Small wonder that some of America's biggest companies want to get in on the deal. On Dec. 5, AT&T (T
), IBM (IBM
), and three venture-capital firms, including Intel Capital, announced that they were backing a new Wi-Fi wholesaler, Cometa Networks (see BW Online, 12/13/02, "The Big Boys' Mad Dash into Wi-Fi"). By 2004, Cometa plans to roll out a constellation of at least 20,000 wireless-access points, known as "hot spots."
In major cities, wireless users would never be further than a five-minute walk from wireless connectivity. In the suburbs, road warriors would be a mere five-minute drive from wireless nirvana. The technology's advocates are cheering: With AT&T, IBM, and Intel (INTC
) behind Wi-Fi, the technology appears poised to move beyond its grassroots to the big time.
DOING THE MATH. Cometa Networks, though, is no slam dunk. As anyone who lived through the tech boom knows, a few big names behind a new technology don't guarantee profit. And Cometa still hasn't spelled out exactly how the business would work. It says it'll set up the hot spots and sell bandwidth to cellular companies and Internet service providers, who in turn would sell Wi-Fi service to their customers.
However, Cometa gives no details about how and where it plans to put the hot spots or how much the service would cost. In a recent conference call, Cometa executives remained hazy about how the network would be used, saying only that they were confident that the country's 100 million Internet users would find new ways to use their laptops once the network was rolled out.
Let's look at the plan. Cometa Chief Executive Larry Brilliant says 20,000 hot spots strewn in shopping malls, hotels, and restaurants will provide coverage in the 50-largest U.S. cities. The company has previously been reluctant to put a figure on the build-out cost but is now estimating it at $30 million.
SATURATING SEOUL. Cometa, though, will need far more than 20,000 hot spots to provide truly ubiquitous coverage, says Anthony Townsend, a research scientist at New York University's Taub Urban Research Center and a founding member of grassroots Wi-Fi coalition NYC Wireless. Wi-Fi signals aren't ultrapowerful. So users must be within 850 yards of a hot spot to get connected. According to Townsend's calculations, Cometa will need 250,000 hot spots to "wire" America's 50-largest cities -- more than 10 times the number Cometa has slated.
Townsend's figures pale in comparison to the density of hot spots that are being rolled out internationally. South Korea Telecom is installing 10,000 in Seoul alone. In Tokyo, NTT DoCoMo (DCM
), one of seven local Wi-Fi providers, has installed 1,000 hot spots. These rollouts are even more significant in light of the dense population of Asian cities compared with sprawling U.S. commerce centers such as Los Angeles or Dallas.
Installing even 20,000 hot spots around the U.S. won't be easy. Cometa asserts the task will be straightforward, thanks to the presence of AT&T's national network and the skill of IBM Global Services. But setting up wireless-access points will require cutting deals to gain access to thousands of privately owned office buildings, airports, and restaurants. Also, many major convention centers -- prime hot spot locations -- already have exclusive contracts with other telecom providers.
LATE START. Take Las Vegas-based telecom provider Smart City, which has sewn up exclusive deals with convention centers in cities from Anaheim to Washington, D.C. Meanwhile, early startups, such as Wayport and Boingo, have beat Cometa to the punch at airports, a key location where business travelers often remain stranded for hours at a time.
Cometa Chairman Ted Schell adds that his company has no plans to draft "roaming" agreements with other providers who have already locked up prime spots. The reason: Cometa is targeting business users and must retain full control of its network in order to guarantee quality of service.
Finally, there's the question of whether anyone can successfully charge for Wi-Fi. Access points are cheap -- hooking up the technology starts at around $200. These access points are so easy to operate and maintain that some believe hotels, cafés, bookstores, and airports will simply install them on their own. Sound crazy? Manhattan's Bryant Park has had a free Wi-Fi network in service since June. The park spends more per month on trash bags than it does to run its Wi-Fi network, according to system operators, who spend between two to five hours per month on maintenance.
ALTERNATE PARADIGM. In fact, some network experts think Wi-Fi's diffusion may end up resembling the spread of the fax machine rather than the development of a centralized communications network. Wi-Fi, like the fax machine, is a product rather than a service. The users will build the network by installing cheap wireless-access points and buying wireless cards for their laptops that allow them to log on to the airwaves. As with the fax, the more people who set up access points, the more powerful the network.
When fax machines were first introduced in the 1980s, several big companies planned new fax-delivery services. In 1984, visionary FedEx CEO Frederick Smith introduced a service called ZapMail that he hoped would replace jet fuel with ink toner. The plan: FedEx would buy the then-pricey fax machines and place them in every FedEx office. Customers who wanted to send a fax would have FedEx pick up their documents and bring them to a local office. Within the hour, the documents would then be faxed to the FedEx office closest to the recipient. FedEx would put the fax in an envelope and hand-deliver the service.
At the time, it made sense. ZapMail began as a value-added service that leveraged FedEx's core strength -- reliably delivering information overnight. It also saved customers the trouble of installing and maintaining expensive equipment. But ZapMail ultimately failed as the price of fax machines plummeted. Rather than pay someone else to send a fax, businesses just bought their own machines. FedEx shuttered ZapMail only 12 months after the launch -- and $190 million in losses.
"LAND GRAB." ZapMail may prove a cautionary tale for Cometa. Right now, Wi-Fi seems like a new, whiz-bang technology that requires corporate oversight. But in time, business users and individuals may not see the need to pay someone for Wi-Fi service. After all, bandwidth is sold at a flat monthly rate. That means there's no cost difference to a hotel, restaurant, or public park if 1,000 or 100,000 people log on to their network.
"This is a corporate land grab. Ultimately, though, users may realize they can make this work on their own," predicts Dewayne Hendricks, CEO of the California-based Dandin Group, which promotes wireless technology in remote areas. That would be good news for Wi-Fi. But bad news for Cometa. Black covers technology for BusinessWeek Online