Internet access for some!
August 1, 2006 8:03 AM Subscribe
The Boston Wi-Fi Network may be constructed within the year by a non-profit corporation. An appointed task force has produced a report (pdf) which recommends building the network on the cheap and allowing providers to compete over the chance to provide service. It won't be free though. Can this possibly work?
I don't really understand the point of this movement towards free Wi-Fi. My assumption is that private companies are more efficient at distributing broadband services than the government. And the truly poor can't afford food, let alone computers, so they won't benefit. The fringe (lower middle class) won't benefit because they'll receive less of the really beneficial services from the government since government spending is being diverted to running and maintaining this service. And the middle and upper classes can already afford broadband, so this is almost like a handout to them.
Maybe I'm missing something. Public libraries already have computers with broadband which can be used for free. And I'm guessing that services like transportation, healthcare, and housing are far more important to those who are in greatest need of government help (when comparing to Wi-Fi). So what's the point?
posted by SeizeTheDay at 8:40 AM on August 1, 2006
Maybe I'm missing something. Public libraries already have computers with broadband which can be used for free. And I'm guessing that services like transportation, healthcare, and housing are far more important to those who are in greatest need of government help (when comparing to Wi-Fi). So what's the point?
posted by SeizeTheDay at 8:40 AM on August 1, 2006
My assumption is that private companies are more efficient at distributing broadband services than the government.
Not in the United States, they're not:
With a hefty push from the government, South Korea's telecommunications providers have built the world's most comprehensive Internet network, supplying affordable and reliable access that far surpasses what is available in the United States, even in those homes that have their own broadband setup.
Basically, Philadelphia saw an opportunity to set itself apart from other cities and attract creative, young people to the taxbase. The push-back in Pennsylvania came from the state's subsidized communications monopolies who saw competition from the public sector. Laws were bought to stifle competition and broadband adoption was slowed down yet again.
posted by Blazecock Pileon at 8:46 AM on August 1, 2006
Not in the United States, they're not:
With a hefty push from the government, South Korea's telecommunications providers have built the world's most comprehensive Internet network, supplying affordable and reliable access that far surpasses what is available in the United States, even in those homes that have their own broadband setup.
Basically, Philadelphia saw an opportunity to set itself apart from other cities and attract creative, young people to the taxbase. The push-back in Pennsylvania came from the state's subsidized communications monopolies who saw competition from the public sector. Laws were bought to stifle competition and broadband adoption was slowed down yet again.
posted by Blazecock Pileon at 8:46 AM on August 1, 2006
Indeed, the US' subsidized monopolies on telecoms make our infrastructure craptastic. My loathing for the telecoms industry knows no bounds. The cable company I get my internet access from may be incompetent, but the phone company is mean.
posted by Skorgu at 8:53 AM on August 1, 2006
posted by Skorgu at 8:53 AM on August 1, 2006
Does it work for Philadelphia? My impression is that other than a few pilot areas, Philly is nowhere close to a full citywide rollout of Wi-Fi. In fact, other than a few small communities, or municipalities that have hotspots in a handful of locations, I'm not aware of any city-wide Wi-Fi that's actually in production.
The thing that nobody ever wants to say is that:
* Wi-Fi doesn't easily go through concrete, plaster, or insulated walls. Everyone who has an access point knows this -- it's often difficult to get coverage through an entire house, much less a city block or a quarter-mile. The number of access points you'd need to actually get Wi-Fi in the home is far greater than the number the cities are building out. These networks are only going to work acceptably in the streets, and if you want to use them in the home, you'll need to rig an antenna in a window or on the outside of your home. Who will want spotty connection from a rigged-up antenna for $15-20/month versus DSL for the same price? Think about all the people who get poor reception with their cell phones in their homes -- Wi-Fi will be worse.
* The mesh networks that the cities are using, primarily manufactured by Tropos, severely reduce the available bandwidth to most of the access points. A smaller percentage of "supernodes" will have connectivity to the Internet -- if you're on an access point one hop away from the connection, the maximum bandwidth is halved as the base stations send Internet traffic amongst themselves over the wireless. Two hops, you're quartered. Three hops, if other people are using the network, forget it. If the service takes off, the quality of the connection will be poor -- and again, they're looking to charge fees similar to DSL for the honor.
