Transcript :

LEO COMPUTERS SOCIETY  -  Oral History Project
Interview with Ray Hennessy by John Daines, 27 April, 2012.

John Daines :  This interview of Ray Hennessy has been recorded by the LEO Computers Society as part of an Oral History project to document the earliest use of electronic computers in business applications.  Any opinions expressed are those of the interviewee and not of the Society.  Copyright of this interview in recorded form and in transcript remains with the LEO Computers Society, 2012.
Interviewers are volunteers drawn from the Society’s membership.  My name is John Daines.  I started working for LEO in 1961 on LEO II and stayed with the company and its successors for 40 years.
Right, Ray.  Perhaps you would like to introduce yourself and talk about your early life before you joined LEO.
Ray Hennessy :  Thank you, John.  
I started in LEO in 59, as you know.  Up to then I had little idea what I wanted to do with my career.  I was born in Sutton in Surrey in December of 1934.  My parents were both working on the Daily Herald at that time.  They were committed Socialists and had been very active in London politics during the 20s.  
My father had been a reporter very briefly in 1926 before the slump and the General Strike had caused cutbacks and he spent the rest of his time at the Herald typing stories that people phoned in.  My mother was a librarian at the Daily Herald and she worked in what they called ‘the morgue’ where the cuttings were stored.  Every day they would go through all the daily newspapers and cut out interesting stories and file them.  It sounds dull but I often helped her on Sundays when she was still working just after the war and I found it fascinating, the amount of historic material that was stored in newspaper files.
As I say, they were both long-time Socialists and they met in Young Communist League in 1926 or soon after when there were fights going on between the communists and the various fascist groups.  I think they probably had quite an exciting young period but I never asked for any real details about that, which I now regret.
During the war, which was not really affecting us too much, I had a road accident and my parents decided, with that and with bombing, I really ought to be evacuated and I went with my grandmother to South Wales for just under a year.  It wasn’t a particularly happy time there and I found that the schooling that I was getting in Wales was far behind what I had had in Sutton.
I think I was a fairly bright kid and we have records that my father and his brother used to exchange messages about how one of my cousins and I were getting on.  It was a bit of a competition, I think, but I never had any pressure to achieve.  So then I went back to Sutton in February 1945 and the schooling I had had in Wales was virtually all wasted so I lost a year in school and was quite old when I took the 11+.  However it possibly helped me because I was able to get a scholarship to the local public school, that’s Kings College School, Wimbledon (KCS).  This was a scholarship which paid for absolutely everything; it wasn’t limited in any way. 
At the school I was very much an outsider because most of them were boys whose fathers were middle management [and] some of them were top management.  It was a fairly minor public school but it had a good educational reputation.  Being the only pseudo-communist on the whole school roll, I tended to stand out a bit.  One of the things about the public school system was that, at thirteen you go from what is called the ‘junior’ school into the ‘senior’ school and you take Common Entrance and at that point a lot of new boys came into the school.  I managed, rather gloriously, to fail English at that level.  I found English very difficult; I was quite a late starter and I didn’t really read much, except, probably, comics so the English test was always a bit of a challenge.
Anyway, I got into the senior school and had to spend an extra year there because of my poor performance and then took O-levels.  I think I got seven O-levels in French, Latin, English, Maths and General Science.  I managed to fail German and I took it again the following year and failed it again, so I thought that languages perhaps weren’t my forte.  
I then went into the sixth form and decided with advice from an uncle who was a farmer that perhaps I could do Agricultural Economics which would be very useful for travelling the world and advising small countries – third world countries – about the use of careful farm management instead of just living hand-to-mouth.  What did I know?  Not very much in those days.  This meant that I went into the Biology science sixth form.  I took four A-levels in 1953 and managed to more or less fail the Botany one which was the subject I was supposed to be taking; not very clever, all things considered.  By this stage I was nearly 19 and I took the Oxford Scholarship just before I was 19 and didn’t get an entrance or scholarship on that occasion.  Went up again in March [1954] but by this stage I was too old to get a scholarship so I just took an entrance exam which was really very, very simple – about Olevel standard – and got a place at Exeter College, Oxford.
I then decided I would take A-levels; I took the Botany A-level again though it wasn’t essential that I did so, and I also read up the text books for Applied Maths and Pure Maths and I got those three A-levels in 54.
In those days we had a choice whether we did National Service before or after going to university and I decided that really I wasn’t up to university yet and took my ‘King’s Shilling’, well, four shillings as it was, to do National Service.
National Service was, for me, very interesting.  I know that a lot of people found it boring or strenuous but I joined the RAF and refused to go for the officer training option and was trained as a radar fitter.  I asked to be posted to south-east England, we all had to put in a preference, and actually got posted to south-east Asia, which, I suppose, was close!  I spent two months in Singapore and then something like seven months in Kuala Lumpur as a radar fitter and I found the experience extremely interesting just to see how the rest of the world lives.  It did make me think again about my Agricultural Economics ambition.
When I finished National Service, which was in south Wales because I had come home early to take exams at Oxford, I then started at Exeter College and was there from 56 to 59.  I got a reasonable degree – a second which was quite remarkable given that I didn’t really do very much work, but I certainly enjoyed the experience.  One of the things that I did decide very, very quickly was that Botany and possibly Agriculture weren’t for me.
Since I had been reasonably good at Maths I thought I’d have a look round and see what opportunities there were in the mathematics arena.  I wasn’t too keen on trying for a university place or a teaching place because I hadn’t really studied Maths for quite a long while –five years by then – but I thought I could use my ability in something.  So I looked in various journals; there was a Statistics journal and an Accountancy journal and one of the common threads running through all the job advertisements was ‘computers’.  
There were four or five firms offering places with computers at that stage.  I suppose I was being reasonably selective but what I wasn’t able to do was to determine what the job actually meant from the advertisements.  So I had four interviews.  
I had one with ICT and the ex-RAF man who interviewed me said that I would be a salesman, and when I said no, I really wanted to do programming because that looked interesting he said “No, no, no. We’ve got gels to do that.”  I learnt much later that, at that stage, ICT only had plug-board programs.  I don’t think I can go into how that works – not on this interview – but it was a very elementary form of programming punched card sorting machines.
I went up to Kidsgrove for an English Electric interview and was interviewed by Wilf Scott and Conway Berners-Leei.  They offered me a post fairly quickly in Stafford.  They were offering £750 a year at that stage but Stafford was well outside the area I wanted to be in which was very much the south of England, preferably London.
There was an interview I had at Borehamwood with Elliot Automation. They had mostly process control and small scientific computers at that time; I think the 803 was just coming on stream then and the work they were proposing was not particularly enticing and also Borehamwood would have meant probably moving out of London.  I think ‘reverse commuting’ would not have been particularly useful.  So again, although they offered me £700 a year to work there, it was not really an interesting offer.
