Scor­pio News

it send external images but it can send its own stored images as well. Ok, so it’s a very expensive FAX machine, but that’s not why you bought it.

Yet another sneaky idea is the ‘Security Photocopier’. Now in high security establishments, they tend to guard their photocopiers not so much against unauthorised use as the copying of unauthorised documents. The problem with a photocopier is that once it’s done its business, that is made a photocopy, no-one knows what was photocopied. Now it’s easy enough to bar access to the photocopier, using a security key is one way, but suppose someone with a security key intended to be naughty and copy things he wasn’t supposed to anyway. There’s no way of knowing what was copied. Alter the logic of the Imaging System, so it requires a password or key to use it – that way you know who used it – because the system logs who used it. As it copies it compresses the image(s) of what was copied to the disk and ties it to the log-on. That way you not only know who used it, but what was copied as well. The only problem with that is if it was a stolen key or password – but that’s a physical security problem. Again, you wouldn’t buy an Imaging System as a photocopier, but it’s a useful adjunct to it’s normal use.

The Software.

At last – on to the software. Now all the hardware is well tried boxes of works, each of which has been designed to do it’s job, either on its own or in conjunction with other boxes of works. The thing which hangs it all together is the computer software which runs it and if anything, it’s the software which is the most important part of all. Now I have some very fierce criticisms of some approaches to the software design of some imaging systems. Most of my criticism stems from one of two (to my mind, mistaken) assumptions made by the software designers; based on the fact that as Imaging Systems are expensive and the software will inherently be a low volume dedicated market compared with, say word processing software. This means that the cost of software design which is high in any event, will be amortised across a relatively low volume of sales. The first assumption seems to be, “As we aren’t going sell many anyway, let’s throw the software together as cheaply as possible.”, the second assumption, is the reverse, “As the software is going to cost a bomb, let’s make it as thorough and complicated as possible, that way the punter thinks he’s getting his money’s worth.”. Much software I’ve seen falls into the latter category, and as a result you need a Ph.D. to drive it. Of course I advocate a middle road, I’m a firm believer in KISS technology (translated in two ways, either ‘Keep It Simple and Stupid’ or as an instruction to the programmer ‘Keep It Simple – Stupid’), I prefer, the ‘Keep It Simple – Stupid’ approach. That way you end up with ‘bomb-proof’ (another Computerese term which means ‘no idiot can make it go wrong’, and it’s relatively free of bugs) software which is easy for an unskilled operator to use, and which covers 95% of all possible uses. It’s always the remaining 5% of clever ideas which don’t work and cause complication. The only trouble with KISS software is that it’s