Municipal WiFi

[August1991]

This was discussed in the thread about why we should have the CBC.

The City of Toronto is proposing a city-wide wifi network.

Transmitting and receiving units on lamp posts would let people with wireless-ready computers and other gadgets connect almost anywhere, creating a city-wide version of the wireless zones available at certain coffee shops and hotels.

It would also let people use wireless connections at home, short-circuiting phone and cable companies that sell internet access.

There would be a price, however.

Last week, Philadelphia announced a deal with a commercial provider, EarthLink Inc., for construction of a network using 4,000 municipal light poles to cover 350 square kilometres.

EarthLink has agreed to provide 22 free wireless hot spots in tourist areas and other locations while offering discounted connections to poor families at $9.55 U.S. a month.

CBC

In theory, there could be several private networks and they could could compete for customers. But it usually makes no sense to have more than one network and this requires either State-ownership and provision or State-regulation. A fundamental problem is that it is usually costless to offer the service to one more customer.

We are a few short years away from having essentially free wireless service in urban areas for both computers and voice communication.

As an institution, governments were created precisely for providing this kind of service.

[Renegade]

Rogers Communications and Bell Canada have agreed to share spectrum to support a broadband wireless network capable of reaching more than two-thirds of Canadians.

The two companies will share their tower and network transport infrastructures. However, marketing and delivery of services over the network will be a separate concern. In other words, the two companies will potentially compete for customers.

Rogers and Bell Canada Ink National Broadband Wireless Deal

Actually its not true that state ownership is the only viable way to go. The story above proves that competitors can also cooperate when it is in their interest. In this case it is in both of their interest to build one network and share bandwidth however still sell it separately. No state ownership or government intervention is necessary.

[August1991]

Actually its not true that state ownership is the only viable way to go. The story above proves that competitors can also cooperate when it is in their interest. In this case it is in both of their interest to build one network and share bandwidth however still sell it separately. No state ownership or government intervention is necessary.

I never said that state ownership is necessary, I merely argued that we are likely to get less service at higher cost with a private provider.

WiFi and broadcasting both have high installation costs but then it costs essentially nothing to offer the service to a new customer. No private provider would ever charge nothing for the service.

Admittedly, state ownership poses its own problems but I think frankly that these are the kinds of problems we must overcome in this century.

Toronto is imitating Philadelphia. In the case of Toronto, a 6 km square zone in the centre of the city will offer Internet service using hydro poles.

We rely on government to build roads, and for exactly the same reason, we should rely on government to provide Internet access.

I would dearly like to see governments charge road-users for the congestion they incur when driving. London England has started, in a very primitive way, to do this.

[Riverwind]

WiFi and broadcasting both have high installation costs but then it costs essentially nothing to offer the service to a new customer.

August, adding users to a WiFi network is not zero cost - in fact it can be quite costly depending on how much they use the network. Wireless LANs have limited bandwidth and if too many users try to use the service then the service will become useless to everyone. Adjusting to different levels of demand in different areas of the city would require constant maintenance and upgrades to the network. Furthermore, wireless LANs are often subject to interference and 'deadzones' which will deny the service to some people because it would cost too much deploy an extra router or two to clear up the deadzone.

In the end, Toronto's WiFi network will likely be too slow and unreliable to provide efficient service for most people, however, the presence of the publicly funded network would make the private services more expensive because the potential subscriber base is smaller due to the public network.

Lastly, any enterprise run by the gov't is likely to be infested with inflexible unions that will drive up the cost of providing the service and lowering the quality.

I think gov'ts should stay out of the business of providing IP service just like they stayed out of the business of providing phone service.

[lost&outofcontrol]

August, adding users to a WiFi network is not zero cost - in fact it can be quite costly depending on how much they use the network. Wireless LANs have limited bandwidth and if too many users try to use the service then the service will become useless to everyone. Adjusting to different levels of demand in different areas of the city would require constant maintenance and upgrades to the network. Furthermore, wireless LANs are often subject to interference and 'deadzones' which will deny the service to some people because it would cost too much deploy an extra router or two to clear up the deadzone.

In the end, Toronto's WiFi network will likely be too slow and unreliable to provide efficient service for most people, however, the presence of the publicly funded network would make the private services more expensive because the potential subscriber base is smaller due to the public network.

Lastly, any enterprise run by the gov't is likely to be infested with inflexible unions that will drive up the cost of providing the service and lowering the quality.

I think gov'ts should stay out of the business of providing IP service just like they stayed out of the business of providing phone service.

You can easily limit the amout of bandwidth per user stopping people's ability to leech off the majority of the bandwidth. Your strawman defense is weak; the services will suck so we shouldn't do it. Those damn unions ! We already pay on average more for internet access than many coutries, the telcos are always trying to screw us out of a bit more. vonage files complaint against shaw

[Riverwind]

You can easily limit the amout of bandwidth per user stopping people's ability to leech off the majority of the bandwidth.