LEO was quite different.  I had, first of all, an aptitude test and I think that might have been at Elms House.  Elms House was one of the Joe Lyons offices just opposite Cadby Hall in Hammersmith.  The test consisted of various elementary programming tasks.  We were given guidance on what sort of operations programs could do and asked to do various things with this.  As far as I’m aware I think I did pretty well on that; later evidence suggested that I did.  
I was then asked to go for interview at Hartree House.  Hartree House was a new computer site, which LEO Computers had opened over the Whiteleys department store in Queensway in W2.  That was quite interesting, just to be there and actually see a computer, which was a little bit startling – the size of it.
I was interviewed by Doug Comish and I got on quite well with him.  I eventually actually worked for him – he was a senior Sales Manager.  While I was with Doug someone came in and said, “Lets go along to my office.” So we walked along the corridor and this man whoever he was, was only interested in whether I did any sport.  At the time I was only umpiring hockey matches for the college, having played in the first two years but for the third year I thought that perhaps I ought to be mainly revising?  Anyway, when I said I did umpiring he lost interest entirely and said, “Right, get on” and we went back to Doug Comish’s office.  Doug asked me why on earth I’d said that and I said it was true, which didn’t hold much water as far as he was concerned.  I said, ‘Well who was that, anyway?” and Doug said he was David Caminer, the Marketing Director who I did meet later on.  But Doug seemed to be impressed and offered me £700 a year to start in September and after a summer working on a beer lorry, in a bar and the college library, I duly did.  
So really there wasn’t too much question which job I would take, it was more: how much would they offer me.  Doug offered me on the spot £700 a year, which was the same as Elliots and a bit less than English Electric, but as far as I was concerned the type of job looked to be much more suited to my ability.  I had a flat in London, well, I shared a flat with four girls which was interesting in itself, so working in London was absolutely straightforward and very good.  I started in LEO in the middle of September 1959.  
 JD: Okay, so that’s your general background and takes us up to where you have joined LEO. If you would like now to talk a bit about your career with LEO: which computers you were involved with, and what were some of the roles.  What sort of things do you remember, impressions perhaps of some colleagues or managers, any memorable incidents, including any funny ones if there were. When you joined did you feel you were at the forefront of innovation?  Just generally about your time.   
Ray Hennessy: Well, I joined LEO as a very naïve undergraduate with some experience of the world but a very superficial view of what things were going on, so as far as I was concerned it was the job that mattered and I had no idea at the time that this was really ground breaking stuff that we were doing.  Since then it’s become evident that LEO was very much a pioneering company, which had an ethos of finding things that needed sorting and sorting them and getting on with sorting them.  
I started my programmer training immediately at Elms House where I’d had the first aptitude test.  We were only there for a week when the training was transferred to Hartree House, over Whiteleys.  This was where I had first seen a computer, the LEO II/5.  When I got there the industry of the people who all seemed to be staying on past five o’clock or 5:30, end of day time, rather amazed me because I was used to a student life which was really quite lazy, I would say.
Anyway, I buckled down to the training course, which was fascinating.  It was something where, every evening, I would type up all my programming notes and make sure that I knew exactly what had happened each day.  And I think I was a pain in the backside for most of the instructors.  It was always the raised eyebrows from people like John Smythson when my hand went up to ask yet another question which most people seemed to think was very naïve and not worth asking but I tried to tease out the funny by-ways that computers actually had in those days.  Of course we were programming in something very close to machine code so it was quite low level and I found the whole concept of programming strategy and structures absolutely fascinating.  
One of the instructors on the course was Mike Josephs.  He was talking about restart facilities: how you recover a job which has failed or you run a job again because, for instance, a magnetic tape had become unreadable, and that was a very intricate subject which I don’t think Mike fully understood himself.  So again I was asking all these little teasing questions, which must have put everyone else off me, I think.
Other people on the course were Mike Trevanion, who eventually went into a central organizational role, I believe associated with the operators and you met him, I think, John; …
 JD: Uh-huh.
 RH:  … Alan Gudge who stayed as a systems analyst and programmer for most of his life and was still at the eventual company when I left; and David Yates.  David Yates was quite interesting because he obviously picked up programming very, very quickly.  He’d got a degree, I think, from Cambridge, and one of the things that amazed me was he left after a year and went to the National Physical Laboratory, where he became a very important research physicist.
Now, one of the introductions to the course was given by David Caminer.  David Caminer, whom I’d met briefly at my interview, was a very irascible person, but as Marketing Director he was involved in every sale that LEO was attempting to make both in the bureau and to sell the actual LEO machines.  What I didn’t know then, and have only recently discovered, was that he effectively invented system design, flowcharting and the whole idea of system analysis and conversion into programs.  He was the brainchild for all of that particular aspect of computers.
After the course, which was a five week course, my first job was very, very odd.  I was given a pack of cards and a print-out and told to create the listing for the program.  The way we wrote programs in those days was, we had a coding sheet which had fields for what we now think of as reference addresses; we had to list all the reference points and the coding sheets would measure an offset from the reference points.   So we would have, say, ten storage words for an Address and the program would list that as  “Address” plus a number.  These sheets were then punched up onto paper tape and ‘interpreted’.  Each of these addresses would be converted into an absolute address in the machine store.  What they wanted me to do was to take the list of absolute addresses and reconvert it backwards into a listing which included all of these ‘symbolic’ addresses, which was quite challenging and not actually particularly easy to do.  I spent three weeks doing that and about 30 years later I learnt that the listing that had gone astray which I was asked to recreate had actually been found while I was doing it on the top of a cupboard, but no-one liked to tell me about it! 
It was at the end of those three weeks that I was moved into a main programming office where I worked with John Aris and Mike Jackson on the British Oxygen project.  The British Oxygen people had ordered a machine.  It was going to be LEO II/7 but we were actually developing the Invoicing and Stock Control suite on LEO II/5 which was the machine at Hartree House; it was basically a bureau machine so British Oxygen had rented out so much time on it for us to do our work.  In practice, the task that they presented us with, which was very well specified, was actually quite a lengthy task so we did most of our work on their own machine.  Their machine was installed in Edmonton in northeast London and we started going up there sometime in the middle of 1960.  It was quite a trek.  John Aris who was the senior programmer at that stage used to pick me up in his little Morris Minor and we’d bumble our way through northeast London to get there.
The machine they had - I can’t remember the exact configuration – but it had a full complement of drums and magnetic tape decks and printers and we had, I suppose, something of the order of a year to get this Stock Control and Invoicing job programmed.  We finished it more or less on time and the overall result was that we produced a system which worked – John Aris was very, very good at checking programs and making sure there were no glitches in them before we actually went on the machine.  But, of course, the customer having not really seen a computer system before now said “Oh, we don’t want it like that; we want it like this.” So the next thing we did was to completely re-write the suite.
