I started work at ICL (https://www.icl1900.co.uk/) in Manchester at the beginning of July 1973. I had always been impressed by ICL’s presence in Manchester. Here was a leading company in new technology and it had a large establishment less than 10 miles from my parents’ home. In the early part of 1973 it had also shaken up the small computer market in the UK with a cheap machine (called the 2903) using the latest electronics. At Univac we lost business time and again to ICL .
The interview process had led to my being selected from a bunch of people and I joined the “coding section” of the Design Automation Department (aka DA). I won’t go into the events that led up to my start much, but there was an aptitude test which apparently I failed miserably – probably because of lack of confidence, lack of sleep (I had driven up from London for the event) and a general fear of such things. This tainted my career at the company forever apparently because it was on my personnel record and prevented my moving out of DA. There had been two finalists; the other was an estate agent that was looking for a move into what he thought was “a thing of the future” (how right he was). I got the job because the team leader (John Donnelly) and department head (Gordon Adshead) were impressed by my degree and they thought the other guy was a bit of a “wuss” (or words to that effect). A second person recruited at that time was a young school leaver that was planning a career and company-sponsored education in computer science, called Don Holmes. Don and I became good friends. I also bonded to some extent with a person called David Hilchesson and another experienced coder called Vijay Sharma. Dave had been part of in interview process. Vijay lived near John Donnelly, close to Hollingworth Lake.
Frankly my first impressions were not good. My first day was marred because of a misunderstanding. I was sitting at the desk of somebody that was on holiday. The phone rang and I answered it. The person at the other end started yelling a swearing at me. He thought that I was the guy that normally sat there. The “yeller and swearer” is a good friend of mine now so I will refrain from naming him. I have to say that this event left a scar that stayed with me for many years.
Another issue was the obvious petty jealousies and snobbery (both inverted and otherwise) that permeated the group. Not only was the company full of class distinction like the levels of eatery (canteen , silver noshery, golden noshery, etc.), and the way the general manager would show up each day in a chauffeur-driven Jaguar, there was plenty between individual employees. Also there was the indignity of the time clock for salaried staff. None of this had existed in London. Time clocks weren’t even used in my days working in cotton mills. It all seemed very nineteenth century.
Also, I was privately disgusted by the abuse of overtime (though frankly, as time went on, I became as guilty as anybody else): many people would stay at work for hours and play cards, read or whatever they could do to pass the time. Of course decent pay and being treated like adults would have gone a long way towards getting rid of this.
West Gorton, the Manchester suburb in which we were located, was a notoriously deprived area with a lot of crime. There were badly built 1960s housing projects. Car theft and break-ins in the company parking areas was common and some areas were locked during working hours. On at least one occasion we called the police because we saw a break-in underway from our window and the thief was caught. On another occasion, there was an outbreak of theft from pockets of jackets left on chairs. The thief was caught because he walked into the building wearing a white lab coat. When challenged he pretended to be a doctor and said “I’m looking for a nurse” at which point he was taken aside by security people.
Corruption became an issue at ICL. There were two things that stood out. Shortly after joining the company somebody said to me “have you bought your ham”, well I had no idea what this was about, but enquiries revealed that the person delivering the mail each day would sell ham that came from the canteen. So I joined the scheme. It was very good too and for several weeks I purchased this. Suddenly it stopped. The post-person changed and the canteen manager was arrested. Apparently, they had been fencing meat stolen from hijacked refrigerated trucks.
Anyway, there were a lot of things that made me feel a little bit alienated from the coding section team at first, but I did like the people even if they were a bit rough around the edges: true Mancunians like me. The person whose desk I had borrowed on that first day was called Colin Lekenby. For fans of 1960s British pop groups, his brother was Derek Lekenby – the keyboard player in Herman’s Hermits. Another person in the adjacent programming team had been at the same school at Graham Nash (of the Hollies and Crosby, Stills and Nash).
There was an early issue because I had been hired at a salary that was more than that received by some others. I had actually taken a sizable drop in pay to get the job which I rationalised because of the opportunity and the lower living costs in Manchester. The consequence was a period of unpleasantness. Also when I really started on the work, I realised that my ambitions were well beyond what was done in the coding group.
The job was part of the project to develop “New Range”. This was to be a product line that would bring ICL’s machines ahead of the competition. The market was almost entirely dominated by American companies where even the smallest had a huge domestic customer base compared to what existed in the UK. The vast American customer base meant that ICL did not have the reserves to embark on big projects (a problem that also caused trouble in other industries like automobiles). The result was that ICL’s development was supported by a lot of government money.
