Rupert Goodwins' Diary

MIMO is creeping out of the lab and onto the shelves. This technology -- Multi-In, Multi-Out -- is what happens when you stick more than one antenna on a radio. Do it right by adding circuits to delay and amplify the signals from each antenna in the right way, and you can treat each physical path between transmitter and receiver antennas practically as it own channel. Result: lots more throughput and reliability, lots fewer blackspots and awkward signal fades as you move about the office or home.

The latest device to sprout aerial excrescences like a cress-haired doll is the Linksys WRT54GX access point. Like its rival Belkin, it uses an Airgo MIMO-enabled chipset: it has three long blank antennas that look, frankly, like a hatstand. But with the right cards in the clients, it can pootle along at a reported 30Mbps while running streamed videos. MIMO works.

Or rather, it would if the companies concerned weren't so busy trying to sabotage the thing. The official next-generation MIMO-using Wi-Fi standard is 802.11n, which hasn't yet degenerated into a UWB-style Mexican stand-off but is trolling along slowly. Too slow for numerous vendors, who are offering pre-N MIMO kit, but which is pretty well guaranteed not to work with anybody else's in the new super-duper modes. Everything will work as ordinary Wi-Fi, we presume, but that's hardly the point.

The point behind standards is to grow the market. A fragmented market is not attractive to the users, at least not where interoperability matters. But we know how much the users matter.

The other area where we're being let down is in industrial design. There are any number of great physical designs that suggest themselves for devices that have to have multiple antennae. A wall-hanging starfish. A sunburst. A Sputnik. A model four-masted schooner, perhaps with solar cells in the sails.

All these delights await a jaded world, just as long as the manufacturers get their act together, talk to each other and take care of the users. It's not hard.