John Aris was a – well – I would call him a genius in his own way.  He was a very dedicated person and was extremely clever.  He just failed to get a first at Oxford and I must admit that I could never understand why he didn’t get one.  One of the things he did was to produce a Restart package.  Restarting was essential for computers in those days because valves used to fail, little bits of the circuitry might become corroded or wires might just slip off pins and things were noticeably more fragile than they are today.  The idea of a computer working untouched for days upon days upon days was almost unthinkable.  We had permanent engineers on site to put things right as quickly as possible, but what it did mean was that, if the machine had gone wrong, you had to restart your job.  You could always, of course, go back to the beginning and do it all again, but since some of the runs would take anything up to five hours, doing it all again would use up a huge amount of computer time.  
What John devised was a system of restarting which enabled us to put on all the magnetic tapes a ‘restart point’ every 20 minutes or as short a time as you like – long or short as you wanted.  When we put that restart point on the tape, on one of the tapes, or possibly two of the tapes, we also dumped the whole program and we also put, in front of that, a loading routine.  And the last thing we did was to punch a little card.  This card virtually had just a number on it and that was the number of the restart point so, if you had a hundred restarts in a run, or 500, it really didn’t matter, there would be a number in a field on the card which was the number of that restart point.  
To restart a job, all the operator had to do was simply load that card into the card reader.  Nothing else had to go in at all.  The card had a very short program which wound the tape with the program dump on it back to any restart point.  And at that restart point it loaded the program from the tape and that is all that card did.  The program loaded from the tape would look at the number on the card and say, I need to go to restart point X and it would then rewind all of the tapes forwards or backwards to restart point X.  Having got to that point it would store all of the restart control totals, whatever was required to enable the whole run to continue and away it went again.  It took of the order of two minutes to do a complete restart and everything was completely automatic.  I don’t think many other people used John’s package as British Oxygen did, but I know that the company incorporated the technology he’d invented into its various products . 
There were some other people working on the project that I can remember.  I don’t remember the names of any of the engineers but I know that they did reappear in my later customer jobs.  The two programmers who were working [there] were Sam Waters and Keith Davies.  Keith Davies eventually became a Regional Manager for various overseas areas of the company and left sometime in the 1980s. 
Sam Waters was quite different.  He had been brought up in Shepherds Bush and left school having taken some O-levels and got them, which was fairly unusual for that area and he then joined Lyons, [it would have been in 1956 or 7] as really just a clerk.  He was quite good at maths and when the LEO I computer site was being built up he was asked if he wanted to become an operator.  So he went over and operated LEO I for about a year.  At that point his obvious skills were recognized and he went on to being a programmer.  The thing about Sam was that he was absolutely meticulous.  You could guarantee that when Sam produced a program it had been bench checked before it ever got near the computer so that it was virtually perfect.  It might not do exactly what was wanted, it might need amending, but Sam was very, very good at getting his programs right – something I could never do.  He left LEO about three years after the period I am talking about and went to LSE as a lecturer in ITii.  I couldn’t understand why he was doing that as he had to take a cut in salary to do it but he said he was fed up with getting home and having his children saying “Hello, are you my new uncle?” Which he felt meant that he really wasn’t enjoying their childhood.  So he took a 9 to 5 job at LSE.  Interestingly he eventually ended up as a professor at University of West of England in Bristol University which, for a lad who had come from the back streets of Shepherds Bush was, I think, an absolutely majestic achievement.  
The British Oxygen team was working for about five years. In the middle of it John Aris was promoted to team leader because Mike Jackson moved to Freeman’s. Freeman’s was a mail order company, they were developing an IT Department and were very keen on using LEO computers and they had quite a few bureau jobs working with us. 
When we’d finished the British Oxygen job, I think sometime about the end of 1960 [actually about April 1961], the next thing I did was to be the senior programmer on a team writing for the Hotpoint Bureau job on LEO II/5. I must admit I don’t remember too much about the actual job but people I was working with included Richard Beckinsale, John Parker, who eventually left and joined W D and H O Wills in Bristol and Mary Hambley, who I believe went to National Westminster Bank or to ICT. Other friends have met her since but I’m not too sure where she left to go to. We were at that stage probably supplying most of the homegrown computer programmers in the UK. 
I was interested at that stage in trying to broaden my experience and I had one or two interviews with other companies. One was a bureau which was in Canada.  The way they offered the job was to say “well, we’ll take your current salary and we’ll increase it by ten per cent”. So I said “so if I’d been earning five times what I’m earning now you would have paid me…, so it’s not pay for the job is it?”… and eventually I decided that I wouldn’t be able to work in this sort of environment where people doing the same job were being paid vastly different amounts of money. 
I also had an interview with CDC who were looking to get into various aspects of British computing, and I know they had a very intensive psychological acceptance test or interview test which went on for about two hours and at the end of it the guy said “oh you’ve done very well on this test, in fact I think you’ve done too well really for us to offer you a job”, which also struck me as a rather curious way of going about things but is now much more common, people being overqualified. 
After about three months in 1961 [actually late summer 1961] I transferred into Doug Comish’s team who were a sales team but my job was to write system specs. And the one I most remember is a specification for Midlands Electricity Board which took up two volumes of A4 binders. We didn’t get the order then but I believe we got an order for them much later. Certainly the work that I was doing at that stage was quite different from programming. Well, it was a different level of programming. I didn’t have to worry about the how things were done but the what was being done at various stages of the system, and that was quite an interesting new direction as far as I was concerned and I came back to that later on and more or less specialized in it. 
At that time most of the bureau jobs were being run on LEO II/5 at Hartree House but we had, I understand, a shift on demand at the British Oxygen machine up at Edmonton and a lot of our development work for bureau tasks was actually done up there. It was a rather long way away from our machine and we were constrained to put in trials overnight and that always slowed things down we found, so we were usually begging time on the II/5 machine, the in-house machine at Hartree House.
I suppose at that time we were starting to get fairly intensive competition from IBM and ICT. ICT had some rather small machines but they had a very large installed punch card base so they were able to move their machines into customers which had their punch card systems. IBM also had punch card installations and they were very active in international markets and they saw the UK as being ripe for plucking, which I suppose it probably was. They used to say in those days that no one ever got fired for ordering IBM equipment. I don’t think that still applies but it certainly drove the whole computer market for many decades. 
Now let’s see. I suppose at the beginning of 1962 I was transferred to a small marketing services unit run by Mike Josephs, it was just the two of us and a secretary and our task was to provide a focal point between the sales teams and the development and research teams so that we could at least get some idea of what equipment we needed to develop for future projects. My job was to produce a fortnightly listing of all of the development orders, bureau jobs, equipment that was coming along and so on and send that out to the regional offices. 
The regional offices by then were starting to appear in other countries. We were moving into the LEO3 era and we were setting up installations in South Africa, Australia and Czechoslovakia curiously. The Czechoslovakian project was quite interesting. Ralph Land was involved in that and he was instrumental in getting all of the sales force, that was anyone who was working in sales who wasn’t just a programmer, to go out to Prague for their first international computer exhibition and all of us went out there for one or two weeks. There were perhaps twenty odd when I went out there. I flew out with David Iggulden who was the press manager then. A very, very colourful character was David, very amusing; never seemed to be focused on computers but had some excellent contacts through journalists. 
