Analysis (2): Guy Kewney on Psion

But before all that can happen, two things have to get sorted. First, the 16-bit operating system that underlies today's Psion 3 series has to be upgraded to 32-bit. And second, what Psion wanted from the failed takeover of Amstrad has to be integrated, and it isn't what it seems.

Amstrad, as a supplier of PCs, is of very little interest to anybody. What Psion wanted was the cellphone maker, Dancall, and key to its future is the potential for integrating all personal communications.

Ideally, your phone would come in three parts. You'd have an (almost invisible) earpiece with built-in bone-conducting microphone. You'd have the cellular radio. And you'd have the control -- a display and keyboard or voice-activated console. They'd be connected invisibly; infra-red links are a first attempt to get this organised, but a very, very low-power microwave link would work fine over distances of a couple of feet without affecting human flesh or wristwatch technology. And if the console is just a screen and keyboard, why should it not be a PC?

The question of how small, is the real nub of the problem. The first Psion organiser, back in 1984, had 2Kb of memory. Today's 3c has 2Mb. But huge though this rise in capacity has been, the processor hasn't advanced at the same rate; in 1984, the 8086 processor as used in the standard PC had been rendered obsolete by the 286 of the PC AT. But the Psion 3 had the same type of chip in it, and the 3a used the same 16-bit technology, and the 3c and the Siena all have the same stuff. To be sure, the software has evolved somewhat; but what was a boldly innovative leap of technology when the Psion 3 first appeared, now looks somewhat elderly.

This would, obviously, be a marketing disaster in a market of sensible people. In fact, Psion buyers are rushing to buy more and more, despite the gradual lengthening of tooth shown by the design; it's been enormously popular, and sales continue to boom. And the launch of the 3c marks this fact; apart from really quite small innovations it remains familiar -- and expensive.