Digital audio of a recorded interview with Ninian Eadie, who joined LEO Computers as a trainee programmer, going on to play a senior role in ICL and then Fujitsu. 

Interviewer: John Ferguson
Date of interview: 6th October 2017
Length of recording: 36m42s
Format: original .wav recording 186.03MB (transferred to .mov video for presentation on YouTube 133.42MB)
Copyright in recording content: Ninian Eadie and LEO Computers Society

Transcript editor: unknown

Abstract: Education, prep school, Winchester, Balliol College, Oxford (PPE). After National Service in the Navy joined LEO as trainee programmer. His LEO career included teaching on training courses, joining LEO teams assisting customers like Cerebos, defining LEO III software, acting as sales manager for LEO’s joint venture with Rand Mines in South Africa, and culminating in the role of Project Manager for the LEO III and System 4 Post Office project, the largest and most technically and politically complex project in LEO history. Following the merger which created ICL in 1967, he took on a series of increasingly senior roles for ICL and its successor companies, including managing ICL’s international sales organisation, acting as assistant to ICL’s chief executive and under Fujitsu became (1993 -1996) Group Executive Director for Technology before retirement. Amongst his outside interests he became an internationally recognised Dinghy racer.

Transcript :

Interview of Ninian Eadie by John Ferguson 
JF: It's the 6th of October 2017 and I'm John Ferguson. I'm interviewing Ninian Eadie to give us the story of his involvement with LEO Computers from the earliest days.
  Good afternoon Ninian. We are recording this interview as part of the LEO Computers Society Oral History Project. The audio version and the transcript will be lodged at a central archive and made available for researchers and members of the public.
 Perhaps you'd like to introduce yourself Ninian.

NE: I was born in 1937 in the West Midlands. My father was a Civil Engineer, and my mother a piano teacher who re-trained as a physiotherapist. She worked at the Royal Navy Hospital Haslar

My father was a self-made man, who started work at the age of 11 in a solicitor’s office. He was a great believer in education and sent me to Winchester College (one of England’s premier public schools) where I did my A levels in History and English.

