Municipal wireless called 'telecom's Bay Of Pigs'

After tossing around ambitious plans to set up vast wireless Internet systems, a number of cities are getting out of the game.

Figuring in the costs, infrastructure headaches, and the just plain unknown aspects of wireless communication, the whole idea is being branded as anything from premature to a major fiasco, with companies and cities abandoning half-drafted, half-built systems.

In my home town, plans to build a citywide wireless network are sitting in a holding pattern, primarily because it was determined it would need about three times the number of antennas as originally estimated. The Evening Post (which owns the Charleston Post And Courier) and Mount Pleasant-based Widespread Access had spent about a half-million to get wireless going, and are cutting their losses right now. So far, according to the Post and Courier, the only thing the city had to show for the expenditure is a signal along a few streets near the Medical University of South Carolina. That's all. That network is tapped into about 200 times a day.

Financial black hole? Try "telecom's Bay of Pigs," Tim Wu of Columbia's School Of Law, said recently in a Slate magazine article. Strong words, painting a picture of a total train wreck in the making. But it's turning out to be true, so far.

Last month, the city of Charleston terminated its contract with Widespread Access, and the Evening Post has also cut its ties.

This came to mind partly because of an article in the Sacramento Bee (registration required). That town's efforts to build a data net are running into financial problems. According to consultant Phil Belanger of Novarum, it takes nearly 50 wi-fi transmitters ("nodes") per square mile. All of which costs. Sacramento was planning to do it on the cheap, with 18 to 20 nodes per square mile, according to a city official.

"It will cost a lot more than most cities had planned to build a network that provides robust, consistent indoor coverage," adds Glenn Fleischman, editor of Wi-Fi Networking News.

In Mountain View, 80 miles away, Google has built a wireless system to blanket that town. The only problem is, it just doesn't work indoors. According to the Bee, the network rarely penetrates walls. You'd have to find a spot near a window or -- horrors -- take things outside if you really want to get your geek on.

I'm extremely lucky. I live in a free wi-fi hot-spot, and I can probably wear out an industrial-strength router all by myself. Sure, the cost of the wireless access is figured into my rent, but I definitely get my money's worth here. Through my wireless connection I maintain this blog, address my news addiction, stay in contact with fellow musicians (including sending sound files and chord charts), keep in touch with my scattered-all-over-the-country family, listen to Internet radio, and chat regularly with my friend up in D.C. Plus downloading open-source programs and operating systems to feed my inner geek. OK, I probably spend more time on the Internet than a grown man should, but it's because I can.

To my experience, wireless isn't perfect yet. The connection is subject to the whims of the same forces that affect all radio-frequency waves. Your signal may -- or may not -- make it through walls. I've had to do some major rearranging to get the best connection, moving the computer right next to the kitchen. All kinds of factors affect my connection, including storms, heavy Internet traffic on the system, perhaps the phase of the moon, and probably the steel plate in my neighbor's head. I've had 24-hour stretches of down time. And, yeah, with my Linux system (I never do anything the easy way, do I?) setting up the wireless card was a contentious pain in the butt. By and large, though, I'm extremely pleased with the result.

I'm of two minds when it comes to free (well, never really free) municipal Internet access. While it's good to see such a resource become available at no end-user cost, I have major reservations. Like, the fact that a city or county government would be involved so closely in the operational side. I have yet to see a government build a better mousetrap, or a cheaper one. As soon as a government gets involved in an enterprise such as this, you can expect the total cost to be much more than what its really worth and the quality to be inferior to what you can get on the open market. And no, I'm not talking through my tinfoil hat here; this is a fact of life.

Trying to make a municipal wireless Internet system is going to be a lot like running a bus service. It's not going to make money, no matter how you dress it up. To make the system attractive at all to the consumer, it has to be as convenient as possible and as cheap as possible, an equation that does not fit in anybody's idea of a business model. I can't think of any bus service that actually pays its own way.

My idea of a perfect system will be one that's built and maintained by your friendly neighborhood hacker community, folks making it work for the love of it. Kind of an open source model. All of the software on my computer is built the same way, by a loose crew of experts doing it for the hell of it and accepting any brilliant ideas from the end user. My web browser (Firefox) is built the same way, by a bunch of hackers doing it 'cause it's fun. Unfortunately, Firefox (well, its parent company Mozilla) is going a lot more corporate these days and much is lost in the translation. Firefox is a lot more bloated than it used to be, and the newer versions just doesn't run as well on my system (though it still beats the pants off Internet Explorer). But if it was operated by a municipal government, they might even forget they're making a Web browser.

If the whole idea is to simply make the system work, the open-source model might be the most promising scenario. But it would be practically impossible to do it for love without starving to death.