At that exhibition I believe we actually clinched one of our first orders for LEOIIIs in Czechoslovakia, but we had a very successful sales project going on in what is now the Czech Republic. I don’t think there’s anything in the Bratislava half of the country but most of it was in or about Prague.
Mike Josephs was more or less office bound and his job was to collate the marketing requirements of the sales force and liaise with all of the production and research people to try and make sure that we were producing the right equipment. He became very expert at that but at some stage for some reason he became disenchanted with LEO and left and I never did find out exactly which company he went to, it was a north London company. 
The sort of people we were liaising with, apart from John Pinkerton, was Tony Barnes, who was the production manager. He was concerned that we weren’t putting his production commissioning and delivery staff under too much pressure with new equipment so there was a certain amount of infighting going on at that stage: what should we be looking at, what could we be looking at and what could we afford to be looking at were the sort of arguments that used to go on. [0:40:19]
In April 1963, which was possibly another reason why Mike Josephs left, LEO was sold by the Joe Lyons organisation to English Electric and English Electric LEO Computers was formed. We discovered that the management mechanisms that were going on in English Electric were significantly different from the Lyons ones. Lyons and LEO were very innovative, they were forever producing new ideas and new systems and new products and new ways of doing things. English Electric tended to be much more based on what I would call “traditional management style”. Everyone had a role and that role they carried out for year upon year upon year, there was no suggestion of retraining people or giving them new challenges or new ideas. This had proved very successful for English Electric and they were very central to the country’s electricity generating and transmission systems, which of course were starting to grow after the deprivations of the war.
English Electric had two machines, sorry, three machines that they were producing. Two of them were still very much under development but they had the KDN2, a very tiny machine which was used for process control and was quite successful, they sold many hundreds of those. For commercial work there was the KDF6, which is also a small machine. In our terms it was very much less than LEO2 or LEO3 but of course, it was very much cheaper. It was housed in one cabinet and that incorporated the operator’s desk so it was very much a bijou product. It was quite successful in terms of getting into companies. The third machine they had, which was at that stage still under development was the KDF9. This was basically a system, sorry a research systems product with very highly efficient scientific functionality. That was sold to quite a large number of universities and research establishments was probably the best of their bunch, it was certainly a quite a powerful machine. 
The managing director of this company was Wilf Scott who’d interviewed me what? about five, six years earlier. I don’t think he remembered me. I didn’t, I don’t think I really remembered him except for his name. Conway Berners-Leeiii was still around fronting up the research and marketing operation for the English Electric products. He soon got interested in LEO as well. So he was very much a polymath as far as computers went. He will come up again as well later on. 
The LEO activity at this stage was very much geared towards LEOIII and largely government and public, public utility and public local government projects. There were a few commercial users but most of the users were starting to look like being those operations which had to be seen to be British and we sold a large number of machines to government and utilities and a few to some local government operations. 
I was still trying to perform this marketing service function of gathering together information and circulating it to the field but English Electric had a very efficient operation up at Kidsgrove under Bob Menzies. They clearly were much more geared up to producing lots of documents and circulating them, producing glossy leaflets and so on so that that function was all transferred up to Kidsgrove and I then moved back into the front line systems design. 
So we’re up to soon after the merger, probably about the end of ‘63, maybe early ‘64. I was moved to John Aris’s team. John Aris had gone from programming into systems and then into sales some years earlier and he was responsible for all of the government sales, so my task was to liaise with the MPNI Newcastle site where they had LEO II/6 and I also took responsibility for the project of The Paymaster General’s Office where we were negotiating to supply them with the LEO III. There’s more to that story later. MPNI  (Ministry of Pensions and National Insurance), Newcastle was the biggest clerical office, certainly in Europe and possibly the world. There were 10,000 clerks working there. The pensions job simply printed out pension books, it was no more ambitious than that. Unfortunately LEO II/6 was a bit under-powered for the job as it was turning out, as more and more pensioners were living to a greater age. And one of the ways they were looking to convert was to buy a LEO III of some sort. This eventually became LEO (System) 4/50, and I’m jumping ahead a bit now but when the LEO (System) 4/50 was being sold we had bought in some RCA machines, I can’t remember the number. 
John Daines:  Spectra 70/45. 
 RH: That’s right, Spectra 70/45 which was also being sold by ICT under another badge. Our System 4/50, was going to be a revamp of that and in fact quite a big extension to its facilities. 
We were developing our machine rather slowly and when we were due to deliver to MPNI the machine wasn’t ready, so eventually when the machine was delivered about a year late we actually installed a second machine gratis for them to be able to do more work on it. I think that was probably quite a good deal for the government at the time. Yes. I had someone working for me on the MPNI projectiv so I  didn’t keep too close tabs on what was actually going on. What I did know however was that we were late with delivery and we had to work quite hard to make up the lack of confidence in the company on that occasion. I think we actually regained their trust because the company and its successors got quite a lot more business. 
Other jobs I was doing at that stage were Paymaster General’s Office where we sold them a LEO360 and that was then upgraded, and again because of late deliveries we had to deliver two machines to them to get their work done. I also looked after the LEO III installation at Customs and Excise in Southend. The Inland Revenue were starting to get very interested in using computers. And The Ministry of Labour (MoL) who while they were the  Department who paid Unemployment Benefit(UB) they didn’t actually have the computers for calculating UB - they were at MPNI. 
[Note. UB was a National Insurance benefit paid by MoL on behalf of MPNI]
MPNI later became Ministry of Social Security then Department of Social Security(DSS)
then The Department of Health and Social Security (DHSS) and is now called The Department of Work and Pensions (DWP) having incorporated The MoL. I’m not quite sure why government departments like to change their names, it must be a hell of a job trying to design stationery for them. 
When we were working on the Unemployment Benefit project which was at DHSS, Reading, John Aris used to go down there, again in his little Morris Minor, and he would come and pick me up. I was living in Balham in those days and so I would travel into town and he’d pick me up somewhere along the Underground line and we’d trundle out to Reading before the M4 was built all the way, which was actually quite an interesting experience with lorries and cars zipping past and John bumbling along in the slow lane. I think, probably his mind was working just as fast then as it normally did in the office. 
We sold quite a lot of machines to government and I seemed to get involved in various aspects of the system design for these machines and I therefore got to know quite a lot of people in Treasury O&M which was the authorizing department for purchase of equipment, not just computers but any technical equipment. They were quite interesting civil servants to deal with, you never quite knew what they were thinking, at least with the customers they had a very clear idea what they wanted to do on their computers. 
So we come to 1968 when I was still working with John Aris, and Tony Benn, the Minister for Technology, decided that, (for) the UK computer industry to become a force to be reckoned with in the world, (it) had to actually focus itself on one company. So he brokered a marriage between ICT and what was then English Electric Computers. I wouldn’t say that it was a marriage made in heaven, there again were quite significant management styles. 