After Winchester I did my National Service in the Royal Navy and was commissioned as Sub-Lieut RNVR. (Royal Naval Voluntary Reserve)
After National Service I went up to Balliol College Oxford. I was the youngest of seven children, but I was the first member of my family to go to university. I read Philosophy Politics and Economics (PPE). So you can see that my education did little to prepare me for a highly technical career in computers.
I finished at Oxford in the summer of 1960 It had always been my intention to go into industry. In those days, Oxford graduates were much sought after .because only 5% of the population we're going to University -: very different to today. So I got a short list of companies in different areas which I thought were innovative. One was the early maker of fibreglass, one was the Dexion ‘slotted angle’ company, another was a perfume company (Yardley), and one or two others. I started going round and having a look at them. When I went to Yardley in the East End, the chap who interviewed me said, "I wouldn't join this company if I were you because all the Directors are family members, so there are no prospects for you here. What else have you got on your list?" And then I went through the people I had talked to, and he said, "Computers, really are the thing of the future, so you really ought to do that." And so that is how it came about. I went to LEO and I had an interview with David Caminer and I was put through the tests…...
JF: Where was this?
NE: This was in Hartree House (over Whiteleys department store in Queensway.)
NE: They gave us this little programming test where you were taught how to program in an hour, and then you were given various programs to write to see if you could optimize the lengths of the paths on the program (The candidate was taught a simple instruction code making use of three storage registers and was then asked to write programs to carry out tasks while making use of the smallest number of steps).On the basis of that, and an interview by David Caminer and by Doug Comish. I was given a job offer. I was still waiting to hear from any of the other companies, but LEO impressed me tremendously because I got my job offer almost the following day. Because they seemed hyper efficient. I accepted it on the condition that I wouldn't join them for six months because I wanted to go to the United States. (Editors note. This is famous Aptitude Test. It took a whole day and was taken by all potential programmers, operators and engineers. Note it happened before you were offered an interview. In other words, you had to pass the test.)
In August of 1960 I set off to drive from New York to Los Angeles. I toured San Francisco, Los Angeles and San Diego. Then I came back by Greyhound bus through New Orleans and Florida and eventually back (on the Queen Elizabeth liner)
As it turned out, I joined LEO in the January of 1961, albeit theoretically I had been working for them for six months. 
When I joined LEO the first thing they did was to put me onto a programming course. This was an Intercode course. It was a one week course I think from memory. Intercode was the auto-code which was replacing machine code. (Editor: Intercode was an Assembler language).
I joined LEO at the time that the LEO III had been announced. It was the first transistor machine. I never had anything to do with the LEO II, and Intercode was the main programming language for LEO III. Prior to that, of course machine code was very much the main programming language. (Editor: LEO I and LEO II were programmed in an assembly code featuring, amongst other facilities, relative addressing and macros) After my one week Intercode course, I was asked to teach the subsequent two or three courses for customers.
At the end of that period, I was asked if I would work for Bob Gibson, who was doing the language definition for the CLEO compiler, which was the high-level language destined to replace Intercode. These were the very early days of Grace Hopper and COBOL, so I read up on COBOL and so on and I did some work on the language definition. But after a short while, I went to David Caminer and said that if he wanted to get anywhere on this high-level language he really ought to employ somebody who knew something about high level languages. He took my advice and it passed on to other hands. 
I was then moved to being a junior sales consultant, or Junior Consultant as they were called in those days, this is at the end of 1961, working for Mike Jackson on production control. Of course we weren't really into production control systems at that time, so this was all new stuff. Mike and I went up to Glasgow to bid for a contract with Singer Manufacturing Company, famed for their sewing machines. This was my first exposure to a very, very large industrial company. They had 20,000 people working on a two and a half acre site. At one end of the factory they took in pig iron and trees, and at the other end came out sewing machines. They made every nut, bolt, screw and casing for these machines. It was totally vertically integrated. We tried to propose a cost control system. We talked to lots of people, none of them understood how their cost control system worked. We were not successful with the contract. I can't remember who got it, Honeywell, I think. What I do remember is that they went through about four or five different manufacturers 