Sure, but eventually you have so many users that the available bandwidth per user is useless.

Your strawman defense is weak; the services will suck so we shouldn't do it.

Whenever the gov't gives something away for free the demand for the service always outstrips the money available and the gov't is forced to reduce quality to control costs. We see this health care and we see this roads - we will see the same behavior in publicly provided ISPs. The net result is the networks will cost tax payers a lot of money and people will always complain about the poor service.

We already pay on average more for internet access than many countries, the telcos are always trying to screw us out of a bit more.

Regulating companies with near monopolies on important services is a job that the government should do - the problem does not go away if a few neighborhoods in downtown T.O. get free wireless. That said, you facts are wrong. Internet access in Canada is considerably cheaper than in places like Europe and Japan. The link you provided is not no a problem unique to Shaw - the US cable cos are trying to do the same thing - it is up to gov'ts to slap them down with proper regulations.

[August1991]

August, adding users to a WiFi network is not zero cost - in fact it can be quite costly depending on how much they use the network. Wireless LANs have limited bandwidth and if too many users try to use the service then the service will become useless to everyone. Adjusting to different levels of demand in different areas of the city would require constant maintenance and upgrades to the network.

If congestion became a problem at certain hours of the day, then the city could consider imposing user fees. I have a suspicion though that technology will quickly overcome this congestion problem. In any case, my argument in favour of public provision of Internet access does not turn on whether network is congested or not.

Furthermore, wireless LANs are often subject to interference and 'deadzones' which will deny the service to some people because it would cost too much deploy an extra router or two to clear up the deadzone.

In the end, Toronto's WiFi network will likely be too slow and unreliable to provide efficient service for most people, however, the presence of the publicly funded network would make the private services more expensive because the potential subscriber base is smaller due to the public network.

These are technical problems that a private provider would also face. Your last argument supports my view that a single provider makes sense. Admittedly, it could be a private firm, regulated by the government, but I suspect that that would cause worse problems.

Wireless Internet access is fundamentally different from other wire networks such as cable-TV, hydro or line telephones. It is more similar to broadcast radio or TV, but without the ability to include advertising revenues.

Lastly, any enterprise run by the gov't is likely to be infested with inflexible unions that will drive up the cost of providing the service and lowering the quality.

I think gov'ts should stay out of the business of providing IP service just like they stayed out of the business of providing phone service.

I believe that Alberta government operated telephone service in Alberta. As to unions, it is precisely this kind of problem that we must face in this century.

Governments have an important role to play in an economy (indeed in society) and we must discover a better way to make this happen. The old style Leftists are just as wrong as the Libertarian Right.

Whenever the gov't gives something away for free the demand for the service always outstrips the money available and the gov't is forced to reduce quality to control costs. We see this health care and we see this roads - we will see the same behavior in publicly provided ISPs.

In fact, we don't see this with neighbourhood roads or long distance highways between cities. (OTOH, Montreal streets are in a terrible condition because of potholes. This is a problem of our political system.)

Regulating companies with near monopolies on important services is a job that the government should do - the problem does not go away if a few neighborhoods in downtown T.O. get free wireless.

I'd be willing to consider that option if it worked better.

----

Sparhawk, I think my main point is that a person should not have to pay to obtain "normal" wireless access to the Internet in an urban area. Why? Because once the network is installed, it costs society nothing to provide that person with access. If a person faces a charge to obtain Internet access, they may choose not to use the Internet and society will be worse off. I was making a similar argument concerning culture, ideas, the CBC and broadcasting in another thread. We have too little art and artistry and too much pollution and environmental damage. This happens because our governments fail us.

Socialists want our governments to do everything and Libertarians want our governments to do nothing.

No one has seriously considered what governments should do (except the federal Liberals which have governments do whatever is expedient). Two cases in point are post-secondary education and health care. Governments need "barely" be involved in such activities.

[Riverwind]

Sparhawk, I think my main point is that a person should not have to pay to obtain "normal" wireless access to the Internet in an urban area. Why? Because once the network is installed, it costs society nothing to provide that person with access. If a person faces a charge to obtain Internet access, they may choose not to use the Internet and society will be worse off.

Every you say depends on your assumption that adding additional people to an existing network costs nothing. That is an invalid assumption based on my knowledge of network technology. However, I agree with most of your arguments provided there is a usage based fee imposed on all users. This fee need not be large nor does it have to recover all costs but it does have to large enough to remind people that the shared resource is finite.

[geoffrey]

Sparhawk, I think my main point is that a person should not have to pay to obtain "normal" wireless access to the Internet in an urban area. Why? Because once the network is installed, it costs society nothing to provide that person with access. If a person faces a charge to obtain Internet access, they may choose not to use the Internet and society will be worse off.