 JD: Right, Ray that looks like we’ve come to the end of a quite significant part of your career once the new ICL company was formed in 1968 so perhaps if you could talk a bit about that now? 
 RH: If we thought there were problems merging English Electric and LEO we hadn’t really seen anything yet. The merger between English Electric Computers and ICT was very traumatic. There was quite a different management philosophy between the two companies. ICT were very much geared to the selling ethos, they were concerned to move product and to get into bed with purchasing officers and to get the equipment put in. I believe they were not quite so worried about the suitability of equipment that went in but maybe that was my LEO background. The English Electric people by this stage were much more concerned to ensure that the customer got equipment that would suit their requirements. 
It caused quite a bit of friction higher up in the company and a large number of people were put in very awkward positions, one of them I know, Peter Aylett, was more or less edged out and eventually retired really without anyone taking any notice of what he’d been trying to do in the company.  And David Caminer who had been working very hard in probably most of our large selling operations was moved into Europe; and John Aris was left more or less high and dry until he took up the Paris office. When he eventually came back from Paris probably about 1973/74, something like that, there was absolutely nothing for him to do, no one could think of any job for him to do so he left and went to W.D. and H.O. Wills who I think by this stage were known as International Tobacco and then later on became Director of The National Computing Centre. 
What I was doing at this stage was also rather under review and fortunately for me one of the jobs that did come up was a research project for what was then The Post Office. It was the arm of The Post Office which ran the telephone system and is now known as British Telecom. One of the things that they were interested in was managing and running data traffic across their networks. They were looking at the possibility of a switched circuit system which meant that data channels will be set up and would be switched through their main exchanges, or possibly what was becoming known as a packet switched system, I’ll talk more about that in a minute. But what they did was to get three companies to do some analysis of the market and some suggestion on how research might progress. It was ICL, which had just been formed, Standard Telephones and Cables, STC, who were responsible for installing most of the telephone switching exchanges in Britain, and Plessey who were responsible for private switching equipment. 
I spent the first six months working under John Winterbottom who was one of John Pinkerton’s senior research engineers. There were about seven of us drawn from all parts of the company at the time. There were people from Stevenage, the old ICT development installation, they were telecoms engineers. There were marketing people who had been working at Putney. There were other development engineers from Kidsgrove, which was originally an English Electric site. The overall requirement for ICL was to try to analyse what would be the best method of transferring data and what the loads might be in the future.
 The main control that we had at the time was exercised by a steering committee which was very hands-off so that all of the management of our project devolved on John Winterbottom, who was very knowledgeable on almost all telecoms techniques. Not the nitty-gritty so much as the overall strategies.
We also liaised with The National Physical Laboratory - who was that? Donald Davies was in charge of it then, and The NPL were building an experimental system which would packet data into packets of about 128 bytes, or maybe smaller in the case of their experimental system, and route them through networks which were not dedicated to specific connections between, say, a terminal and a computer. The network was general and was known as the packet switched system. The idea was based on something being developed by the main research project in America known as ARPA. I can’t remember what ARPA stands for. 
 JD: Advanced Research Project Agency. 
 RH: Advanced Research Project Agency, thank you John. And they had something called ARPANet which was linking government offices and universities together and was developing packet switching as a technology. 
In amongst our team we had all sorts of skills and ideas and there was a certain amount of, shall we say, thrashing backwards and forwards between people who said “oh we’ve got to do it this way” and other people who said “no, we’ve got to do it that way” and other people, myself included, who said “well, let’s look at both ways”, which was perhaps a slightly easygoing way of looking at it but in fact it paid off in the long run. In this first contract we spent six months and my job was to analyse the user market and I went round the company and I looked at magazines and I considered how many of the various types of terminals there were in use and we did a forecast. The forecast was rather finger in the air forecasting but we did actually produce traffic load forecasts for 5, 10 and 15 years ahead, that was ahead from 1970. And in 1985 I got the report out and actually checked the market because by that stage there were quite good marketing surveys and I think we were within about ten per cent of each of our forecasts at that time except Fax machines. Fax machines, which were actually data transmission machines at that time, had not taken off as people had expected, it was found that small computers were coming in and people were able to communicate between computers and there was no need to actually fax documents. I’ve a feeling that most fax machines at that stage were in solicitors’ offices because they actually had to have a piece of paper which could be shown by a paper trail to be precisely the same as another piece of paper, particularly for instance for house conveyancing. But nonetheless the number of fax machines was only thirty per cent of what we thought they would have grown to by that stage.
John Winterbottom was a very bright guy, he’d started his research time at Rolls Royce and then moved to Minerva Road working for John Pinkerton. And the one thing about John was like all good research engineers he was forever looking at new ways of doing things and new things to do. He found that the ethos in ICL was not quite to his taste but surprisingly he transferred to IBM in Havant, which I would have thought would have been even less to his taste but he seemed to enjoy it, stayed there for the rest of his working life. v
When we produced our paper around about April or May 1970 we came down very heavily on the use of a packet switched system. The alternative of circuit switching, which virtually meant you had ninety per cent of the circuit dedicated but there was no flexibility if there was circuit overloading. If too many people came on line you could run out of circuits very easily, we felt that was quite a restriction for circuit switching. Whereas (with) packet switching if you were running out of capacity all you did was put in another link and this would double the capacity immediately without any introduction of new software, virtually no new hardware, just pick up a link from somewhere and use it. We modelled it on various scenarios for future use and using something called Markov Chains, which I don’t quite understand but it actually was used and was approved by the Treasury as a method of simulating active work, and it  showed very clearly that packet switched services were incredibly more superior to circuit switched. Less equipment was needed and response times were faster and upgrade times were faster. 
The data speeds in those days were very, very slow. The normal rate that you would get out of a telephone line could be as low as 600 bits per second. If you’d got a dedicated line you might get 1,200 or possibly 2,400 if it was a really carefully routed link. 2,400 bits per second was comparatively slow. It’s perfectly good for teletype, it was okay for the green screen videos that we had at the time, but it was not much use for computers. We were then marketing I think the 2903 which was a small distributive processor. One of the functions we put it in for was where we’d have a group of 2903s in offices all linked to a central computer, but getting a decent line speed for those machines was proving quite a challenge. 
The STC and Plessey submissions produced designs for new switchgear which would enable The Post Office, as it was then, to install circuit switch circuits for high speed packet data links. The supergrid, the high level grid in those days was running at one point five megabits per second. They were upping that reasonably quickly but it was still - in today’s terms - very, very slow, and normally that speed of line was divided into eight voice circuits, I think it’s eight. No it might have been sixteen voice circuits. And that would be what would be used for the circuit switching equipment, which really meant that you were getting no greater speed than was currently available on dedicated lines. Anyway The Post Office then asked for further work to be done with designing and building systems. The ICL team were asked to devise the strategies for installing and running a packet switched service, STC and Plessey were asked to specify the circuit switched equipment. 
