4 ms·

Honestly, the more network is out there, the better. There is probably an upper limit on the number of networks any given city can support, so it's going to be

by Nrsolis 14y ago

Honestly, the more network is out there, the better. There is probably an upper limit on the number of networks any given city can support, so it's going to be natural for some combination of content sources to be partnered with fiber distribution networks to revenue share. It's enormously costly to deliver "wireline" fiber to the home service. Most of this is labor cost, but you also can't discount the impact of property taxes and upkeep on infrastructure that is supposed to weather the elements and wildlife (including human) for upwards of 30 years. Fiber is a great solution but wireless is still the most economical way to deliver access for the "last-mile". If we ever find a way to provide low-cost, high bandwidth wireless service within a one mile radius that doesn't make the NIMBY types have an aneurism, then fiber-to-the-home will seem as quaint as an individual copper pair to every residence. disclaimer: In 1997-1998 I worked at a municipally owned city utility that was able to deliver 10mbps symmetric Internet access to a development of homes in FL. Bellsouth subsequently had laws passed to prohibit political subdivisions from engaging in the provisioning of telecommunication services.

jaggederest 14y ago
Fiber is fundamentally no different than wireless. It's just at a high enough frequency the light is visible. That means you get much higher bandwidth. There's just no way around that. If you could rig up visible-light laser communications, you wouldn't need the fiber. Fiber's just a convenient container.
- Cogito 14y ago
  Fundamentally, the two are very different. Whilst the information carrier, photons, is identical in both cases, the medium is extremely different. In the case of fibre we have a highly controlled, flexible medium with extremely good quality of service and very little attenuation. This comes at the price of requiring relatively expensive infrastructure between network nodes. Wireless technology on the other hand travels through air as a medium. You remove the end-to-end infrastructure requirement, but the trade off is much higher attenuation and lower quality of service. Additionally, you require some amount of 'direct line of sight' depending on frequency. Lower (and hence slower) frequencies are able to refract around mountains and even off the outer atmosphere, providing quite good coverage, but visible light has no such capability. Repeater satellites or LOS base stations would be required to effectively use such a frequency. This is why HAM radios are great for talking to people kilometres away, but can't be used for effective wireless internet, and why hotspots don't cover more than a few hundred metres at most.
  - Nrsolis 14y ago
    Fiber is just a convenient waveguide for certain electromagnetic frequencies. Bad for others. That said, your understanding is a bit "off." Amateur Radio (HAM) has microwave frequencies available for use that would be VERY effective for point-to-point wireless internet. We even have frequencies that are very close to the existing bands and there would be no problem running LTE over those frequencies. The problems with amateur radio are regulatory, not technical. Generally speaking, the higher the frequency, the higher the attenuation. Of course, the higher frequencies are better suited for the kinds of modulation rates necessary for high bitrate communications. So there's that. But all of this is besides the point. No technology exists in a vacuum and must compete against others for particular applications and for particular cost models. That's why telecommunications engineers have no hair and bad tempers. This stuff is hard. For "lower" bitrate communications, wireless is the hands-down winner. The cost to deploy a network for a given level of service to a range of end-stations is just so much easier and cheaper that lots of other nations skipped over building a wireline network and just went straight to wireless. Mobility of the stations is almost a secondary concern. Once you get to the point where your communications are less intermittent and higher bitrate, it's really a question of how far you need to go and how much/how often you need to communicate. Lots of LTE stations are connected to a central office (router) via fiber (gigabit ethernet mostly). Lots more are connected via point to point microwave. Backhaul of bandwidth is something you dont hear talked about a lot but it's a big part of why it can be hard to get LTE/4G/WiMax into a particular location. Economics are a big part of that equation. SO, I wish the Google Fiber guys well and I hope they succeed, but the telecommunications market is so broad and complicated that it would be foolish to look at one company's efforts to shake up a single market and equate that with a sea-change in how the broadband market works. edit: I met Milo Medin (who is now running Google Fiber) and he offered me a chance to work at @Home way back when. In retrospect, I probably should have pursued it.
    - Cogito 14y ago
      Thanks for the in-depth reply. The example of HAM radio was a poor one, I was associating it with Low Frequency [1] radio when it in fact covers a much larger spectrum. I agree that telecommunications is a hard subject, and even more so that different technologies are useful for different purposes. That is why I think the original comment was off: > If you could rig up visible-light laser communications, you wouldn't need the fiber. Fiber's just a convenient container. Something we haven't really talked about, and I'm not sure how relevant it is, is the saturation of the wireless spectrum. Point to Point wireless is obviously a different beast, but for broadcast wireless is there a saturation point where we can't safely send more data over the airwaves? I like that fibre is by construction point to point, not restricted by line of sight, capable of very high bit-rates with relatively low energy (correct me here if this is wrong), and excellent quality of service. Wireless has a lot less infrastructure, particularly when used for the last mile, and is definitely extremely convenient in many situations. I don't think it is the solution for general purpose network infrastructure, though that may change in the future (and is clearly the choice for some countries already). I wonder what the replacement for fibre will be in 20-30 years time - is wireless the only frontier at the moment? [1] https://en.wikipedia.org/wiki/Low_frequency#Propagation_of_LF_signals https://en.wikipedia.org/wiki/Low_frequency#Propagation_of_L...

