Remember the good old days, when Tomorrow's World was still on the telly (and The Gadget Show wasn't)?

Remember when they trotted out folk that introduced new storage technology like laser-disks and CDs, and other sorts of storage?

Remember they said it would last "forever"?

Gallery - Set in stone or written to disc: data through the ages - Image 1 - New Scientist

I don't know if things are quite as glum as NS portrays them. so far, at least, any old storage I've kicked into life has actually worked, but that doesn't mean it will continue to do so.

I'd hate to have to go back to carving my notes into stone tablets with a hammer and chisel, although some might say that those have just bought iSlates are already heading in that direction

I didn't realise the Gallery mention in the preceding post was associated with a much longer article:

Digital doomsday: the end of knowledge - tech - 02 February 2010 - New Scientist

Although this concentrates more closely on the aspect of memory and records, and is a vey good article worh taking a few moments to read, it really is just a special case of a subject I've been banging my drum about for some years.

Although it wasn't as relative to domestic life when I started all those years ago, my life in industrial electronics brought me into contact with two distinct form of systems and circuity.

The first was the was what I might describe as the legacy items, and this included mostly analogue electronics, and could involve anything - including some wonderful toys that I could date back to the the postwar years. The significant factor was that I could probably repair, or have repaired, almost anything that was made up to the end of the 20th century. To use the old Russian analogy, provided it could be repaired by the local blacksmith, just about anything could be refabricated or repaired using something that could be coerced to fit. It may not have been original, but the bits were big enough and rough enough to work.

However, as the 21st century dawned, things changed, and instead of opening boxes and finding generic bits inside that could be cobbled together from a junk box, or the long forgotten stuff lying around on the floor of a factory's maintenance department, what you found was something black, unidentifiable, and unless you could identify and track down the original manufacturer from the number (if it had one), irreplaceable or unrepairable.

While the marrying of new digital technology with just about every electrical device to make something more effective may be a good thing in terms of performance, it does lock the user into that system.

Think, for example, of the future of energy metering in your home.

For years, this has been done using a simple and effective mechanical watt-hour meter, with only one moving part to drive the recording dials. Largely unbreakable and reliable, once made and installed, it should last a lifetime, unless modified by someone trying to get free electricity out of it.

Compare with the "smart" meters we will all be forced to hove in a few years.

There's no denying the reliability of the electronics, but in reality, the measuring components are more likely to drift and become inaccurate over time than the basic lumps of metal that made up the old watt-hour meter. Then there's the question of the memory and code contained within them, and the interface that transmits the data back to the supplier - all these have to be maintained, and kept compatible with future developments.

I'm not suggesting they will be unreliable or fail at an early stage, just noting that they may prove to be more of a liability than their humble mechanical predecessor, and require more costly maintenance. In other words, they have to be cheap so they can be replaced rather than replaced. And they may prove to be limiting in future.

For example, PC design today is stunted and held back because this device is built on the internal architecture of the first PC that appeared over 20 years ago. And we can't really doit any other way, as we can't afford to lose compatibility, and develop brand new replacements that displace the original.