For another eighteen months in which time I was actually leading the team, we had one or two more engineers added to give us more information on up to date circuitry. I ought to add an aside here to say that John Winterbottom’s management style was very much let people have their say and then get on with things, my management style was to let people have their say and then have a fight. We had quite a lot of fights in the team with people saying “we’ve got to do it this way” and other people saying “we’ve got to do it that way” and me eventually saying “come on lads you do it one way and you do it the other way and then we’ll see what the result comes out”. And actually this resulted in a very good design study and we ended up with a good design and the elements at a very early stage of what was eventually to become the Internet. 
Donald Davies’ team at NPL had by this stage built their network and this was running and was showing efficiency almost in advance of what was hypothetically possible. There was data getting through their network much more quickly than seemed even remotely feasible with the equipment they were using so we were fairly certain at that stage that packet switched services could be run very efficiently, and I think the current situation on the Internet shows that we were absolutely right. 
The final presentation that we made to The Post Office, that will have been late in ’64, no, sorry, ‘74 was fronted by Peter Hall one of the directors of the company. Now Peter Hall was quite a good forward thinker, he’d actually come from I think ICT. 
 JD: Ferranti? 
 RH: Was he Ferranti?
 JD: Yes.
 RH: Ah. I never quite knew where he’d come from but he was very different from all the other directors who were looking at business plans and how things went. He enjoyed looking at the nitty-gritty and understanding it and he was very valuable in our presentation. He talked a lot about the commitment of computer firms to communication networks in the future and so on and I think made quite an impression on The Post Office people. However, having said that, The Post Office still decided to go ahead with circuit switched equipment. I’m fairly sure that equipment never ever got used because in parallel with it they also developed what they called an experimental packet switched service, that was EPSS, and that was being used by a number of innovative users and to actually show that packet switching was the way to go and it eventually was converted into the packet switched network which British Telecom now support. 
One of the interesting asides to this, Conway Berners-Lee had retired by this stagevi, he’d stayed up in Kidsgrove at the start of English Electric Computers but he’d transferred down to Stevenage when ICL was formed but I think he was finding travelling getting a bit much and he retired with good grace some time in the ‘70s. But his son Tim Berners-Lee of course, is now renowned, Sir Tim as we should call him, because he, on the basis of work done by a guy called Dave Ackerman in our Stevenage office, eventually devised HTML, the interchange standard for the Worldwide Web,  which is now known as the Internet and he was very much in touch with what ICL engineers were doing in telecoms. 
Right now so that was the end of The Post Office study and The Post Office made what I think to be a very silly decision, but that was their decision to make. 
After this, sometime late in 1970, I’d actually got elected to the local district council and when I was with the other candidates who’d been elected having a celebratory drink all of a sudden a guy called John Smith (J C C Smith) turned up. Now he had been in charge of the MPNI/MSS/DHSS Newcastle systems and he’d transferred down to Reading on the Unemployment Benefit system. He turned up at the pub and asked me if I would like to be transferred down to Reading to work with him again. As far as I was concerned that would mean I was closer to the council I was going to be on for the next three years and that sounded like a good idea. So he then phoned up Doug Comish later on who was in charge of government sales and said “we’d like Ray Hennessy to come and work with us”. And Doug came to me and said “I’ve been thinking about the unemployment benefit system Ray, do you think you might like to go and work down at Reading”? So, ever canny, I said “oh that’s an interesting idea Doug. You know, it’s close to where I live, I think I’d quite like that”. So I moved (my working location) down to Reading, which, going back a bit, I had originally moved to when English Electric were trying to get the unemployment benefit job and it actually went to ICT. And ICT had installed equipment and I was living in Reading & working in London, so it was quite an interesting circle to come from back in 1966/7 when the first order was placed to 1972, or 1970 when they were starting to look at more orders being placed. 
I think the 1970s was the most active time I had. From 1970 through to 1979 I went abroad a large number of times although I was nominally working on the DHSS projects and later on the Inland Revenue project. I actually went abroad to a number of countries to front up presentations for government sales in those countries. I went to Hungary, Saudi Arabia, Egypt, Malaysia, Iraq, Kuwait and Greece, Greece was the last one. In fact I didn’t go abroad apart from the Prague conference before 1970 or after 1979 with the company so it was all in that decade.
There were two major projects that year that I was working on, that decade. There was the Unemployment Benefit project at Reading and various adjuncts to that, which included trying to extend it into the DHSS benefit offices, and also Inland Revenue. On the unemployment benefit project there were something like 300,000 unemployed in 1970, there were something like 500,000 by ‘72 and we were planning for 800,000 unemployed. As we know the figure got up to nearer three million, so the eventual equipment that was installed on at Reading and at Livingston in Scotland was six 2966 systems fully committed with vast arrays of tape decks and disc units. That was one of our biggest orders. 
The other system that was, shall we say large, was Inland Revenue. I actually did the sizing for the Inland Revenue project which suggested that they needed 44 2966s. We added I think two or maybe three development machines alongside those so there was an order which we were angling for for 47 2966’s, which would have amounted to the best part of twenty-five million pounds, which was a big order in those days. Because of the work that was going on in DHSS I was then dedicated to DHSS with a view to extending the network into all the local offices so I was taken off Inland Revenue and the project team was quadrupled and we eventually got the order for the 47 2966’s that it required, and as far as I’m aware that was exactly sufficient for the workload that they had because they could forecast it very easily. 
The problem for DHSS was their forecasts were always hypothetical because they were based on government projects of how many people would be unemployed, how many pensioners we’d have, how much child benefit, how much sickness benefit and so on, and those figures were always highly variable. And as I’ve said earlier the 300,000 unemployed in 1970 was up to three million not very much after we started installing equipment. So we put in six 2966s and I believe they were able to cope with something like the three million unemployed, mainly because about two thirds of a million of those were not claiming benefits so didn’t have to go through our system. 
I also did a certain amount of work with other departments and it was fairly ad hoc. Eventually the DHSS projects were centralized on an office in the Fylde coast in Lancashire. The civil servants all transferred up there but I was not interested in transferring up there and it was very much a matter of saying: well what could I do next? And the then manager, Denis Ramsey, decided that I would do better to be in a central role coordinating future requirements. I’m not entirely sure that Denis and I got on because we didn’t see eye to eye on various things and I suspect he was sidelining me but that’s only a personal suspicion. What I do know is at that time we were putting about 50 people into the Fylde office who were supporting various DHSS projects but they would not buy any consultancy from us. Most of the people we put in were not actually sold, the selling was done by Arthur Andersen Consulting Limited and they put in hundreds of consultants. The curious thing was that the man in charge of recruiting consultants was an Arthur Andersen employee and it was no surprise that ninety per cent of the orders went to Andersens. I was a bit (getting very) cynical by this stage. So my central role in the DHSS project gradually withered away until effectively I was doing nothing but clerical work, which I think was Denis’s idea, I think he was trying to get rid of me. Anyway he didn’t get rid of me that easily. I couldn’t really see too much future in the company after the merger with STC with which I had nothing to do at all. And then there was a reverse takeover with Nortel where various chunks of the two companies were exchanged, which again I didn’t have anything to do with but Nortel were to figure quite largely in what I did after I left ICL. 
