News

No 3G at 3GSM; just 3G SMS

You would expect to see a 3G phone call at a show called 3GSM. In fact, the show in Cannes, Frances this week was far more of a 3GSMS show; the world may be waiting for universal mobile phones delivering data at "up to" two megabits per second, but behind the hype, there are problems. Not technical ones; conceptual problems

There was a 3G phone call to report in Cannes. The call was made in America, and it was presented in triumph by AT&T and Ericsson together - but you needn't imagine this means anybody else will use UMTS equipment in anger in the United States this year. It was just a test call, using "enhanced" GSM phones.

The really exciting stuff in Cannes, was of little interest to the people there; they want more big-money generation. They're dreaming.

The good news about 3G phones is that test equipment is shipping, or at least being developed; and phone equipment is being linked up. But the good news won't turn into a commercial product until next year, 2003, at the earliest.

We'll be able to tell that high-speed wireless data is a reality when the publicists stop saying that it is already deployed. What you have in Japan is not UMTS, but something vaguely similar - very nice, and much higher speed, but with none of the breakthrough advantages that a global 3G network will bring.

And in particular, none of the big-spending data users that the telcos are all dreaming about.

In the real world, the products that are attracting attention are from companies like Ubinetics. This company announced that it has the first customer for its portable 3G test mobile.

This is well worth analysing in more detail.

If you were about to open the first motorway, would you be announcing the first device capable of spreading bitumen across three lanes? Or would you expect such equipment to have been invented, tested, debugged and put into mass production before the first bulldozer cut into the green fields? Obviously, you'd expect this to be something everybody was already working with.

If we were close to deploying proper 3G networks, companies like Ubinetics would not be announcing their first customer for critical test equipment. It would not be shipping early equipment for testing wireless networks. And its customers would not be boasting about how the new Ubinetics equipment gave them a head start over the competition, as Agilent is doing with the TM200 "field test mobile" this week.

If the networks were in a position to start rolling out anywhere outside a laboratory, we wouldn't have AT&T upgrading its GSM network to GPRS just to place a single test call.

In Cannes this week, all the new ideas were SMS based. Multimedia Message Service, MMS, was the exciting stuff. High-compression video streaming ideas were rolling off the press release desks. You don't have to compress data down to 9600 bits per second if you have UMTS, because, remember, this is going to provide "up to two megabits per second" of data. You only start doing SMS and data compression if your major markets for the next two years are all ordinary GSM.

Don't get me wrong; I'm as excited about the UMTS dream as anybody; but as long as people keep hyping it, you can be sure there's nothing there. And while we can be jaded if we like, about SMS, it's nonetheless as good as it gets.

At Cannes, there were dozens of exciting new ways of using SMS. They are all worth taking seriously, and they are all going to make some money for the people who deploy them. They aren't what the big guys want, though. If you talk to Vodafone, MMO2, Orange, Hutchison, Telia and the rest of them, they'll admit that they are making less and less money out of SMS. What they want is a way of selling data to bulk users, at high rates.

The days of high-rate data are over.

BT postponed its rollout of ADSL because it would have been a serious threat to the existing leased-line business. People were paying $10,000 a year for a dedicated data line that ran only 64 kilobits of data through it per second; ADSL gives 512 kilobits per second for $500 per year. And within two years, people building new houses will expect 10 megabits per second over fibre; it's already available in parts of Sweden, for example, at ADSL prices.

In a world where a megabyte costs a fraction of a penny, people simply aren't going to spend large sums on data over UMTS. They will use UMTS when its prices drop to similar price levels.

And that simply won't happen for at least three years, and maybe longer. And if it doesn't happen in five years, it may never happen, because the world won't wait for the big networks. The customers will do it all themselves with Bluetooth and WiFi local networks over the Internet.