Our part of the work was to code drawings made by the hardware developers into the language of the Design Automation system. My boss was the team leader called John Donnelly, his boss was the operations manager (I forget his name – John something or other) and above that was the department head called Gordon Adshead. Gordon. or +Z, as he signed himself, has left a massive impression on me. Gordon was the undisputed technical mastermind behind the whole department and well known worldwide authority on what we were doing in DA. Gordon was the architect of DA’s software system at the time – called DA4 and it was at the heart of the development procedures for New Range. Because many of DA4’s features were used for design simulation and verification in addition to record keeping, a serious problem in DA4 could cause a big flaw in the final product – an issue that was to cause some heartache at a later date.
When I started, the coding team actually seemed to have some decent work, but early on I had a big scare. The files were kept on magnetic tape. Each update caused all the data plus the update to be copied from one tape to another. For security reasons, there was cycle of 5 tapes. When all 5 tapes had been used, the first one was overwritten. 5 paper tape updates were kept and the oldest was destroyed when a new one was made. This meant that as long as at least one of the tapes was in good shape the data could be recreated. Of course it was me that discovered that there was a problem with one of the tape decks and its inter-block gap was incompatible with the other units. We were down to the last usable tape before one of the programmers (Stuart Vernon) discovered the issue and we were saved. A service engineer fixed the hardware. Stuart was a brilliant programmer, very rough around the edges, a heavy drinker and self taught from what I know. An indicator of how the education system can leave talented people behind, but when the opportunity arises such individual can shine.
Another person that I met during this period was a very large personality called Clive Crossley. There’s not much more to say about Clive here, but he became a very significant presence in life later. More on this in the next phase of my career.
The coding team soon became little more than a paper pushing group that chased work through the computer operations. There was also a shift working scheme because deadlines were tight and the company wanted to make sure that any problems with overnight tasks were fixed promptly – I became restless.
I took a class in COBOL programming at Salford College of Technology and I was allowed to use the department’s computers to write and run programs. My objective bit by bit was to move over into the programming section. This was possible because Dave Hilchesson had made the move. DA4 was written in an assembly level language (an invention of Gordon Adshead I believe) and I was able to get a copy of the documentation, I loved playing with computers and most evenings I would be writing code at home to be tested on DA’s machines when opportunity allowed. In 1975, there was a reorganisation. I was a bit antsy about the changes because they would bring the coding section into a new group that was separated from the programming team. Anyway for whatever reason, Don Holmes and I managed to move over. My new boss was an Iraqi ex-pat called John Hillawi. I was given the task of maintaining the libraries of software models of components. Also, at the same time, the team at ICL’s Stevenage group was being disbanded and merged with our group. I took over maintenance of the software that accessed most of these libraries. I also was given the opportunity to take a 2-week class in ICL’s publicly-available assembly language called Plan. A big step forward!
I think that it’s fair to say that the following years at ICL were the most enjoyable in my career. I loved working on that system and its successor.
We used a family of computers called the 1900 series (http://ict1900.com/) using an operating system called George 3. George 3 had “virtual memory”, we didn’t call it that. The hardware was based on a 1960s design called “Ferranti-Packard 1900”. the 1900 series was very successful, but in many ways it was archaic e.g. its use of octal . However, it used a lot of technology derived from the Ferranti Atlas – a government funded project that was where virtual memory was invented. It’s a long story, but there was a lot of bad feeling about this, because the government sold the patents to IBM and they incorporated this into their highly successful 370 series and as a consequence ICL (successor to Ferranti computers) had to pay royalties on something that they had invented, but that’s another story.
It’s not time to get into the history of computers, but ICL’s contribution through its connection to Manchester University (which built the first programmable machine) and the Leo — the first commercial machine — made it one of the premier places to work in the field, a fact that did not seem to have dawned on the West Gorton management who seemed to have little respect for the Manchester employees. This all came to a head when there were recruitment adverts for people at some of ICL’s establishments in Southern England where the salaries and conditions were plainly superior to those at West Gorton. The engineers were incensed by this. There was a mass joining of ASTMS (Association of Scientific, Technical and Managerial Staffs) and some 1-day strikes. Led, not by rabid communists, but by leading members of the hardware design team. A loud voice was my good friend Richard Bosworth and in DA, my then boss – John Hillawi. Well, the top management were obviously woken up by this, and we had significant pay raises and the time clock went away (among other things).
I have digressed a bit from my work. I loved the programming and I used to write software for fun. My first big success was to speed up the loading of libraries significantly. The existing software had come from Stevenage and was tortuously slow. Possibly the closure of the Stevenage group had occurred before the responsible engineer had time to finish the work; I can only speculate about that. The changes I made were very simple and obvious, but users were very pleased. As time went on I made many changes to this software and wrote some programs in my spare time that were released.