Every you say depends on your assumption that adding additional people to an existing network costs nothing. That is an invalid assumption based on my knowledge of network technology. However, I agree with most of your arguments provided there is a usage based fee imposed on all users. This fee need not be large nor does it have to recover all costs but it does have to large enough to remind people that the shared resource is finite.

Adding additional users would be free up until a point. Eventually you have to expand the infrastructure. It's like a road. The first thousand users a day don't influence each others use, but after that, you get traffic jams and there is a definite cost to adding an additional user to the roadway. It's the exact same principle in shared Wi-Fi.

Not to mention, there would obviously be freeloaders, such a business using it for profitable means. A data related business could easily consume an entire Wi-Fi network, it's a tradegy of the commons concept, especially if it was completely free, unregulated access.

It would become a common resource, not a public good.

[August1991]

However, I agree with most of your arguments provided there is a usage based fee imposed on all users. This fee need not be large nor does it have to recover all costs but it does have to large enough to remind people that the shared resource is finite.

On the contrary, I think there should be no usage-based fee (unless the user imposes congestion costs).

"... remind people that the shared resource is finite." WTF? We don't impose taxes or fees for PR reasons. (And if governments do, that's a sad demonstration, like the sponsorship scandal, of a government's extensive power.)

OTOH, there must be a "fee" to pay for the infrastructure. A city in Canada that is considering a WiFi network will have to build it. That costs money. In general, the money will come from property taxes and other municipal revenues. Some people will pay more than others. Well, that's life.

Adding additional users would be free up until a point. Eventually you have to expand the infrastructure. It's like a road. The first thousand users a day don't influence each others use, but after that, you get traffic jams and there is a definite cost to adding an additional user to the roadway. It's the exact same principle in shared Wi-Fi.

True, but I think that technology will solve this problem. How much bandwidth can one person use in one second?

Not to mention, there would obviously be freeloaders, such a business using it for profitable means. A data related business could easily consume an entire Wi-Fi network, it's a tradegy of the commons concept, especially if it was completely free, unregulated access.

Businesses using the Internet are freeloaders? WTF?

This is not a "tragedy of the commons" concept; rather, it's a "miracle of the fishes" concept. Land (the commons) is limited but has no owner. WiFi is unlimited and needs no owner.

[geoffrey]

Businesses using the Internet are freeloaders? WTF?

This is not a "tragedy of the commons" concept; rather, it's a "miracle of the fishes" concept. Land (the commons) is limited but has no owner. WiFi is unlimited and needs no owner.

Easily. It's not hard to overwelm a Wi-Fi network with very little traffic. Our multi-million dollar Wi-Fi throughout the university is regularly bogged down and slow due too excess traffic. Definitely a tradegy of the commons, as bandwidth is limited and has no owner.

To increase the bandwidth is incrediably expensive past a certain point, laying fibre optic lines isn't cheap at all.

[Riverwind]

On the contrary, I think there should be no usage-based fee (unless the user imposes congestion costs).

That is my entire point. University networks are already having problems with students running extremely bandwidth intensive applications like VOIP or BitTorrent. Any free network must have fees that are connected to the amount of data a user transfers.

A tiered fee structure that only affects users that abuse the resources would also be effective.

[lost&outofcontrol]

Is there an estimate for the project ?

[lost&outofcontrol]

That is my entire point. University networks are already having problems with students running extremely bandwidth intensive applications like VOIP or BitTorrent. Any free network must have fees that are connected to the amount of data a user transfers.

A tiered fee structure that only affects users that abuse the resources would also be effective.

Why not just put a cap(256Kb/s) on the free service and uncap(lets say 1.5Mb/s) it for the ones willing to pay a small fee.

[August1991]

It's not hard to overwelm a Wi-Fi network with very little traffic. Our multi-million dollar Wi-Fi throughout the university is regularly bogged down and slow due too excess traffic. Definitely a tradegy of the commons, as bandwidth is limited and has no owner.

I wasn't thinking about a university wi-fi.

What if an individual is watching a movie in real time, downloading another movie and talking on a cell-phone at the same time? And what if a business is receiving multiple access requests?

IOW, the university example is simplistic. It is like saying that a university in 1880 built a telephone line between two buildings and so many people wanted to use it that the university decided to abandon a telephone network.

Networks are unlike personal relations, and broadcast relations are unlike networks. WiFi Internet is something else again.

I frankly think that technology will solve these constraints shortly. Access to the Internet will be like breathing air. In theory, there could be congestion but in practice there won't be. (Well, fresh air is not free.)

If there is congestion because of heavy users (like Inco uses the air in Sudbury) then specific solutions are possible.

----

Returning to the thread's topic, a new user who logs on at 3 am to an installed network in a city costs society nothing. Hence, that user should pay nothing to access the Internet.

Any fee will wrongly discourage the user from seeing what the world has to offer. It would be like preventing a microbe, 800 miliion years ago, from having access to oxygen.