In the other corner is new technology called WiMax, which is being pushed heavily by Intel. WiMax promises to be many times faster than the Verizon-style approach. The downside of WiMax is that it will require the construction of a new network of transmission towers, something much more expensive than the simple upgrades the cell companies are doing. What's more, WiMax is still a few years out.

(There's a third approach called Flash-OFDM, but it's too soon to tell how much traction it will get in the market.)

One of the below-the-surface issues in this battle is the emerging corporate rivalry between Intel and Qualcomm, the San Diego wireless-technology company. All of the phone-based approaches depend on a key Qualcomm technology known as CDMA, on which Qualcomm has a design monopoly analogous to the one Intel has with its Pentium microprocessor. The research outfit iSupppli, for instance, estimates that Qualcomm gets roughly $10 in royalties for every cellphone that uses CDMA, which includes many of those in the U.S.

With Qualcomm already a junior version of Intel, the last thing Intel wants is for Qualcomm to have even more strategic power than it already has. But that is exactly what will happen if phone networks begin to be commonly used for the sorts of data applications that have always been Intel's forte.

Thus, one of Intel's motivations in pushing WiMax is to keep Qualcomm in check, and to allow Intel to play the same dominant role in the emerging world of wireless computing that it now plays on the desktop. And so, as this market starts to grow, watch as these two companies maneuver around each other. It's not yet Apple vs. Microsoft, but give it time.

The second bit of handicapping about wireless computing involves not who might win the battle for the market, but whether there is in fact much of a market to win in the first place.

One of the fondest dreams of observers of the technology scene is that there are legions of would-be Web surfers willing to pay good money to be able to go online anytime, anywhere. But this may be an instance of psychological projection. The sorts of people who engage in this sort of punditry tend to be online addicts, and assume the rest of the world is as well.

Questioning the popularity of wireless may be counterintuitive; after all, hasn't Wi-Fi been a huge hit? In fact, it has. For a recent colum, I drove around my ZIP Code and counted more than 5,000 hot spots using software that detects such signals. But virtually all of these were houses and apartments where Wi-Fi was being used essentially as a network-cable replacement, such as to connect a computer in the front of the house with the DSL modem in the back.

As for people paying $50 or so a month to be able to access the Web wherever they are -- that's a different story. With a few exceptions like T-Mobile, the business of providing wireless connectivity has been a wasteland, akin to the online pet-food companies that sprang up during the Internet bubble.

The simple truth may be that outside of homes and offices, and outside of obvious white-collar tar pits like hotels and airport lounges, there really may not be much of a market for high-speed mobile Internet connectivity. The proverbial killer app could always emerge, like being able to watch movies over the Web, the way they do in Hong Kong. It's just as likely, though, that people would prefer to watch them at home.