It turned out that ICL had a major, and very expensive problem. As a computer hardware company operating in many countries in the world, they needed stocks of spare parts everywhere. And every time an engineer changed out a spare, the faulty unit either needed to be replaced or repaired and returned to stock. Just those units in transit between local stores and central repair had become a major balance sheet item. And then there was customs duty, it was a major cost of the spares in some countries. ICL’s Spares Division wanted to find an answer, and had recruited an expert from the USA to help.
The answer was to build repair centers around the world, and repair units fast and cheap in the country where they broke. The problem with this idea is that the main volume of items that were in the repair loop were complex logic PCBs (Printed Circuit Boards). This was exacerbated by the fact that ICL was a world leader in the use of complicated, and expensive, multilayer PCBs.
The job I was offered, and decided to accept, was to go to the USA with the new manager and help buy the test equipment, learn how to program it to test PCBs, then return home and recruit and train a team to do it. The salary I was offered made it difficult to refuse, but I was under no illusions that it would be easy – it wasn’t.
There are only two words to describe Phoenix: Hot, Dry. My first trip to the USA was certainly exciting. We arrived first in Chicago, and went though immigration there – having done this many times since, that first time sticks in the mind – American Immigration Officials are terrifying, and I was in the line (Line, not queue) behind about a hundred Arabs coming to the USA to immigrate – the officer was not a happy chappy by the time I got to him! Through that bit, no time to see Chicago, but onto another plane and off to Phoenix – the door opens and that Heat Hits You. My first Air Conditioned car, my first drive on the right, my first American hotel, phew!
But, inevitably I got acclimatised in a couple of days and we started work looking at the proposed tester, and decided we could make it do what we needed. After about a week of that we had to go to San Diego to see another supplier, so my boss asked if I would like to drive – it was a 10 hour trip (24 hours round trip including the business meeting) and I would get to see the country – well Desert actually. We headed off before dawn, and arrived in Yuma, a one-horse town in the desert just as the rain started and the flash floods arrived. As the only Limey in the diner, guess who got the blame for the rain! Still the rest of the trip was great fun, and the Laguna mountains exceptional.
My other recollection of that trip was being asked by a very pretty bank teller to ‘just talk to me’ while she changed my traveller’s checks (sic). Apparently she fell in love with my accent.
Stevenage, then Letchworth, then Sydenham
We were based in the same offices in Stevenage that I had worked at three years before, but it was different division and none of the people that I had known before were there. We started to size the task and soon realised that there was not going to be enough office space for the team in Stevenage. We took over an old factory in Letchworth (where I had grown up, and my parents still live) as initial offices for the new unit, but even that wasn’t going to be sufficient.
Soon I was planning a move to South London, to Sydenham, a suburb where there was a large warehouse with an upper floor completely unused, and better still an air-conditioned computer room ideal for the massive computer we would need to develop the tests. My wife was overjoyed at the propect of a company-paid relocation to the South London she knew from childhood, so the house went on the market and we moved into a rented apartment while we found a new house.
Simulators and test patterns
I shan’t bore you with the details of the software we wrote, indeed my memory of it is not up to the task of a full technical exposition. It involved working out how to diagnose a fault in a section of electronic logic by deducing the inputs necessary to stimulate outputs that could be used to pinpoint the faulty component. This is maths. To be precise it is Boolean Algebra, and the way to solve the equations is to use a computer to simulate the logic and throw lots of input patterns at it and see if the outputs are unique. You need a computer as there are not enough hours left before entropy does for the Universe to do it by any other way.
This is why we needed the big computer, and why we needed that big computer room. And I got an office, and I was also, surprisingly, Senior Manager on the site. Bearing in mind I hadn’t been any kind of manager six months earlier, this was a rude awakening. I imagine it’s a bit like a naval midshipman suddenly finding that all the other officers have been washed overboard and he is now the captain – well, it felt like that to me anyway.
I coped. The team grew to 25 graduate engineers, I recruited some operators for the computer – first time I had female staff, and then black staff – absolutely no problems at all – apart from my worry that I might inadvertently do something that might be seen as prejudiced.
And then suddenly the operation was just running along, doing it’s job, and I was, I realised with a start – bored. On top of this, the American manager resigned to go home, and I was not particularly enamoured of his boss – who suddenly became mine.
Now, in the course of setting up the department, I had had a lot of contact with the Test Engineering department in Manchester – many of the people with whom I’d worked in my first ICL job were there (although none of them were in managerial roles so I think I had made the right decision). In that group they had developed a Simulator (that software for doing Boolean maths) and it was this that I had used to set up my development system. I was approached by one of their management team with a view to setting up a new unit for them in Bracknell – to the west of London – where a new computer development group was forming.
There were two good things about this idea. Firstly it would let me use my newly formed management skills in a new team, and there would be another company-paid move which would get us out of London. I took the job.
From one perspective it was good move. We bought an good home in Send Marsh, a village near Guildford, this being halfway between my job in Bracknell, and Ros’s job in Reigate (she was a senior programmer at a Chartered Actuaries). From another perspective it was disaster.
It turned out that I had been told less than the truth, and my innocence led me to believe I knew what I was in for. The pitch had been to liase between the Manchester group and the new Bracknell group and help them set up their Engineering test operation as a clone of Manchester’s. The facts were that the Bracknell group had been set up expressly to break the grip of the Manchester cost base, and to do things differently – using technology imported from the USA after ICL had bought Singer (a US small computer maker). I was a spare part, and couldn’t do anything for either camp. After about 3 months I jointed the Bracknell Team as Manager of their test engineering group, and built a new team to do much what I had done for Spares Division.
This time though, I knew exactly what not to do.