* Wi-Fi interference sucks. Lots of things compete for the 2.4ghz space, including many microwave ovens. It's one thing when you run your own home network and can control what's going on in your own house to avoid interference, but when the access point is two blocks away? The chance of connectivity being impeded by interference is high. Since Wi-Fi works on unlicensed, unregulated frequencies, anyone can (inadvertently) jam the neighborhood signal.
Free Wi-Fi to parks and libraries and common areas is a wonderful thing, but setting up neighborhood access points and charging for service will be a debacle.
posted by I EAT TAPAS at 8:55 AM on August 1, 2006
The thing that nobody ever wants to say is that:
* Wi-Fi doesn't easily go through concrete, plaster, or insulated walls. Everyone who has an access point knows this -- it's often difficult to get coverage through an entire house, much less a city block or a quarter-mile. The number of access points you'd need to actually get Wi-Fi in the home is far greater than the number the cities are building out. These networks are only going to work acceptably in the streets, and if you want to use them in the home, you'll need to rig an antenna in a window or on the outside of your home. Who will want spotty connection from a rigged-up antenna for $15-20/month versus DSL for the same price? Think about all the people who get poor reception with their cell phones in their homes -- Wi-Fi will be worse.
* The mesh networks that the cities are using, primarily manufactured by Tropos, severely reduce the available bandwidth to most of the access points. A smaller percentage of "supernodes" will have connectivity to the Internet -- if you're on an access point one hop away from the connection, the maximum bandwidth is halved as the base stations send Internet traffic amongst themselves over the wireless. Two hops, you're quartered. Three hops, if other people are using the network, forget it. If the service takes off, the quality of the connection will be poor -- and again, they're looking to charge fees similar to DSL for the honor.
* Wi-Fi interference sucks. Lots of things compete for the 2.4ghz space, including many microwave ovens. It's one thing when you run your own home network and can control what's going on in your own house to avoid interference, but when the access point is two blocks away? The chance of connectivity being impeded by interference is high. Since Wi-Fi works on unlicensed, unregulated frequencies, anyone can (inadvertently) jam the neighborhood signal.
Free Wi-Fi to parks and libraries and common areas is a wonderful thing, but setting up neighborhood access points and charging for service will be a debacle.
posted by I EAT TAPAS at 8:55 AM on August 1, 2006
Does it work for Philadelphia? My impression is that other than a few pilot areas, Philly is nowhere close to a full citywide rollout of Wi-Fi.
That's correct, hence why Verizon and Comcast are (deliberately) stretching this out many years longer than the roll-out should have taken.
posted by Blazecock Pileon at 8:57 AM on August 1, 2006
That's correct, hence why Verizon and Comcast are (deliberately) stretching this out many years longer than the roll-out should have taken.
posted by Blazecock Pileon at 8:57 AM on August 1, 2006
My (newly-adopted) small city (18k people) has free Wi-Fi in the downtown waterfront area. Apparently it's completely supported by local businesses via unobtrusive on-screen ads and didn't cost the city a dime to implement. Can't understand why when this tech is implemented over larger areas or in bigger communities someone always tries to find a way to make it expensive to the local users or taxpayers.
posted by caution live frogs at 9:01 AM on August 1, 2006
posted by caution live frogs at 9:01 AM on August 1, 2006
caution: that waterfront area is around a quarter-mile to half-mile of open space, right? Wi-Fi does acceptably in that environment -- the range is decent in open space, but as soon as it starts hitting solid objects, the performance degrades spectacularly. When you start looking at coverage in a city like Boston or Philly, where there's narrow streets and solidly-constructed old buildings, often with thick brick walls and plaster, you need to bump up the number of access points you'd use for open space in the same area by 4 to 8 times.
Now, think about how large Philly is. They're looking to cover 135 square miles. This is likely 540 times larger than the area in your small municipality's waterfront. Since it's dense with buildings and narrow streets, they need to use 4 to 8 times the number of access points. All of a sudden, you're talking 3240 times the cost. You're talking a couple of orders of magnitude here.
