The other day I was checking the numbers on my bank statement and it suddenly struck me that my calculator is approaching its 21st birthday. It's a Hewlett-Packard HP15c and has been used almost every day since 1980. I started wondering what makes this calculator such a long-lived product and thinking about other contemporary items of digital electronics.

In our1980 office only the secretaries had personal computers on their desks and these were 8085 machines running CP/M and WordStar; the IBM PC had not yet been launched. Digital watches were all the rage and cellphones were some way in the future.

I looked around the house and could find no other digital devices from those days that were still in use. In the workshop, it was a bit different in that I have a Racal digital frequency meter from that time that is often used. I have plenty of older analogue test gear, but that's another story.

So what is it that makes the calculator and the frequency meter different from other devices that are long since obsolete? For a start, they both do one job only and they do that very well indeed. The calculator is about the right size, it's robust, its display is easy to read, the batteries last ages and the keyboard is excellent. The frequency meter has enough resolution and accuracy, a clear display and is easy to use. In either case, it's hard to think of other features that would make the devices better in some sense.

Another factor is that the technology is appropriate. What I mean by that is that, from the user's point of view, the devices do the job satisfactorily. Even today, both companies make products that do much the same job, however, the insides are totally different. This benefits both the manufacturers and the users in that the costs and prices are reduced but the fact that the functionality remains much the same shows that the original design was good.

Anyway, is it a good idea to make digital products that users will keep for over 20 years? It depends on the type of product and the maturity of the requirement. Clearly, it will be a long time before PCs reach the point where technology improvements have no significant benefit to the user, but what about things like cellphones? If a user wants a compact phone just for voice calls then almost any of the current products have the potential to last a good while and it shouldn't be necessary to upgrade every year or so. Unfortunately, the business model of the mobile phone manufacturers means that production runs last less than a year and we are continually persuaded to buy the latest model; however, it makes calls no better than last year's.

So what can we learn from all this rumination? If your products have a clearly-defined functionality and the customers don't want too many bells and whistles then it ought to make sense to go for a design built to last. This might cost more but it will build a loyal base of customers who will recommend the products to others. As technology advances, use it to keep the costs down but keep successful functionality and user interfaces. If more manufacturers did this we would have better products and happier customers.