Municipal Wi-Fi: A Promise Unfulfilled?

My first experience with municipal Wi-Fi was disappointing. Two years ago, after finding out that a nearby town had covered its downtown area with free Wi-Fi service, I walked over to a coffee shop, ordered a latte, and powered up my PowerBook. No signal. I fiddled around with the Internet Connect utility a bit. No signal. I looked around, and noticed that there were some people working on laptops, and that most were sitting near the windows. I wasn't. So I walked over to one of these people and asked, "Are you able to get onto the Internet from here? I'm not getting a signal."

"Not right now," she said, "but sometimes if you sit right next to the window you can pull in a signal."

The town where that happened, Carrboro, NC, is one of more than 400+ U.S. towns, cities, and counties that have a municipal Wi-Fi project in some stage of development. To be fair, Carrboro, a densely populated town of 17,000 that borders Chapel Hill, isn't very boastful about its downtown Wi-Fi coverage; it's difficult to find a reference to it on the town's Web site. "Our plan has always been to build it out modestly," says Carrboro Mayor Mark Chilton. "We've taken a very limited approach." The town spends about $10,000 a year on the network, with most of that money going to add another access point. Free Wi-Fi coverage is important enough to merit a mention or two in the town's 2006 "Economic Future" report. On the other hand, the network's official Web site hasn't been updated in more than four years.

"We don't want to overbill it," adds Chilton, who emphasizes that while municipal Wi-Fi was never intended to draw businesses to the downtown area, it does serve many in Carrboro's creative community, who use the public network to work outside on nice days.

The Dream Of Muni Wi-Fi
In some ways, Carrboro is representative of most U.S. towns and cities that have some sort of free municipal Wi-Fi network in the planning stages, already partially in place, or fully deployed -- usually in a limited fashion, or in a state of political or economic limbo.

Many people have heard of municipal Wi-Fi, either by glancing at a trade press headline or, if they live in one of the cities planning or implementing such a system, through the local media. It's the locals who know the most -- because city politicians first spoke directly to them, believing they would be the great (and grateful) beneficiaries of such systems, which promised free Wi-Fi access throughout a metropolitan area, and low-cost (or advertising-supported) high-speed Wi-Fi connections to houses and apartments.

Finally, you'd be able to sit in a park on a sunny summer morning, pull out your laptop, and get some work done without missing the good weather. You'd be able to replace your pricey cable or DSL connection with Wi-Fi that would be just as fast -- at about half the cost.

But recently some big, heavily-touted muni Wi-Fi initiatives, most notably in Chicago and San Francisco, have been cancelled, having been felled by a number of problems. The consultants and analysts I spoke with agreed that local politicians, seeking voter support, latched onto the idea a few years ago and made commitments prematurely. "It was a politician's dream," says consultant Craig Settles, "a 'chicken in every pot' type of promise." Do People Really Want It?
One rub is that demand for free, or low-cost, citywide Wi-Fi has been weak or nonexistent. Most who wanted or needed home broadband access already had it through a cable or DSL line, and weren't clamoring for price decreases, especially if it meant switching to an untried Wi-Fi network. Similarly, there was no great outcry for the ability to access the Web while sitting on a park bench. It might be a nice thing to do once or twice a year, but in many parts of the country it's too hot, too cold, or too wet to work outside most of the year. And free, cozy Wi-Fi access is rarely far away these days, as many cafes now provide it as a way to draw in, and keep, customers.

In addition, while many of the problems that municipalities face are small, even trivial (the "last few feet" hitch that I ran into in that cafe just needed a simple wireless bridge that costs between $60 and $90 dollars), it is just one of the many problems muni Wi-Fi faces nationwide.

The savviest modern politicians talk about, and actually spend much of their time, listening to people, both in person and via polls and surveys, to try to find out what's needed and what's wanted. They're following leadership wisdom that's been passed down for thousands of years. "If you want to lead the people, you must learn how to follow them," reads the Tao Te Ching.

But many local politicians didn't keep that in mind when they made their municipal Wi-Fi promises and plans. "They tried to shove the technology down people's throats," says Settles, sharply articulating the consensus among those currently watching -- or involved in -- muni Wi-Fi projects. "In many places it was wrong for what people wanted or needed to do. But there are votes in grandiose thinking."