In 1991 ICL decided to downsize their workforce a bit and they were going to do it from the oldest down. They offered anyone over 55 an enhanced pension package which would increase their pension by forty per cent. Since I was looking to be made redundant by this stage because I had “plenty of other things to do, thank you very much”, I took that pension offer with alacrity. There were very clear signs that we would be financially secure by this stage. I had been 32 years with effectively one company and 27 years of that I’d been working on and off with the Ministry of Pensions, MPNI right through until it became Department of Social Security, DSS. And I met some very, very clever civil servants in that time, most of whom didn’t get any recognition for what they’d been doing, which is the way of the world I suppose. 
 JD: Well, thank you very much Ray, that’s a long and very interesting career. Perhaps now you’d like to have a few words about what happened after you left ICL, the impact on your home life, what you may have done in terms of any professional bodies that you belong to, and carry on until your retirement. 
 RH: Thank you very much. When I retired with 1,400 other people there was a huge exodus from ICL and I think ICL suffered from this for quite a while. I was really out of the loop by this stage so although I kept in touch with various members of ICL who’d come with me right the way from LEO days I didn’t actually feel any need to keep in touch with ICL as a company. By this stage it was in the throes of being sold to Fujitsu and it all seemed to be very much big business as she is writ and not the ethos of finding problems and solving problems that we’d started with at LEO. 
But I wasn’t able to get away from computers that quickly. Very soon after I left ICL I got a telephone call from a guy called Doug Mathie. I don’t know where Doug Mathie started but he used to live in Bristol and had been a project manager at the DHSS unemployment benefit project in Reading and I’d worked quite closely with him for about eighteen months. By this stage he had transferred within STC and then STC had linked up with a company called CCS, I don’t know what CCS stands for. This was an American company which had actually then been hived off and was now totally separate from the STC set up but that was how he got into them and he was heading up a project which was actually producing The Post Office’s or the British Telecom’s preferred Directory Enquiries service. 
Directory Enquiries had been run on the ICL CAFS machine which is really too technical for me to go into as I didn’t really fully understand it but we might add a note on that to the transcriptvii. The system on the CAFS machine which ICL had put forward worked perfectly well and very quickly but it was hardware based and British Telecom decided for reasons best known to themselves they wanted to go to a software based system and CCS had a software based system running in the States. Needless to say it didn’t work for The British Telecom but at least they had the expertise of building such systems. 
Doug wanted someone to write a specification for the terminal for this system and he thought that was the sort of thing I could do as I’d done various terminal specifications for DHSS during my time there. So I went over to Bracknell where he was then based and for three months I wrote the spec for his terminal and that was eventually the terminal which was used on British Telecom’s directory enquiries service. That’s clearly been upgraded by now but it was very successful when they put it in. 
Once that had finished, and that was now about March/April 1992, Doug was transferred into Nortel, Northern Telecom, which had this rather strange relationship with STC and ICL of partial reverse ownerships, which I never understood and never particularly wanted to understand, but as a part of all of this set up Doug had got moved in to Nortel, and Nortel was then split off so he was actually no longer a part of the ICL team at all. What he was doing was he was building a system to enable a company called Energis to have their new network, a dedicated network of transmission lines, to have a calling card system which would authorize use of this network and enable various companies who were using it to pay whatever the line charges were through their calling card. Effectively it was a credit card for use on a telephone network. 
Nortel had three or four different suppliers working for them and the only one I really had anything much to do with was Logica. Logica were doing the central system build and also the overall consultancy for the job, that’s the technical consultancy. Doug was the project manager and he was in my view one of the best ever project managers there’s ever been. His project consisted of two or three punch cards and he would carry those punch cards with him all the time and he would feed them into various computers, how he did all this I don’t know, and then that would produce a project plan. His project plans were legion. If someone came along and said they hadn’t got something done when they said they would, in the same telephone call he’d be able to run his computer and say “right, you’ve got to speed it up or else you’ll be a week late and a week late means you’re responsible for it and it will cost you … …”. It’s remarkable how much people could speed up their project when they were told they might be liable for fifty million pounds worth of damages. 
What I was asked to do was to build the contract document which would specify the overall system. What they had at that time was a technical spec which was built up by the engineers.  And I say “built” because it really had building blocks from all their specification and was a document of about 2,000 pages, mostly diagrams, circuit diagrams, lists of lights, blinking lights, flashing lights and so on and so forth and was very much a techy’s delight. It was no good for proving to anyone that somebody was delivering the right part of the system. 
Two of us were asked to consider how this might be done and the other guy simply edited the document and reduced it to about 500 pages, which anyone could have done and it did nothing for showing what had to be done. What I did was to say “well, I don’t know anything about this technology, I don’t know anything about the contract, what I do know is if you want to have a contract you have to specify what it does not how it does it”. And so I produced a new list of headings and said we would organize it in such and such a way with various bits and pieces defined by what was happening at the line, what was happening at the computer terminal, what was happening on the card, what was happening in the program and so on without trying to tell the developers how they should do that. This was quite interesting because when we had the initial meeting I was asked by their project leader whether I thought I could do the job and I fairly cavalierly said “I’ve no idea, I haven’t seen it yet, you’ve just given me this document and I don’t know what it contains. I don’t know whether I can do it, I think I can but that’s only from experience”. 
And a year later when the project was very well advanced and one of the Logica consultants said to me “I was impressed with that answer, I’d never have said to any potential customer ‘I’ve no idea whether I can do the job or not’ but that was the thing that swung it for you”. So I think honesty generally pays. 
This project lasted about three years part-time. It was very lucrative of course, consultancy rates and I was drawing quite a fat pension by then so it was very pleasant for us to be able to take holidays which we’d not really taken for a number of years, but I’ll talk about that later. 
When the eventual project was finished I had the help of Fred Skeat, one of the ICL retirees who’d originally been in Southern Rhodesia but got out when it transferred to Zimbabwe and joined us all at Reading. Unfortunately for Fred when he was retired he started getting heart trouble and didn’t really live very long after he’d left ICL and the Nortel contract. I think that Fred was one of those people who, when he was given a job to do, he did it. He didn’t particularly mind what sort of job it was, it was usually in systems but he was one of the people who you could rely on to get a task that you presented him with done and done properly. 
As far as I was concerned at this stage I was really running down my computer activities, I didn’t try to get any more contracts after about 1995 when I finished at Nortel. 
I’d got married back in 1960 when I was working in Edmonton – I was working at Edmonton in those days on British Oxygen for the best part of twelve hours a day so I was actually away from home much of the time, too much of the time really. We’d got married and moved to Balham and the train journey from Edmonton to Balham took the best part of an hour and a half so I was three hours a day travelling and about ten to twelve hours a day working so married life got off to a fairly rocky start. We had three children during which we moved out to Reading for the unemployment benefit project and I then spent the ‘70s as I’ve said going abroad very, very frequently and I think this more or less demolished any sense of family life. I didn’t see the children very much as they were growing up and we got divorced in ‘81 finalized in ‘82. 
