field anyway then you’re certainly going to have the infrastructure there already anyway to support GPON over whatever fibre is in the ground anyway. And are you sure that by “copper’s capabilities” you’re not actually referring to progressively closer deployment of infrastructure to the customer to overcome copper’s limitations? Because the move from dial any number to the local exchange to a node (FTTN) to now the curb (FTTC) certainly seems to reflect that more adequately.
- gopher_space 5y ago
  Are there more people available to work on one vs the other, or would you be trained to handle both?
ZeroGravitas 5y ago
Do you have any links to Korea or Japan hitting these insurmountable problems? Last I heard they had better internet than the UK fairly consistently for decades. Though, if they did hit a problem, just sensibly regulating shared national infrastructure of high industrial, cultural and economic importance seems like it would solve anything that came up better rather than shouting "free market" and "regulation bad" while you hand things over to American corporations in an act of brazen corruption.
- simne 5y ago
  > Do you have any links to Korea or Japan hitting these insurmountable problems? You remembered me talks, about how enhanced was Hong Kong, so they in 1990s built very high-speed internet for all. But than I read wikipedia, and seen, Hong Kong have 6800 people per square kilometer (Germany 200, France 300, Ukraine 80, Russia less then 30). Sure, if you have more than tenfold number of clients, you have much more possibilities to pay for expensive new technology. But even for not poor western Europe, fiber technologies was too expensive for wide usage until mid 2000s. Sure, backbones built on fiber from early 1990s, but backbones is not all internet, in reality 90% of resources spent on last mile. And as I said before, until late 2000s, only copper last mile was economically effective without Asian density. Also was possible fiber backbone, and wireless last mile, but air speed also very limited (should read expensive :) ).
  - 411111111111111 5y ago
    > But than I read wikipedia, and seen, Hong Kong have 6800 people per square kilometer (Germany 200, France 300, Ukraine 80, Russia less then 30). Hong kong is essentially a city. You'd have to compare it with Munich (5k/km^2) in Germany or Paris (20k/km^2) in France to get equivalent numbers
    - simne 5y ago
      I talked with colleagues in early 2000s, and they said, Paris central parts with high density was connected to broadband relatively early, but farther distribution was very slow, exact as I said.

radicaldreamer 5y ago

Copper is still trash though compared to passive fiber... and expensive to run. Loads of electricity and amps required to use it and a lot of fine tuning and interference management and maintenance. A ton of costly, endless work when you can bypass all of that by laying some passive fiber...

quink 5y ago

> upgrading is basically as expensive as laying new line That assertion is highly dependent on too many factors and even recent developments to be of much use. That general assertion needs to be backed up with everything around the soil type, the copper diameter, labour costs, protocols and type of fibre, etc., etc. And if you’re going down the route of having VDSL ISAMs in the field anyway then you’re certainly going to have the infrastructure there already anyway to support GPON over whatever fibre is in the ground anyway. And are you sure that by “copper’s capabilities” you’re not actually referring to progressively closer deployment of infrastructure to the customer to overcome copper’s limitations? Because the move from dial any number to the local exchange to a node (FTTN) to now the curb (FTTC) certainly seems to reflect that more adequately.

gopher_space 5y ago
Are there more people available to work on one vs the other, or would you be trained to handle both?

ZeroGravitas 5y ago

Do you have any links to Korea or Japan hitting these insurmountable problems? Last I heard they had better internet than the UK fairly consistently for decades. Though, if they did hit a problem, just sensibly regulating shared national infrastructure of high industrial, cultural and economic importance seems like it would solve anything that came up better rather than shouting "free market" and "regulation bad" while you hand things over to American corporations in an act of brazen corruption.

simne 5y ago
> Do you have any links to Korea or Japan hitting these insurmountable problems? You remembered me talks, about how enhanced was Hong Kong, so they in 1990s built very high-speed internet for all. But than I read wikipedia, and seen, Hong Kong have 6800 people per square kilometer (Germany 200, France 300, Ukraine 80, Russia less then 30). Sure, if you have more than tenfold number of clients, you have much more possibilities to pay for expensive new technology. But even for not poor western Europe, fiber technologies was too expensive for wide usage until mid 2000s. Sure, backbones built on fiber from early 1990s, but backbones is not all internet, in reality 90% of resources spent on last mile. And as I said before, until late 2000s, only copper last mile was economically effective without Asian density. Also was possible fiber backbone, and wireless last mile, but air speed also very limited (should read expensive :) ).
- 411111111111111 5y ago
  > But than I read wikipedia, and seen, Hong Kong have 6800 people per square kilometer (Germany 200, France 300, Ukraine 80, Russia less then 30). Hong kong is essentially a city. You'd have to compare it with Munich (5k/km^2) in Germany or Paris (20k/km^2) in France to get equivalent numbers
  - simne 5y ago
    I talked with colleagues in early 2000s, and they said, Paris central parts with high density was connected to broadband relatively early, but farther distribution was very slow, exact as I said.