Wireless Philadelphia
The biggest muni Wi-Fi success story, both in terms of the number of people and geographic area it covers, and the publicity it's received, is Wireless Philadelphia. EarthLink has built about half of the network, and aims to complete the basic installation by the end of 2007.

The network is entirely owned and operated by EarthLink and currently covers about half of the city's 135 square miles. Many have praised Philadelphia for making home access free or low-cost to the city's poor (the cost for most users, after an initial promotional price of $6.95/mo, will be $19.95 for home access), but others have criticized the network for not being strong enough to penetrate walls or reach upper floors in skyscrapers. Putting in a few wireless bridges is easy, even for a small town like Carrboro. Philadelphia, on the other hand, will eventually need to plant thousands of the devices to make its consumer offerings attractive. The change in scale makes a difference, both logistically and financially.

The Philadelphia "success story" has been costly for EarthLink, which announced record losses last summer and laid off 900 employees, a move that proved costly to its municipal Wi-Fi division. In addition to Philadelphia, EarthLink is in the process of putting together Wi-Fi networks for New Orleans, Milpitas, Calif., Corpus Christi, Texas, and Anaheim, Calif., and is continuing negotiations with two Virginia municipalities. But otherwise, the company has put the brakes on muni Wi-Fi. "We will not devote any new capital to the old municipal Wi-Fi model that has us taking on all the risk by fronting all the capital, then paying to buy our customers one by one," said EarthLink President Rolia Huff. "That model is simply unworkable."

Wi-Fi For Municipal Services
What appears to be workable, says Jerry Grosso, a spokesperson for the EarthLink Municipal Networks division, are systems like the one in Corpus Christi, where municipal services are the backbone. These aren't sexy stuff that holds out the promise for a lot of votes come Election Day, but rather support systems for mundane -- but critical -- city services. Some police and fire departments have already benefited from municipal Wi-Fi buildouts, which enable them to communicate more quickly and reliably, file reports remotely, and gather detailed information while in the field. Other municipal Wi-Fi services promise greater efficiency for field workers who do things like fix sewer and power lines, read water and power meters, and carry out other day-to-day tasks. Some services promise greater revenue via such innovations as Wi-Fi-enabled parking meters.

But many cities, including Philadelphia, haven't been jumping at such offerings. So EarthLink has to finish "Wireless Philadelphia" before it can really make an effort to recover its capital expenses, says Grasso. When it's fully built out, EarthLink will increase efforts to sell a small variety of services. The company plans three offerings: a standard consumer offering ($19.95 per month for speeds comparable to DSL); an occasional use product similar to T-Mobile's plan, which would enable users to buy network access by the hour, day, or in three-day chunks; and an enterprise solution for businesses, government units, and universities. No advertising is involved in any of these ventures.

Despite the reluctance of Philadelphia and other cities, making municipal services the "backbone" of muni Wi-Fi -- with the municipality paying the lion's share of operating expenses -- may be the only workable model, says Gartner networking and communications equipment analyst Ian Keene. "If you look at practical applications for the municipality itself, (municipal Wi-Fi is) alive and kicking. Lots of field employees say they can save one or one and a half hours a day with wireless connectivity."

Still not excited? Westminster, in London, is installing Wi-Fi-enabled security cameras that can identify illegally parked cars and issue tickets without an on-site witness. In theory, the number of parking tickets should increase dramatically without much additional cost, and city coffers will swell. "Parking enforcement is the killer application that everyone is looking for," a Westminster City Council member said in early September. That's not the type of thing of thing U.S. city politicos were boasting about a few years back. The Nuts And Bolts Of Muni Wi-Fi
Because of the enormous geographic area that muni Wi-Fi networks have to cover, and the huge number of people (and surveillance devices) that access the network, the hardware and software setups bear little resemblance to home or cafe Wi-Fi networks. To a large extent, the hardware devices share the same radio bandwidth; 802.11b/g protocols utilize the 2.4 GHz band, which is unlicensed; ham radio operators, microwave ovens, wireless phones can share this spectrum, causing interference for consumers. But police and fire departments and other public safety officers have access to the 4.9 GHz band, and usually use equipment that enables Wi-Fi communication using that licensed (and more secure and reliable) bandwidth.