Back to the career then. I suppose there many things that I considered to be successes in addition to the component model library and the associated software. I was very proud of my simple program to track critical paths through a component network, my attempt at automatically creating a symbol of a component on an information listing, a spares program that I wrote at home to show engineers graphically where the available space was to be found on a PCB. Then I speeded-up some disk access routines. It’s a long list and I was having fun. I also started studying computer science at home and I also took a year on a Associates Degree class at the Open University. My real learning was by personal study reading books on relational databases, Donald Knuth’s books on algorithms and books on computer logic. I also managed to become a licentiate at the British Computer Society. Sometime in this period I became very friendly with an engineer in the diagnostic group He suggested that I try for one of the openings there. This failed because apparently my not great aptitude test result from years before had followed me – nothing is lost in a big bureaucracy.
Away from work, life during the ICL years was chaotic to say the least. The fact that we were located in one of the seedier parts of Manchester, which was then at its low point, meant that there was a lot of rough pubs nearby. There was a culture of heavy drinking among many employees of DA too. Some had serious issues in my opinion, but I won’t go there. Anyway I survived and was not drawn in to the excesses that plainly took over the lives of some of my colleagues.
In 1980 I attended a European Conference on Design Automation at Brunel University. It was something of an eye-opener and I heard about the burgeoning business of Electronic Design Automation. Also somewhere around this time I attended a 2-week class on S3; the systems programming language for the new 2900 series computers. It was held, by coincidence at ICL’s training center in a building that was located a few miles from Royal Holloway College. As an aside, S3 leaned heavily on Algol 68’s way of doing things.
Another event of minor significance at that time was what I think was a visit from somebody from a Manchester college, part of what is now Manchester Metropolitan University, they were looking for people to fill places in post graduate programs. I applied and I was accepted in an Master of Philosophy program in Computer Science, which was a bit lower that doctorate, but not bad. Frankly though I did not want to commit to the effort and turned it down. Maybe I should have gone for it – life since then would have been a lot different if I had!
Soon after, a job of team leader within a group to build the successor to DA4 came up and I applied – I got the promotion. It was a 2-grade level leap and the management was reluctant to give me the pay raise. That was the beginning of the end of my time at ICL. However, it coincided with some new rules that were intended to discourage certain staff from leaving the company – that included me and it was gratifying to know that I was valued. There was a stock purchase scheme, a guarantee of 6-months notice before redundancy (in return I had to give 3-months notice of leaving) and a jump in the number of weeks of paid vacation per year (to 5 or 6 weeks, I cannot remember which). All this was brought in during a period when the company was led by a young American recruited from Texas Instruments.
Anyway, I was somewhat discouraged by a few things. First there was the nonexistent pay raise, second there was the falling off in interest in a new system. Also a number of key people were leaving DA – Mike Hewson moved to the semiconductor subsidiary of a larger company and others followed him; eventually my boss Roger Bell also went there. So I started to look around.
First I signed up to employment agencies. Frankly they were useless. Their business was filling openings for COBOL programmers in commercial applications. There were some requirements for PLAN programmers, but not much for people like me. Definitely not with very technical applications. This lesson in overspecialization stayed with me.
So I started to look in the Computer Press. Specifically Computer Weekly. I saw an ad for people to work at Cirrus Computers in Manchester and Fareham in Hampshire. there was a need for somebody with database expertise in particular. Electronics and Design Automation were mentioned. I applied and found myself being interviewed by some very familiar names.
The name Cirrus was often seen around DA, but frankly I had not taken too much notice of it. There was obviously a big connection with ICL, but what it was had not dawned on me. Anyway, I sent a letter in response to the ad and received a call from a Gordon Robinson – not somebody that I knew, but he knew of me. Anyway I was asked to go for an interview over two days. It’s a complicated story, but I chose to take the train rather than drive. Gordon told me that the opening for which I was to be interviewed was in Fareham, not Manchester as I had hoped. So OK, I went along with it all.
Things did not seem to be going my way. First of all, the train was delayed so much that the first day of my interview was abandoned – I was supposed to arrive at lunch time and it was well into the afternoon when I made it to Fareham. Gordon was very nice. he gave me a ride to the hotel called the Maylings Manor. The second problem was that I was expected to start soon, but the 3-month notice of leaving would get in the way. Another issue was that I had obtained a municipal grant to renovate my house in Manchester – there was a snag in that I had to stay there for 5 years or pay the money back (it was only 1 year since I had received the money). Anyway, Cirrus was willing to wait out the 3-month period, the rules were changed about having to stay almost at the same time as I applied to Cirrus (phew) and the interview went well. it seemed. I met Clive Crossley and John Bowthorpe. John and I had a mixed relationship during the following years, but overall it went well. John waved the letter of application in front of me and said “this is good”, but I learned later that he was not too keen on hiring me. I did get the job though.
More about John and the rest of the Cirrus team on the next page.
Of course as soon as I told ICL, they released the money that they owed me from the promotion. The people at Cirrus said that if I could not move then at least I had something from the experience!