I've said this before, so I should apologise: but let's say it again - the PC era is coming to a close.

When IBM quits the business, that's the sign the doubters were looking for, surely?

As I remarked to you recently, I recall coverage in the SAME 9 O'Clock BBC news brodcast that the paperless office was "just around the corner", and that Clive Sinclair was deeply disappointed by the BBC not having adopted his new "Spectrum" computer as the BBC machine. That kind of date-stamps the whole thing. It's took WELL over a decade for email to surplant "snail mail", so even if the technology existed, that's hardly the point. It has not only to exist... but also to be cheap and (comparatively) reliable. We're certainly reaching the end of an era - My current Athlon-based system looks superficially the same as my original (then almost cutting-edge!) Amstrad PC1512. But my frist stroll around a large computer fair in over six months earlier today revealed wireless keyboards and mouse combo's for just £16, flat-screen monitors for under £150... and (to my considerable surprise) hardly ANY "cube" systems, either "barebones" or fully built.

Just as the keyboard and mouse can now be cheaply divorced from the system box, and wireless home network components can be bought for small change, so to are the other components required to complete the transition emerging - but they're emerging very, VERY slowly, and prices are dropping slower yet. Same deal as the "paperless office." The time for change will be right when ALL of the components needed to produce something better are available... and affordable. Some of the missing ingredients (some? MOST!) are cheap broadband - both wired, and wireless. When THEY arrive, I can perceieve the dawn of "each man his own WAN" - a concealed system box at home, holding the whole thing together as a kind of "hub", and remote inputs and outputs distributed around the house and one's person. The day of the "Smart phone as dumb terminal" will be upon us. That seems to be the direction we're inevitably headed... but there will doubtless be many gimmick-ridden sidetracks along the way. One stumbling block (as I've already suggested elsewhere) is the major problem of getting data into the system and out of it. Keyboards are hard to beat, and there's a minimum size to them (Oddly the show today was heavily laden with conventionally-wired keyboards that were smaller than you'd find on most laptops) I/O between user and system remains the stumbling block to change.