We all know that Wi-Fi is big and getting bigger. Indeed, in the States, Wi-Fi access is becoming as common as a cafe latte thanks to businesses like Starbucks, which have installed T-Mobile 802.11b hotspots. How big is Wi-Fi? According to Gemma Paulo, In-Stat/MDR senior analyst, about 21-million Wi-Fi cards will ship worldwide this year, and she expects 33 million to ship in 2004. Allied Business Intelligence's (ABI) senior analyst Tim Shelton predicts that the number of WLAN enabled notebook users will be up to 58 million users by 2008.

Simultaneously, wireless telephony networks using 2.5 and 3G offer smart phone users a sub-set of TCP/IP networking services, typically Web browsing and e-mail, but with far broader coverage than hotspots. Unlike Wi-Fi, which is exploding, hard data on 2.5 and 3G is hard to find since contract details and market share are often only described in the most general of terms. What we do know though is that, according to the Dell'Oro Group, that in first quarter of 2003 3G mobility infrastructure revenues declined by 14% since the last quarter of 2002 and that in 2002 the entire mobility telephony data infrastructure shrank by 16%.

2.5G networks using General Packet Radio Service (GPRS) can theoretically have data transfer rates of up to 114Kbps. In practice, horrid transfer speeds of 400 to 1000 bytes/sec are par for the course. While Wi-Fi vendors also make speed claims based on misleading top data signal rates, for example 802.11b is commonly cited as having 11Mbps throughput, their real world throughput of 4 to 6Mbps is much higher than 2.5 and 3G technologies and on Wi-Fi the full array of TCP/IP network services are at your beck and call.

Expert Opinions

"The WLAN industry will continue to experience stellar growth as deployments in several key markets take place," predicts ABI senior analyst John W. Chang. Some of that growth will come at 3G's expense. Specifically, "As WLAN moves toward 54 Mbps, it is apparent that 3G cannot compete with the data rate of WLAN. Though 3G will be deployed worldwide due to its voice capacity benefits, telecom carriers are seeing WLAN hotspots as the immediate revenue generator for data services."

3G handset vendors aren't about to throw in the towel though and it's easy to see why, "Commercial deployments of sophisticated, high-end devices mean higher margins across the entire value chain," explains Tim Shelton, also a senior analyst at ABI. For example, Nokia's latest handset, the Nokia 6650, includes advanced features such as a built-in camera, and an applications processor that requires 2.5G bandwidth or higher and, more to the point, "These features boost margin (and enable) the handset OEMs to sell the devices at a premium. Further, armed with these handsets, wireless subscribers will be poised to consume enhanced services from operators, with a downlink speed of 384 kbps, thereby allowing operators to boost their Average Revenue per User (ARPU).

Still, infrastructure build-out costs are getting in high-speed data telephony's way. 3G base stations price start at six figures and move up from there while, according to Art Tyde, COO of Sputnik, a Wi-Fi software company, a Wi-Fi hotspot can be set up for as little as $1500.

Julie Ask, Jupiter Research senior analyst, explains further, "In Europe, 2.5G data services have some traction, 3G spectrum has been allocated and paid for, 3G deployment has begun, and Wi-Fi growth lags the U.S. At this point it is far more likely that broadly deployed data services will roll out over ubiquitous 3G networks vs. Wi-Fi hotspots.

"In the U.S., 2.5G data services have little traction -- and our research indicates there's also little consumer demand, 3G is barely on the roadmap for most of the carriers, and networks of Wi-Fi hotspots are emerging. Carriers are investing in Wi-Fi more as a defensive position than clear expectation of revenue since build out costs are in the low hundreds of millions for Wi-Fi versus billions for 3G."

One of the reasons for this is that Wi-Fi can use existing telephone company infrastructure. For example Bell Canada has started converting underused telephone booths into hotspots. Other telephone companies, like British Telecom (BT) and Verizon, are also working on such conversions.

For users, cost may be the deciding factor. While it's much easier to get a data connection with a handset thanks to 2.5/3G's greater range, the cost tends to be much higher for much lower bandwidth. Indeed, public Wi-Fi may yet end up largely being free.

Kenneth L. Dulaney, Gartner's Mobile Computing VP thinks, that companies shouldn't build "hotspots to make a profit. You do it for customer service reasons (hotels, Starbucks, etc.) or for cost offset."

Ask agrees, "Return on (hotspot) investment will be driven from vertical/business applications."

In-Stat/MDR Senior Analyst for Wireless Component Technology Alan Nogee wonders "if any kind of fee-based hotspot system can survive. Customers aren't willing to pay a high usage fee for services, especially when so many free hotspots exist. If service fees rapidly drop, and roaming agreements could be put in place, some (pay) hotspots may survive, but the rest will be free."

Sputnik's Tyde says that hotspots are already becoming "like a utility. In every new technology park or business hotel, people are beginning to expect Wi-Fi the same way they now expect the phone and broadband."

Still, Dulaney thinks the technologies can co-exist. "They are not competitors. They are compliments. You use one in the wide area sense and the other (WLAN) in the local area sense."

Next Page:

2.5G/3G and Wi-Fi in one Device?