Written in a coffee shop in Seoul and despatched to silicon.com via a really fast, and I mean really fast hotel network at about the same silly connection price charged in the UK
Every now and again I do something really dumb. In my defence the dumbness is mostly semi-dumb as it is generally forced upon me by circumstance. It's not that I don't know what I'm about to do is dumb, it's that I have no choice. So here is my latest and most costly lapse of sanity.
About 12 to 14 months ago several events happened at once: my assistant's laptop suddenly appeared moribund with a pink screen; my laptop lost LAN connectivity via the RG45 port; a new staff member joined the team; and my son was about to set off for university for his first year.
I got quotes for repairing the broken laptops that exceeded 50 per cent of the cost of new ones and both units were well out of warranty. So I knowingly went out and purchased four identical laptops over a three- to four-month period. Yep - I know, really dumb but I had no choice. I had to get everyone up and running and back to work pronto.
So, as a result, the last two months I have seen one battery and three logic board failures! How come? Correlated component purchase! The battery failure may well be a random event - and I do hope so - but the logic board failures are almost definitely down to the supply of a single or multiple components.
Here is the inside line as far as I can determine. The manufacturer concerned purchased thousands of capacitors from a single supplier but there was a single bad batch that escaped detection. Yep you got it! I purchased some of that bad batch. Worse still of course, it is well beyond human dexterity to replace a single component or group of components and it becomes necessary to change a very expensive board.
The moral of this story is obvious: always spread out your purchases to avoid correlated and concatenated failures in the future. It really is painful to find yourself with continuous and unexpected machine failures, data recovery, new replacement machine purchase, set-up and loading, whilst at the same time trying to keep business going.
My back-up and contingencies are pretty thorough but inevitably some files and documents get lost, distorted, damaged and/or corrupted. And the reality is I only lost a week of evenings working on these problems but I also repeated my initial error by having to repair three machines at the same time! It turns out that getting out of this correlated purchasing loop is expensive and the only route is to buy yet another new machine to break the cycle.
I have one other tip. I never buy the latest machines or equipment. I always operate six to 12 months behind the bleeding edge in the hope that all the initial bugs will have been ironed out and the production and supply chain will have become stable. But it might be that the rate of change might have now reached a point where true stability is never reached... but I do hope not!
Peter Cochrane is an engineer, scientist, entrepreneur, futurist and consultant. He is the former CTO and head of research at BT, with a career in telecoms and IT spanning more than 40 years.