before eventually closing the plant, which was unable to compete with the new Japanese sewing machines. Today, not one of those 20,000 people remain in Glasgow.
After that in early 1962, I was allocated as a Junior Consultant to Doug Comish. We were just starting to get moving on LEO III sales. I think Doug had a target of one or two machines for the year. He sold six or seven machines, which was absolutely amazing. I was his bag carrying salesman, so he was doing the selling, and I was just getting in the way. Our focus was on Retail customers, and one of the more famous accounts which we won from IBM was the H J Heinz account. 
We also sold a machine to Cerebos Salt and later in '62, I was made on-site Systems Consultant responsible for helping them to recruit programmers and to start programming.
Cerebos was in Willesden. The main thing I remember about Cerebos was that when I joined they had a lot of trouble working out where I fitted into their hierarchy. They had seven different canteens, and the main problem was which canteen I should be allowed to eat in. I think I ended up somewhere in the middle management canteens. 
Anyway, I was at Cerebos when David Caminer asked me, in the middle of '62, if I would go out to South Africa as the Sales Manager for the joint venture with Rand Mines. They had the second LEO III, LEO III/2. The first LEO III was a bureau machine in Hartree House where it was doing various bureau jobs. LEO III/2 went out to South Africa.
LEO Computer Services, as it was called, was being run by Leo Fantl. He was one of the original LEO I programmers.. We had a Native Payroll of 120,000 staff which was already up and running, I think, or just going live when I went out there. It was a pretty important payroll, The data transmission was a native in a van, who went down the Reef in one direction dropping off the pay slips, and coming back in the other direction, bringing the data back to the machine. That worked pretty smoothly, except for the time that he was in an auto accident when, of course, the data didn't arrive. We were always pretty frightened because if you didn't get the native pay out on time, crowds of people with pangas were liable to come and introduce themselves at the computer centre.
 We were also handling the Mine Stores. All the bureau work was done by LEO Fantl and Joe Crouch. The machine itself was maintained by Lew Weatheral who later on became the managing director of LEO Computer Services, and was subsequently fired. I was supposed to sell, what they called mainframes, or very large computers, to other companies. There were really no computers in South Africa at that stage. There was an ORION machine (Editor::ORION was the LEO III sized business computer manufactured and marketed by Ferranti) at SA Mutual. There was another machine at SANLAM, and IBM had a machine in the Government science research labs, and that was about it. These machines were very, very difficult, if not, impossible to sell, and most people hadn't heard of computers at all. This was also true, in the UK. When Doug and I were selling, more of the time was spent explaining how a computer could be used for commercial applications than was spent on getting the prospect to choose a particular computer. It was a pretty frustrating time. I did eventually manage to get a company called Consolidated Glass on the hook for bottle design. But apart from that I didn't really manage to sell any computers at all. 
Anyway, at the end of '64 I said I wanted to go back to the UK, because it was pretty much a dead end really, and also the politics in South Africa were poisonous. Of course, in business, selling computers, you couldn't participate in the politics. Because if you did participate in the politics, and a lot of my friends did, then you very quickly ended up being locked up.
JF: I’d have thought that you wouldn’t have wanted to get involved in South African politics.
NE: Our friends were diminishing, more and more of them ended up behind bars. I got married while I was out in South Africa, so we decided to come back at the beginning of '65. I negotiated with David Caminer, and he offered me the job of looking after Local Government accounts. But he had a lot of difficulty finding somebody prepared to go out and replace me in South Africa. By the time I got back in the beginning of March of '65, my job had been given away to Tim Holley. And so I ended up as David Caminer’s, who was the Managing Director in those days, (Editor: Sales Director, until he was displaced by Ken Barge) ending up as his ‘gofor’, which, I suppose, was quite good training for being a ‘gofor’ to other people later on in my career
English Electric had bought LEO while I was in South Africa, so we didn't know anything about the English Electric acquisition at all.. But when I got back David decided that we needed to re-brand the company, the LEO company, as the new English Electric LEO Company (I was asked to find a new name for the company and Wally came up with ‘Computer Makers’, but Lord Nelson decided that it was to be English Electric Computers!) I was put in charge of the corporate rebranding, which I did with a company which David selected called Wolff Olins with a chap called Wally Olins, Wally Olins subsequently became the ultimate corporate branding person in the United Kingdom. He did Orange and lots of other companies like that.
