Municipal Wi-Fi Networks In Trouble

imamac writes "According to an AP story, municipal Wi-Fi is going nowhere fast. A think tank research director quipped, 'They are the monorails of this decade: the wrong technology, totally overpromised and completely undelivered.' Subscriptions to the services are much lower than expected and lawmakers are concerned that millions of dollars will have gone to waste that could have been better spent on roads or crime-fighting. Satisfaction with the quality of service has also been low, which give some insight into the low adoption rate. Is municipal Wi-Fi just a bad idea, has it been poorly implemented, or is the technology just not there to support such an endeavor?"

Comments

But if we don't have it

[Anonymous Coward]

The WiFi will go to Shelbyville!

A think tank research director quipped

[Ice Station Zebra]

"I'll say anything if you give me money."

WTF?

[rsilvergun]

As a tech, I'm dying for these things. I'm getting more and more wireless networks where it just doesn't work because there's too many people with wireless devices in the area. I had one house with 6 wireless networks in range, cell phones, wireless security systems, 2.4 Ghz wireless on the land line, and even a few wireless mice and keyboards floating around. It was too much.

At the risk of being repetative

[zappepcs]

Let me say this, Metropolitan networks, whether Wi-Fi or otherwise need one thing to make them both competitive and financially viable; the metropolitan network needs to be owned by that cooperative body within the municipality's control. That means every last 'last mile' connection.

When the city/county (whatever) owns all the last mile physical plant/infrastructure and ISP's simply rent connectivity to end users the municipality will be functional and profitable. Yes, that is how we would see big bandwidth to every home, and each home would have the choice of ISP services. It is possible to do this and would instantly flatten the cost of entry as well as the rules of engagement.

Then, if you ad Wi-Fi support to parts of the city that is subscribed to by users who already pay... well, it's not such a stretch to support financially.

Does anyone see any downsides to this?

Google's Wifi

[James_Aguilar]

Google's wifi here in Mountain View is not very good. I can't get any reception on it, and I live less than a mile from their headquarters. If even Google can't get it right, city governments probably . . .

The rest of the above sentence is left as an exercise for the reader.

The major issue

[Shaman]

The major issue has been that they have given the contracts to implementors that are paid for the number of radios that they install and by gosh they will install more radios than anyone every imagined. But, see, the 2.4Ghz bands were already polluted BEFORE they started and installing 2.4G radios on every block for several square miles when each mesh radio has a practical range (line of sight) of around 20 miles is really not helping things. And just as bad, the backhaul of the mesh radios is almost always 5.2Ghz or 5.8Ghz, which have only a few channels each to choose from (5.8Ghz has more, but still...)

Don't believe this could happen? Ask anyone that has tried to use the Toronto mesh network downtown. It's flat ugly.

Re:The major issue

[Shaman]

I operate a fixed wireless ISP with 40 towers and over 1,500 (deeply) rural customers. Just so we're clear on the level I'm speaking at.

Actually, you have several things wrong. Mesh radios CAN be set to low power modes but invariably they are not. They are set to blast at or near full power because nearby interference causes issues that only power output can solve. Sectorizing only solves so much. But even those that aren't set up that way still exhibit many issues. At a full 36db EIRP, 2.4Ghz will indeed go 20 miles line of site and beyond, if the noise floor is low enough and the radio is high enough. 5.2Ghz cannot use reflectors and only has a useful range of a few kilometers, but it's the lowest power of the available bands.

Take a look at the 2.4Ghz backhauls that go over 40 miles with standard EIRP. Not that PCMCIA cards will power that far, but the A.P.s will. One company makes a product that claims 216Mbps full duplex over 20 miles, in fact.

So the question to you: If mesh gear worked so well, why is everyone having trouble with them?

As for interference... 5.8Ghz noise levels are horrendous around here, 2.4Ghz is only good for backhaul links for towers that are way out in the middle of nowhere, for multipoint it's nearly unusable, and 5.2Ghz is moderately noisy as well. I'm hoping the 5.4G and 4.9G radios will be available really soon because I need them. Speaking of that, my damn Motorola OFDM radios still can't be set to 4.9G even though it says right on the box that they support that band.

Then there's 900Mhz... the interference in the top of the usable unlicensed band made it unusable and if two WISPs in an area decide to use 900Mhz, they'll both lose... and the beat goes on.

The only real way out of the mess is to go with proprietary WiMax type products, and if you see another one of my posts, that's not a completely infallable solution, either.

Re:the answers:

[SQLGuru]

The reason that subscriptions are so low is that they've concentrated on the business districts.....

1) Most businesses have their own network (which, BTW, is faster than the service provided)
2) Most CBD's are "vacant" during the evening when individuals would be using it.
3) It doesn't make it to the 'burbs where I live.

Layne

Worst since the monorail eh?

[bizitch]

I've sold municipal wi-fi to Brockway, Ogdenville, and North Haverbrook, and by gum, it put them on the map!

Re:Harry Browne said it best...

[spun]

Uhhhh, I am an anarchist. An anarcho-syndicalist. We think your concept of individual ownership of natural resources will automatically lead to an oligarchy of a few owning class people. We believe in democratic control of natural resources by the workers actually using those resources.

Government is a control structure, and like any it can be abused. That includes the control structures of private ownership, which aren't as efficient as you may think. In studies of privatization, privatization of competitive industries works well, while privatization of natural monopolies has always failed.

With government, there is a system of checks and balances. In the free market, there are no checks and balances to curb the runaway positive feedback loop of wealth accumulation. There are no checks to stop the exploitation of the natural failure modes of the free market: information imbalance, natural monopoly, and externalities. I have yet to hear a Libertarian give a cogent explanation of how their system would deal with those three factors.

You accuse me of not understanding Libertarianism, I accuse you of not thinking through the consequences.