Next, you need to think about access rights. The access points are fairly easy here -- a city can drop them on light poles. But you need to actually get *Internet access* to a number of those access points, and that starts getting expensive. You can either build fiber to particular locations (which is pricy), or you can use pre-WiMax technology to beam the Internet access to the access points. The issue with pre-WiMax is that you need towers and roof/wall clearance on buildings to mount the antennas, and those towers and roof/wall clearances don't come cheap unless you're lucky and have municipal buildings in exactly the right places. Your municipality can run DSL to a shed in the middle of the waterfront, but, Philly needs to get Internet access out to a large number of access points throughout a 135 square mile area.
Finally, the goal of Philly's wireless is to get Internet access into the *homes* of people. Free wi-fi for people checking their E-mail on a sunny day is one thing, but people use their home connections more aggressively, for downloading video, using VoIP, hitting web sites hard. You need a lot more bandwidth, and that extra bandwidth ends up being a lot more cash.
Just for sake of reference, Philly already *has* free Wi-Fi in a number of parks, and have had it for a couple of years now. Free Wi-Fi in the parks is relatively easy. Covering an entire major metropolitan area is a completely different story in terms of technology, project coordination, infrastructure and the way people will use the network.
posted by I EAT TAPAS at 9:29 AM on August 1, 2006
Now, think about how large Philly is. They're looking to cover 135 square miles. This is likely 540 times larger than the area in your small municipality's waterfront. Since it's dense with buildings and narrow streets, they need to use 4 to 8 times the number of access points. All of a sudden, you're talking 3240 times the cost. You're talking a couple of orders of magnitude here.
Next, you need to think about access rights. The access points are fairly easy here -- a city can drop them on light poles. But you need to actually get *Internet access* to a number of those access points, and that starts getting expensive. You can either build fiber to particular locations (which is pricy), or you can use pre-WiMax technology to beam the Internet access to the access points. The issue with pre-WiMax is that you need towers and roof/wall clearance on buildings to mount the antennas, and those towers and roof/wall clearances don't come cheap unless you're lucky and have municipal buildings in exactly the right places. Your municipality can run DSL to a shed in the middle of the waterfront, but, Philly needs to get Internet access out to a large number of access points throughout a 135 square mile area.
Finally, the goal of Philly's wireless is to get Internet access into the *homes* of people. Free wi-fi for people checking their E-mail on a sunny day is one thing, but people use their home connections more aggressively, for downloading video, using VoIP, hitting web sites hard. You need a lot more bandwidth, and that extra bandwidth ends up being a lot more cash.
Just for sake of reference, Philly already *has* free Wi-Fi in a number of parks, and have had it for a couple of years now. Free Wi-Fi in the parks is relatively easy. Covering an entire major metropolitan area is a completely different story in terms of technology, project coordination, infrastructure and the way people will use the network.
posted by I EAT TAPAS at 9:29 AM on August 1, 2006
Nice summation Tapas. I like my park wifi.
posted by craven_morhead at 9:50 AM on August 1, 2006
posted by craven_morhead at 9:50 AM on August 1, 2006
PipeDreamFilter: Does anyone else think that the solution is to put APs on lightpoles with a mesh network just like above, but drop some coin on outfitting those APs/mesh nodes with software defined radios and a decent CPU.
For the moment, use .11a as the uplink freq and .11b as the usage freq, then when wimax gets good rollout new firmware and *bam* everyone gets a speed increase and nobody needs to upgrade the hardware.
While you're installing the nodes, drop dark fiber down to the service tunnel and when you service those tunnels, pull fiber along with whatever needs to be done. Bit by bit you light up that dark fiber and then you don't need wireless uplinks in some areas. You'd start out as having islands of fiber-connected nodes around muni-owned buildings and as normal infrastructure replacement proceeds eventually you get the whole city connected. Since laying 50 strands of fiber is as cheap as laying 1, you eventually have the beginnings of a fiber-to-the-curb setup, all you need is endpoints.
Yes, its a lot of up-front expense, but so was the interstate system.
posted by Skorgu at 10:05 AM on August 1, 2006
For the moment, use .11a as the uplink freq and .11b as the usage freq, then when wimax gets good rollout new firmware and *bam* everyone gets a speed increase and nobody needs to upgrade the hardware.
