What’s the big deal? Aren’t thousands of mobile calls “handed off” every day from one “cell” to another without a glitch? They are indeed. But third-generation technology, or 3G, is so radically new that it requires a rethinking of just about every aspect of how mobile phones work, from the handsets to the transmission masts to the software that runs them. For this reason, 3G is a massive engineering and construction project that will take years to complete and cost hundreds of billions of dollars. The magnitude of this effort has somehow been forgotten in the mad scramble to be first out.

The handover problem is a case in point. When you talk on a conventional mobile phone, your call is beamed as a continuous stream of digital data to the nearest receiver. The technology for handing these calls off from one area to the next was worked out years ago. But a 3G phone is different: it bundles up the data into little packets and sends them through the airwaves, one at a time. This creates the impression of an Internet connection’s being “always on,” which is good news. But keeping track of these data bundles from one region to the next is a daunting engineering problem–and, more to the point, a brand-new one. NEC, the Japanese phone company that supplies BT with equipment for its Isle of Man trial, hasn’t had time to work it out.

Handset makers also have work to do. The 3G technology has so many features, only a wonder gizmo could handle all of them, which is why none exists. The phones are not only supposed to work with 3G networks but also with the less sophisticated (but cheaper and more useful) General Packet Radio Service (GPRS) technology already being installed on the continent and also with the current mobile phone standard, Global System for Mobile (GSM). Phones for corporate executives are also supposed to adapt to dozens of other standards around the world. Doing all this requires powerful, custom-built computer chips, which are tough to make quickly.

A device that does so many things is bound to guzzle a lot of power. Prototype 3G phones drain so much juice that they’ve been known to get uncomfortably hot. Batteries that can keep a conventional phone running for days would fizzle in a 3G handset in a matter of minutes. Engineers are searching for alternatives, but at the moment the lack of a long-lasting battery is a major hurdle.

Then there’s the problem of how to design handsets that take full advantage of the capabilities 3G makes possible. How, for instance, do you watch videos on a phone small enough to fit in your suit pocket? “You get a viewer you can plug into your mobile phone so you get a bigger screen,” admits a spokesman for Siemens Mobile. Says a consultant: “What’s needed is a sea change in the machine-man interface.”

Nokia says it will introduce its first 3G phone at the end of next year, but experts are skeptical. “We had to wait nearly a year after the network was ready for the first GPRS phones to reach shops,” says Simon Buckingham, head of telecom consultants Mobile Lifestreams. Some experts don’t expect to see handsets available in serious quantities until 2004. Others say 3G may ultimately prove to be inappropriate for phones and may require some other device altogether.

The one technically straightforward aspect of 3G networks–the masts that communicate with the handsets–happens to be politically fraught. About 700,000 new towers will be required throughout Europe, and finding places for all of them is a monumental task. Space is so tight in Britain that BT is planning to mount low-power versions in telephone booths, and others may even go on churches (in return for £6,000). Lobby groups complain that the masts deface the countryside and the radio transmissions are a health risk.

None of these problems is insurmountable, but neither will they be resolved quickly. Analysts at Forrester Research in the Netherlands predict that even in 2005, when more than half of Europe’s phones will be connected to the Internet, fewer than 15 percent of them will use 3G. That’s a measure of this technology’s complexity and immaturity.