We re-did everything, every sign in the company, all the visiting cards, all the brochures, the letterheads, the layout for the secretaries to type things, how reports were to be reproduced - we did a thorough rebranding job. Towards the end of that period, because we were just beginning to announce the System 4 machines, I also wrote the brochures for the System 4 launch. Of course in those days the machines were launched two years before they actually arrived. At the time we were launching System 4 we were in the middle of delivering LEO IIIs, so you couldn’t get a System 4. 
So towards the end of 1965 the company offices were moved from Hartree House in Queensway to Stag Place in Victoria. I took over from John Smith as the Consultant, I suppose Senior Consultant, responsible for The Post Office. 
Of course these days The Post Office doesn’t have a very exciting ring to it. But at that time The Post Office was THE main hub of activity for Government in computers. It's worth reflecting on how this came about. Because if you go right the way back to Bletchley Park and all that, and the Bombe and so on, when they decided that the mechanical Bombe was not really fast enough, then they decided to build an electronic decoding machine. Then they had a look around to say, "Who do we know who knows anything about electronics?" The only people in the UK who really knew anything about electronics, at least on the defence and government side, was The Post Office at Dollis Hill. So Tommy Flowers was recruited to build the electronic Bombe, the Colossus, which in the end came too late to be of much use during the war. But this meant that Dollis Hill was regarded as the ultimate authority on electronics. (Editor: Because of its widespread implementation of Telephone Exchanges the Post Office, and Dollis Hill, in particular was regarded as the natural place to seek the expertise to implement Max Newman’s design for the Colossus. Despite what Ninian says the Colossus did play an important part in the final victory in the War Colossus cracked the more complicated Lorenz code. The Bombes were electro-mechanical plug board controlled machines which helped crack the earlier Enigma code). 
When LEO I came along and commercial computers started to be making themselves felt, and the Government decided that it needed to do something about computers in Government, then it turned inevitably to The Post Office, and selected The Post Office to be the pioneers of computing in Government. 
(Editor: There had been a number of early implementations of computers such as those at the Ministry of Pensions & National Insurance, Board of Trade and H M Customs and Excise, but when Wedgewood Benn was appointed Minister of Technology by Harold Wilson he decided to accelerate the introduction of computers. Having been Minister of Posts, he made use of the Post Office to spearhead this endeavour, relying heavily on the technical side on staff from Dollis Hill.)
Their brief, as far as I could see, was to work out how you could use computers to do Government work, commercial work, and to encourage the British computer industry. But at the same time it was a brief not to make the British computer industry too comfortable, so they should on one hand give them computer orders and on the other hand they should torture them in the process to make sure they kept on their toes. When I went to The Post Office, David Caminer and Mike Josephs had sold the first two LEO IIIs for Telephone Billing and these machines had been installed in Charles House in Kensington. And Mike Jackson, I believe, had written the programs for Telephone Billing. My job, you could say, over the period I was there, was to sell and install the successor machines for the application roll-out, the LEO 326, and simultaneously to sell the following machines, System 4s, which had been announced and were destined ultimately to replace them, we thought. When I took over we had these couple of machines LEO IIIs doing Telephone Billing. They used the mark-sense Uptime reader, on which the exchange operators marked the calls. Half a million of these cards were read each night at Charles House and then the bills were produced from those. (Note from Tony Morgan Chief Commissioning Engineer LEO: Exchange calls were hand-marked onto what were called 'Trunk Tickets'. These went through some device, presumably American, which produced the 40 column cards (with round holes) for the brilliant 2000cpm Uptime card reader from the States. wo operators were needed with the two PO designed wooden racks for the card trays for each reader and 2 readers could be on the go at once. In later years this data was captured from terminals via Front End Processors and 2960s, and transferred via magnetic tapes and the Standard Interface Assembler to the LEO 326s. The application was called TOLD , Telephone On Line Data).