While you're installing the nodes, drop dark fiber down to the service tunnel and when you service those tunnels, pull fiber along with whatever needs to be done. Bit by bit you light up that dark fiber and then you don't need wireless uplinks in some areas. You'd start out as having islands of fiber-connected nodes around muni-owned buildings and as normal infrastructure replacement proceeds eventually you get the whole city connected. Since laying 50 strands of fiber is as cheap as laying 1, you eventually have the beginnings of a fiber-to-the-curb setup, all you need is endpoints.
Yes, its a lot of up-front expense, but so was the interstate system.
posted by Skorgu at 10:05 AM on August 1, 2006
The solution to the problem is WiMax, not WiFi, for a lot of technical reasons. WiFi is a LAN technology, WiMax is a WAN technology, designed from the gitgo for creating metropolitan area mobile networks.
Chicken and egg problem though, slowly being overcome by early system developers, and overseas interests. When the Koreans and Japanese have metropolitan WiMax networks with millions of users up in a year or two, WiFi mesh network builders here will still be trying to make them work.
posted by paulsc at 11:36 AM on August 1, 2006
Chicken and egg problem though, slowly being overcome by early system developers, and overseas interests. When the Koreans and Japanese have metropolitan WiMax networks with millions of users up in a year or two, WiFi mesh network builders here will still be trying to make them work.
posted by paulsc at 11:36 AM on August 1, 2006
It works for Philadelphia, except Verizon and Comcast got their lobbyists involved, stretching out the process several years longer than it should have taken, got service prices raised, and got several laws passed in their benefit to boot. So much for the taxpayer
Not only that, but every other city in PA is banned from making their own city-wide network.
One of the interesting things about the new telecom act (the one that gets rid of Net Neutrality) is that it actually bans those types of bans. So now every city will be able to setup their own wifi networks.
posted by delmoi at 1:20 PM on August 1, 2006
Not only that, but every other city in PA is banned from making their own city-wide network.
One of the interesting things about the new telecom act (the one that gets rid of Net Neutrality) is that it actually bans those types of bans. So now every city will be able to setup their own wifi networks.
posted by delmoi at 1:20 PM on August 1, 2006
My assumption is that private companies are more efficient at distributing broadband services than the government.
Chances are it'll become a public monopoly, funded by government money but administered independently, perhaps with government oversight. Or a Co-Op.
(Come to think of it, Co-Op is probably the perfect way to do this sort of thing!)
And the truly poor can't afford food, let alone computers, so they won't benefit.
Tax breaks for companies upgrading outdated machines. Basic functionality is easily achieved on a 300MHz PII machine (hell, basic functionality is easily achieved on a TRS-80!) These machines have a negative value: they cost more to keep around or to dispose of, than they are worth.
The fringe (lower middle class) won't benefit because they'll receive less of the really beneficial services from the government since government spending is being diverted to running and maintaining this service.
Most of your government spending is toward the military. Perhaps the minimal funds required to run this program can be sourced from that. Hey, a source for older computers, too!
posted by five fresh fish at 10:47 PM on August 1, 2006
Chances are it'll become a public monopoly, funded by government money but administered independently, perhaps with government oversight. Or a Co-Op.
(Come to think of it, Co-Op is probably the perfect way to do this sort of thing!)
And the truly poor can't afford food, let alone computers, so they won't benefit.
Tax breaks for companies upgrading outdated machines. Basic functionality is easily achieved on a 300MHz PII machine (hell, basic functionality is easily achieved on a TRS-80!) These machines have a negative value: they cost more to keep around or to dispose of, than they are worth.
The fringe (lower middle class) won't benefit because they'll receive less of the really beneficial services from the government since government spending is being diverted to running and maintaining this service.
Most of your government spending is toward the military. Perhaps the minimal funds required to run this program can be sourced from that. Hey, a source for older computers, too!
posted by five fresh fish at 10:47 PM on August 1, 2006
The whole "spending on x will take money away from y" is always used in political discourse as something laypeople can intuitively 'get' but is usually diversionary. I'm not saying that budgets aren't balanced by taking x away from A and spending it on B instead, but budgetary spending doesn't remain static from year to year.
posted by Firas at 12:16 PM on August 2, 2006
posted by Firas at 12:16 PM on August 2, 2006
« Older Down with television and air-conditioning! Up with... | I'm not kidding. I DO kinda feel sick right now..... Newer »
This thread has been archived and is closed to new comments
posted by Blazecock Pileon at 8:10 AM on August 1, 2006