by Nrsolis 14y ago

jaggederest 14y ago

Fiber is fundamentally no different than wireless. It's just at a high enough frequency the light is visible. That means you get much higher bandwidth. There's just no way around that. If you could rig up visible-light laser communications, you wouldn't need the fiber. Fiber's just a convenient container.

Cogito 14y ago
Fundamentally, the two are very different. Whilst the information carrier, photons, is identical in both cases, the medium is extremely different. In the case of fibre we have a highly controlled, flexible medium with extremely good quality of service and very little attenuation. This comes at the price of requiring relatively expensive infrastructure between network nodes. Wireless technology on the other hand travels through air as a medium. You remove the end-to-end infrastructure requirement, but the trade off is much higher attenuation and lower quality of service. Additionally, you require some amount of 'direct line of sight' depending on frequency. Lower (and hence slower) frequencies are able to refract around mountains and even off the outer atmosphere, providing quite good coverage, but visible light has no such capability. Repeater satellites or LOS base stations would be required to effectively use such a frequency. This is why HAM radios are great for talking to people kilometres away, but can't be used for effective wireless internet, and why hotspots don't cover more than a few hundred metres at most.
- Nrsolis 14y ago
  Fiber is just a convenient waveguide for certain electromagnetic frequencies. Bad for others. That said, your understanding is a bit "off." Amateur Radio (HAM) has microwave frequencies available for use that would be VERY effective for point-to-point wireless internet. We even have frequencies that are very close to the existing bands and there would be no problem running LTE over those frequencies. The problems with amateur radio are regulatory, not technical. Generally speaking, the higher the frequency, the higher the attenuation. Of course, the higher frequencies are better suited for the kinds of modulation rates necessary for high bitrate communications. So there's that. But all of this is besides the point. No technology exists in a vacuum and must compete against others for particular applications and for particular cost models. That's why telecommunications engineers have no hair and bad tempers. This stuff is hard. For "lower" bitrate communications, wireless is the hands-down winner. The cost to deploy a network for a given level of service to a range of end-stations is just so much easier and cheaper that lots of other nations skipped over building a wireline network and just went straight to wireless. Mobility of the stations is almost a secondary concern. Once you get to the point where your communications are less intermittent and higher bitrate, it's really a question of how far you need to go and how much/how often you need to communicate. Lots of LTE stations are connected to a central office (router) via fiber (gigabit ethernet mostly). Lots more are connected via point to point microwave. Backhaul of bandwidth is something you dont hear talked about a lot but it's a big part of why it can be hard to get LTE/4G/WiMax into a particular location. Economics are a big part of that equation. SO, I wish the Google Fiber guys well and I hope they succeed, but the telecommunications market is so broad and complicated that it would be foolish to look at one company's efforts to shake up a single market and equate that with a sea-change in how the broadband market works. edit: I met Milo Medin (who is now running Google Fiber) and he offered me a chance to work at @Home way back when. In retrospect, I probably should have pursued it.
  - Cogito 14y ago
    Thanks for the in-depth reply. The example of HAM radio was a poor one, I was associating it with Low Frequency [1] radio when it in fact covers a much larger spectrum. I agree that telecommunications is a hard subject, and even more so that different technologies are useful for different purposes. That is why I think the original comment was off: > If you could rig up visible-light laser communications, you wouldn't need the fiber. Fiber's just a convenient container. Something we haven't really talked about, and I'm not sure how relevant it is, is the saturation of the wireless spectrum. Point to Point wireless is obviously a different beast, but for broadcast wireless is there a saturation point where we can't safely send more data over the airwaves? I like that fibre is by construction point to point, not restricted by line of sight, capable of very high bit-rates with relatively low energy (correct me here if this is wrong), and excellent quality of service. Wireless has a lot less infrastructure, particularly when used for the last mile, and is definitely extremely convenient in many situations. I don't think it is the solution for general purpose network infrastructure, though that may change in the future (and is clearly the choice for some countries already). I wonder what the replacement for fibre will be in 20-30 years time - is wireless the only frontier at the moment? [1] https://en.wikipedia.org/wiki/Low_frequency#Propagation_of_LF_signals https://en.wikipedia.org/wiki/Low_frequency#Propagation_of_L...