We had been taken over by English Electric in 1963 and it had been decided to replace the LEO III with a faster version, initially the 360 and then the 326, these being the speed of the stores - so six micro second and a two and a half micro second core stores. But English Electric made a fatal mistake. The LEO III was designed by John Pinkerton who had designed LEO I, after he had been recruited from Cambridge at the age of 27. The LEO III which he had produced was the most beautiful, very tidy machine, which worked immaculately. We had one in South Africa. It was very difficult to support any machine from the UK in South Africa, as you can imagine. We still were able to work this machine 6000 miles away. But with the LEO 326 they decided to replace John Pinkerton with an English Electric designer, because English Electric had obviously bought LEO and knew everything, and they appointed a chap called Tony Headley, a Yorkshire man, I think, to design the LEO 326. (Tony Morgan notes: Tony Headley had originally worked for John Pinkerton at Minerva Road. He was the design authority for IIIF - LEO 360 and LEO 326. He became Development Manager at Minerva Road. The Standard LEO III, IIIS, was designed by Paddy Salem, who had a nascent LEO IV on the stocks with integrated circuits and a rod microprogram store. It was lost in the IBM-compatible System 4 era)
And basically the LEO 326 was the most appalling machine, it didn’t really work at all. It had lots and lots of earthing problems and was very, very edgy. It kept blowing up in the middle of the night, I spent endless hours and hours with Jim Moody, who ran Charles house. When he got exasperated he would ring me up and I would get into the car and go down in the middle of the night to hold his hand. Not much I could do actually. While the engineers, Peter Eversden, crawled all over the machine. Peter even re-wrote part of the Operating System (Master Routine) although he was only the site engineer. But we did get Telephone Billing going on the LEO 326 and subsequently we decided to install a pair of these machines in each of the Post Office regions. So we had a pair in Bristol, a pair in Portsmouth, a pair in Derby and a pair in Glasgow, These were big machines and basically they were a million pounds a pop and I suppose in today's world that's around 10 million pounds a pop for one machine, 20 million for two and, of course, I persuaded them to upgrade all these machines to the maximum number of Ampex tape decks and the maximum core store and so on and so forth to get as much money as possible out of these machines. So I sold, I don't know, LEO 326's, probably, I can't remember, around 10 of them. At the same time I inherited the programming job for Premium Bonds in Lytham St Annes. for which the Chief Programmer was Alan Gudge. And that was a set of programs, unlike Telephone Billing which worked perfectly from the beginning, it was a programming enterprise which never ever seemed to come to an end. At one stage I had to go to the Director of the Post Office and say that, although he had had 300 girls punching for take-on on paper tape for the last two years, that the cheapest and most effective thing would be throw all the work away and start all over again.[laughter] So between the non-working machine and the non-working Premium Bond programs I really had baptism of fire.
I had a lot of programmers, because although The Post Office put their own programmers on, they relied on us to supply as many programmers again, so I had about a hundred programmers working for me on The Post Office account. On the sales side I had a couple of assistants, three, if you include the one who dealt with punch cards, and Rob Tubb, and Rodney Seely later on on the sales side - and then the Chief Programmer and all these progammers,. We also wrote the programs for the Savings Bank because The Post Office was responsible for what we think of today as Telecoms, for Posts, National Saving, and The Savings Bank. The Savings Bank had 20 million live accounts, 20 million dead accounts. So, we wrote programs with The Post Office for those and during the course of that with my chief programmers Trevor Davies, following on from Alan Gudge we developed the LEO methodology for programming writing, which had come down to us from LEO I and LEO II, into a methodology for building what were the largest systems programs in the country at that time.
I mean you need to bear in mind that we had to fit all this in 12,000 words of store and the program with 130,000 instructions was a very large program because these were Intercode instructions. So we had a methodology for specifying what was required very precisely in writing and getting it all signed off by the customer, and for writing the programming specs. In the latter years I used to get them to actually write the programs once very quickly, in six weeks, to scope the size, and then throw all that away, re-scope and flowchart the final thing, and then write that. Because writing the program really was the easy bit, testing it was the difficult bit. So the other thing I used to do was, at the same time as we started writing the programs, I would set up the test programming team in parallel with it, so that by the time we finished the programs, we had a set of test programs, automated test programs, which were used to validate the work that had been done. 
Now as a side-line during this I was trying to sell the System 4 to the Post Office against competition from IBM and everybody else which I was successful in doing and I sold quite a few System 4s, I can't remember how many but four of them went up to the National Giro which was Wilson's (Editor: Prime Minister Harold Wilson) pet project, which again fell to The Post Office because they had set up a bureau and one of the first projects was the National Giro. It was called the National Data Processing Service and it had to be set up by an Act of Parliament and I used to go along to the Parliament in the evening and listen to the parliamentary sessions to get The Post Office license extended to allow them to set this up.
 