I started a new relationship in ‘82/83 and we eventually got married in ’96, but in that time we had been abroad quite a few times. As far as travelling abroad was concerned, privately not for business, these were the first real holidays I’d had and my new partner as well. We went to the Far East on the basis of the money I was earning at Nortel, and then we did a round the world trip calling in  Singapore, Australia, New Zealand, San Francisco and back home. And we had been to the States four or five times for shorter trips. So life was good and I think the end result of being in LEO was that I had a very satisfying career and I’m now very settled in the life that I’ve got. [1:32:35]
I had been a member of The British Computer Society from more or less its inception and I got FBCS, Fellowship of The British Computer Society, very early on in about ‘63 when effectively all you had to do was apply for it. I wasn’t very active in BCS activities. I did go to quite a few meetings when I was in London, or some of them were held in Reading, but I wouldn’t say I was an active participant. 
Most of my private activity was local. First of all in the local government from ‘70 to ’74 when local government was reorganized, and from about ‘70 to probably about ‘86 I was an officer in the local Civic Society and those were my main activities. When I’d finished at The Civic Society we did some voluntary work at a concern called The Inside Out Trust which was working with prisoners to try and get them useful, effective, community-based work. Most of the work tended to be a little bit mundane, like short term prisoners would sort stamps for the Blind Dogs for the, sorry Guide Dogs for the Blind not Blind Dogs for The Guides. The Guide Dogs for the Blind Association is close to Reading and we had the young lads on six month prison sentences sorting stamps for them. This meant they (GDBA) could get more money for the stamps. They were repairing bikes which were then supplied to various other charities for use, sometimes sent to Africa for midwives and nurses to use in country districts. We had one branch, not one I was responsible for, which was renovating hand driven sewing machines which were becoming obsolete in Britain but were invaluable in places like India, Pakistan and various African countries which were making clothing for sale back in Britain. So it was quite an interesting set of projects we worked on. One of the most interesting was they had a lot of units around the country who were actually able to produce Braille books so we had specialist teams of long-term prisoners actually typing books into Braille, and this was such a valuable thing that when the charity stopped operating in about 1999 the Braille unit started a completely new charity and that’s still going. So overall I think a very satisfying career and a very satisfying retirement and all in all I really wouldn’t change any of it. 
 JD: Well, thank you very much Ray. Just finally you obviously did all sorts of things and LEO may have been a relatively short part of a 30 year career but was there anything that you acquired from LEO that affected you for the rest of your career or for your life? 
 RH: I think probably the main thing that it reinforced (and I wouldn’t say I’ve always been completely honest because there are always little peccadilloes that one has in one’s life that other people might say “oh don’t like you doing that”) but as far as business went I didn’t want to be in big business. As I said my parents were Communistic Socialists, they were Socialists all their lives and they held steadfast through thick and thin and they worked hard and they were completely honest in everything they did, and what I felt was that business was really a smooth facade to something which was slightly rotten at the core. Now that’s a bit cynical and rather left wing, but the thing I liked about LEO and the LEO ethos was that the guys who worked there were not just clever they were absolutely straight. There were one or two people who (weren’t). There was one guy I know who had a whole roll of receipts from restaurants which he’d picked up when he was wandering round the tables and he used to put those in for expenses, but apart from that sort of fairly minor thing I think the LEO ethos was “we will do it to the best of our ability and be totally straight with our clients and our staff and we will do that both upwards and downwards”, and everyone was dead straight, and I tried to keep that going throughout my career in ICL. It wasn’t too easy because a lot of other people were what I call “business orientated”, they didn’t care how they got business just that they got business. LEO people were not like that at all. Most of the other people weren’t like that, I mean that was just the big business ethos which is, you know, get the business and worry about it later. LEO were not as bad as that, they were very straight. I think all of them were absolutely straight down the line. 
 JD: Well, thank you very much Ray. It sounds like a way of living. 
 RH: Yes. Well, it certainly suited me. 
 JD: Right. This interview with Ray Hennessy has been recorded by the LEO Computer Society and the Society would like to thank Ray very much for his time and reminiscences. The interview and transcript form part of an Oral History project to document the early use of electronic computers in business and other applications, but particularly in business. Any opinions expressed are those of the interviewee, that is Ray, and not the Society. The copyright of this interview in recorded form and in transcript remains the property of the LEO Computer Society 2012. Thank you.
 RH: Thank you.  

i	 I have probably mis-remembered the name of the second person who interviewed me as I am reliably informed that Conway Berners-Lee was a Ferranti at that time.  The second interviewer was the head of computers development whose name escapes me.
ii	 Sam was recruited for LSE by Frank Land who had been a senior manager at LEO until shortly before Sam left.  Frank has pointed out that my understanding of Sam’s later career is somewhat inaccurate as he went to LSE first to do research into the technical side of computer applications, and then moved to Bristol Polytechnic as a lecturer and senior lecturer until Bristol Polytechnic became University of West of England which promoted him to Profesor.  
iii	 Conway Berners-Lee had moved to ICT by now as part of the Ferranti break-up so clearly this is a mistaken “memory”.
iv	 This was Alan Tufnell.  I lost contact with Alan soon after the merger with ICT and the re-organisation of the Government Sales operation.  He moved, reluctantly, into Defence Sales Support.
v	  John Winterbottom actually retired early from IBM and spent some years lecturing on Computers and IT at a college in Portsmouth.  He is recording this in his Personal Written History for the LEO Oral History project.
vi	  Conway Berners-Lee had retired by this stage the other comments about him are incorrect [see above].  However the comments about his son, Tim, are correct.
vii	  CAFS, the Content Addressable FileStore, was a development at ICL Stevenage.  It was a basic fixed disc store with added electronics on each read/record head.  Each disc in the stack had a head for each surface and in the first design each head had a microprocessor controlling it.  The data on the disc was stored with indentifying characteristcs associated with each field within a record.  When the data was to be searched, the mainframe sent a codified query to the CAFS unit which then set every head to look for that data on its surface.  The search argument was able to define the nature of the field(s) it was interested in and the value of the contents, with various functions, (e.g. equal to, greater than, less than for numbers; earlier than, after, between for dates; equal to , containing, starting with, etc for alphanumeric).  The search would take as long as to rotate the disc stack for the number of tracks that the file was stored on, the files being stored vertically to speed up this part of the process.  CAFS would then return to the mainframe the contents of every record that satisfied the search argument, i.e. the ‘hits’.   A program processing a query could present the hits to the enquirer in any form but the main use would be on a screen.  The Directory Enquiries system was able to produce answers to incomplete queries within 1 or 2 seconds, a speed not matched for many years afterwards.

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.