The networks use intelligent "mesh" technology that requires between 30 and 100 hardened outdoor access points (sometimes referred to as local gateways) per square mile; each of these access points is usually equipped with multiple antennas (many are equipped to transfer data using the 2.4 GHz, 4.9 GHZ, and 5.8 GHz frequencies). Access points can gather data that enables them to make intelligent decisions, using embedded firmware, about data routing.

This strong protection and intelligence is costly. MetroFi is a company that is working on muni Wi-Fi rollouts in nine West Coast cities, including Portland. According to Lou Pelosi, its vice president of marketing, each access point costs between $4,000 and $6,000, not including installation, which is also expensive. MetroFi buys its access points from Sky Pilot, Motorola, and BelAir Networks, says Pelosi.

Pelosi believes that as more municipal Wi-Fi plans migrate from paper to poles, costs will come down with volume. But Keene disagrees. "The price probably won't come down even if one out of every three cities put Wi-Fi mesh networks into place," he argues. "This won't bring huge economies of scale. And maintaining the nodes can be expensive, because interference from other devices using the same bandwidth requires modification." While some access points are capable of scanning for interference and adjusting accordingly, most aren't. And, Keene adds, many muni Wi-Fi failure stories have come about because of lowball estimates on the number of nodes required per square mile. Philadelphia, for example, initially estimated the network would require 24 nodes per square mile, but it now averages 48 nodes per square mile, according to a Novarum survey. Toronto boasts a node density topping 100 per square mile.

Novarum, a broadband consulting firm, recommends that cities planning muni Wi-Fi begin with an estimate of 50. Keene agrees that the number of access points has been a problem in municipal Wi-Fi planning; a low density of nodes is less expensive, but it also results in a noticeably less reliable and slower network. Because many muni Wi-Fi models are banking on wide consumer adoption (they need either lots of paying customers or lots of eyeballs, in the case of MetroFi's advertising-supported model), high speeds are necessary in competition with other options such as DSL and cable modems. The Novarum survey determined that Toronto boasts speeds of 2.2M bps downstream and 1.6M bps upstream, while Philadelphia averaged 1.5M bps downstream. Such speeds are competitive with wired systems.

In a mesh network, an individual computer sends its signal to a local gateway; this gateway then sends and receives signals from a broadband aggregation point, which could be fiber optic cable, microwave, or a combination of the two. The aggregation point connects to a network interconnect facility, which is hard-wired to the Internet.

What's Next? The Best Guess
Thus far, there have been a few true municipal Wi-Fi success stories and several spectacular failures. But more than half of municipal Wi-Fi networks remain only in the planning stages. The broad consensus among analysts and providers is that the only viable business models will be centered around municipal government applications, which appear to be able to provide cities with the ability to provide both better and more cost-efficient services for residents and increase city revenue. This will ensure that providers like EarthLink can recoup their capital costs within a few years. Meanwhile, the Wi-Fi standard, while established, continues to evolve, and it appears that in the near future it will coexist with WiMax, a wireless protocol that can blanket large geographic areas and thus reach even the most rural areas in the U.S.

While the pessimists point to Wi-Fi's collapse in Chicago and San Francisco as clear indicators that muni Wi-Fi isn't a viable concept, optimists point out that it's still the early days for municipal Wi-Fi, and that some high-profile failed ventures indicate nothing more than poor planning on a large scale. The optimists point to the hundreds of U.S. cities and technology companies invested in it, in some fashion, and say it's not only viable for many metropolitan areas, but inevitable. "In September," says Settles, "everyone was getting out of the business, but by January some are going to reconsider. The question is, how many of those cities will be moving forward come January? We're going to have a shakeout, with some pursuing it aggressively, some cautiously, and some giving up on the idea altogether. But the one thing that seems certain is that the people who are still at the table in January will have a much more sound vision than they did just a year ago."