NE: We actually wrote the Giro programs as a turnkey project, The Post Office didn't have any programmers, they did the Systems Specification under my guidance then we wrote it as a turnkey program. We wrote the whole lot on time on budget. We were well ahead of the (live) customers, it took a bit of time to build up the Giro even though they had captive customers for things like the Electricity and Gas Billing and so on
And then, towards the end of this period, having implemented these very, very big Batch systems, I would say having pioneered very large commercial systems, I then got involved in online real-time systems for the first time. Again there wasn't really anybody in this country who knew anything about online real-time systems and The Post Office had this contract from Customs and Excise and the Airlines to do the cargo handling system for Heathrow airport.
I wrote the proposal for that system. I went through the, you know, what was necessary for real-time systems, all pretty much from scratch, and we, I, decided I suppose, that we needed a different Operating System because System 4 wasn't much good at MAC systems and probably wouldn't have been much good at TP systems, as we call them now. So I commissioned a new Operating System from an external company called Computer Sciences that was called LACES Monitor, and that Operating System worked on the System 4 machines for that project. Although by that time I had moved on, it was later transferred on to the IBM systems, which also ran using the LACES Monitor which I had commissioned. 
And then, I suppose, at the end of that period, or in the middle of that, we were taken over by ICT, because in the early days of LACES I was doing these tenders and the Managing Director was Bill Barlow, of English Electric at that stage, and then, as time passed, because, I don't know how many times I had to do the proposal for the LACES System, many times, Arthur Humphreys was the Managing Director by then, and I used to have to go along to Arthur and every time I went he said: "Look at the P & L, you’ve bloody changed it again." But anyway, I was dealing both in person with Bill Barlow and then with Arthur on this, because for those days, it was a very, very big contract. 
So then in 1968 ICT bought us as part of a Government plan to rationalise the UK computer industry. During my latter years at the Post Office, I suppose the last year of LEO and of English Electric LEO, the Post Office probably accounted for two thirds of the turnover of the company. There was nobody else who was able to sell System 4 because they all knew it was going to go out of production, out of business, LEO IIIs were too old, I was the only person in the company delivering and producing revenue, sure, there were one or two other people doing it. And so I was not really involved in any of the fighting over the various different accounts, because it was pretty clear that The Post Office account was a LEO account. I don't think ICT made any bid for The Post Office account, although, you'll hear, if you talk to Doug Comish, that there was a lot of very nasty fighting going on,. As far as I was concerned, the only problem that arose for me was that ICT, which was a very hardware sales driven organization, was that they didn't have anybody like me in ICT, so, they said I must be a Salesman and then David Caminer said: "We didn't think that was quite correct." And Doug Comish said it wasn't quite correct, and anyway, after a certain amount of skirmishing, and one thing and another, in a short period in Marketing, it was eventually agreed that I would be a Sector Manager in the UK. I think there were five Sector Managers, or something, in the UK Sales Organization. I was made the Sector Manager responsible, countrywide, for Financial Services.
So, that was basically my LEO exposure. Obviously you don't want, particularly, for me to go on, what happened to me after LEO, maybe a word of two. I suppose I was the, apart from one or two engineers, probably the last survivor of LEO in the companies which succeeded it, I mean in ICL and all the rest of it. 
In ICL I had a period in charge of Government Defence and Universities and Research sales. When Rob Wilmot came in as CEO, after the company got into difficulties, he made me the Director of Product Marketing, and his ’gofor’, which I was well trained for, having worked for David Caminer, the most difficult person in the world to work for., although Wilmot ran him close. I was then, successively, the President of ICL International and ICL Europe, those are sales organizations. When I finished my career, this is how the company was organized. . We had a Managing Director, who was Peter Bonfield, a Finance Director, who was Keith Todd, a Personnel Director, who was Don Beattie, and then the Executive part of the company was divided into three parts, which were known as the ‘Jedi’ There were three, Group Executive Directors, one dealt with Sales, one dealt with Customer Service, and one dealt with Products. I was looking after Products, which was product development, manufacturing, personal computer distribution, etc, etc. - anything vaguely technical. That was my final job in what was by then, was Fujitsu. 
JF: Well then, thank you very much for spending time with me today, to talk about your experiences. There's just one small statement I'd like to read to you: It's the interview with Ninian Eadie, has been recorded by the LEO Computer Society, as part of an old history project to document the earliest uses of electronic computers in business applications. Any opinions expressed are those of the interviewee and not of the Society. The copyright of this interview, as recorded form and it's transcript, remains the LEO Computer Society 2012. Once again, Ninian, thank you very, very much. It was very interesting talking to you.
NE: Thank you.
Postscript. Ninian competed in the sailing at the 1964 Tokyo Olympics and was still competing in 2018!

Provenance :
Recording made by the LEO Computers Society as part of their ongoing oral history project.

Copyright
The video above may be subject to copyright of the original author and is present only on this website for non-profit educational purposes only. If you are the original copyright owner and wish to have the media removed from this website please contact us using the relevant contact email address listed on our contact page.