Rupert Goodwins' Diary

Wednesday 20/03/2002

A report from US analysts ARCchart makes for uncomfortable reading -- if you're planning a 3G service. Based on current market conditions, cost projections and technology capabilities, it says that 802.11b will take away between 12 and 64 percent of the 3G market.

At first, this seems pessimistic. Nobody's building nationwide networks of 802.11b; it has no integration with the phone system, it takes too much power for portable applications and it has none of the authentication, billing and other magic that makes 3G much more than just a digital wireless network.

The trouble is, GSM has all this. Put an 802.11b chip into a mobile phone, and that's all you need: if you're in an area with 802.11b, you can authenticate yourself via ordinary GSM and get access. I could put up an 802.11b network for a hundred pounds in my front room, and serve an area with maybe a hundred people in it -- compare that to the build-out costs per capita for 3G. And as for the cost and power consumption: this week, I saw a report of a single chip that does GSM, GPRS, GPS satellite and Bluetooth. You know how the costs of things fall when you get them onto a single chip -- and the company's looking at 802.11b for next year.

That leaves the power problem, which is fair enough. But 3G will take just as much juice. No deal.

There is every chance that by the time 3G turns up and starts to offer the couple of hundred kilobits/second, we'll all be gallivanting about getting 11 megabits from the corner shop.