First-Hand:In Her Own Words

Joan Travis (neé Kaye) describes her life as a pioneering programmer

I was born on the 9th of February 1933 into a snow storm. The place was Moor Royd Farm, High Flatts, Denby, near Huddersfield, about 900 ft up on the edge of the Pennines. My father was Stanley Kaye, a tenant farmer of about 30 acres. The farm was mixed, producing milk and eggs and some feed for the farm animals.

I started school at Birdsedge probably after Easter 1938, when I was five. I enjoyed school as I was good at arithmetic, and reasonable with other subjects. In the spring of the school year when we became ten and were due to leave the primary school we took an external exam, called the ‘County Minor’. If we passed we went to Penistone Grammar School in the September, otherwise we went to Scissett Secondary Modern School. I passed and went to the grammar school in 1943. I entered the Lower Sixth Form in 1947 when I was still fourteen. I did Pure Maths, Applied Maths, Physics and Chemistry. I was the only girl in the class. I was timid and shy and somehow had to hide myself in a class of boys. I couldn’t really do that of course, but I kept a low profile.

I took Higher School Certificate in 1949 in the Upper Sixth Form. I didn’t get a County Major scholarship, however. I stayed on a further year in the sixth form to try again, but still didn’t get a County Major. I had applied for University and got a place to do Maths at Birmingham, but without a grant from passing the County Major my parents could not afford for me to go.

My parents were isolated in the country and not worldly wise, Somehow we found out that I could go to the Technical College in Huddersfield and get an external London degree in two years and also get help with the bus fares, so that is what I did. I took Pure Maths, Applied Maths and Physics. I took the general degree, since a special degree would have taken three years and there was my brother and sister to think about as well.

There was one other girl in the first year who was doing Maths before she went on to University. In the second year we took Physics classes with the new first year. There was one girl in that year, Dorothy Fawthrop, and I became friends with her. She was proficient in country dancing and set up a class in the lunch break one day a week. The class was held in the gym. After the class I had some free time and practiced playing badminton with some boy whose name I did not know. I never did get to know his name, or what he was studying, although overall we spent many hours in each others company. We were roughly comparable in skill at playing badminton.

The degree examination was the University of London B.Sc General Examination: 1952 (for external students) and was taken by such places as Technical Colleges, University Colleges and the Military College of Science Shrivenham. Even for the general degree there was the grading of First Class Honours, Second Class Honours and Pass. I got First Class Honours – the only one in my year at the technical college. I was the only female in the thirty-two candidates with First Class Honours.

I did not go to London to receive the degree – it would have cost too much money.

I was still only nineteen and a half and had no idea what to do with my life. I can’t remember what jobs I applied for, except as something like a Scientific Officer in the Ministry of Defence and a job at British Thompson Houston in Rugby. I was probably young for my age and didn’t interview well. In the winter I got a few weeks’ work in Huddersfield going round giving out free packets of a washing powder called Daz. The spiel we had to say started something like – “Good morning (or afternoon), I’m not selling anything. Here is a free trial packet of Daz”. Before then I think washing powders had been white; this Daz was blue. During the time before I got a proper job, I took over the task of wiping the eggs before they were sent off to the egg packing station; I also scalded out the large milk churns which were used for the milk which was collected by the Milk Marketing Board. I expect I also helped with other things on the farm.

Eventually I got a job with Ferranti in Moston, north Manchester, as a programmer. Through Ferranti I got accommodation in a YWCA hostel in Altrincham, to the south of the city. It was adapted from two large older style semi-detached, three storey houses and most residents shared a room with one other girl. The hostel wasn’t particularly near work so I had to get a train from Altrincham to Oxford Street in Manchester then a bus to the works in Moston.

I started on the first of April 1953. Ferranti, in collaboration with Manchester University, built what we (Ferranti) called Mark 1. Mark 1 was what was described as “The first commercially available stored programme electronic digital computer in the world”. In Great Britain at the time there was EDSAC at Cambridge University, ACE at the National Physical Laboratory and the Ferranti Mark 1 at Manchester University. The storage for Mark 1 was on cathode ray tubes for immediate access and a magnetic drum for backing store. Input was by five-hole paper tape and output was also by paper tape. Alan Turing’s Input System was stored permanently on the drum so that programs could be input from paper tape. Editing of programs on paper tape was done using yellow sellotape, scissors and a hole-punch. The hole-punch allowed one character at a time to have any one of the five holes for the character or the sprocket hole (which was smaller) punched using a ‘bodger’. The hole-punch was beautifully made of brass and Perspex. I still have a hole-punch and several bodgers for the character holes, but no bodger for the sprocket hole (this would probably have existed but would have been fine and not really used as the sprocket hole was already there on the paper tape once it had been punched). The tape reader could take a short amount of tape without sprocket holes if paper tape was joined in, but tape was usually joined in without covering the sprocket holes.

Mark 1 belonged to Manchester University and was initially situated in its own room (in a specially-constructed building) in an old part of the University. The University used the computer during the day time. Other companies, including Ferranti, two separate sections of Metro-Vic and the Shirley Institute (associated with the cotton industry) used the computer during the evening and overnight. Ferranti ran a service whereby they assisted other institutions to use the computer in the evenings – these included London Transport and the Atomic Energy Authority at Culham I think. Ferranti also ran a – perhaps the first – Computer Bureau and used the computer overnight for this purpose. Work was done on such things as calculations for the aviation industry, Ordnance Survey, whirling speeds of rotors, radome calculations and the optimum cutting of Perspex sheets.

Normally when a new programmer joined Ferranti they would be sent on a two week course to Cambridge University to learn about programming on the EDSAC. I didn’t get on this course till September 1953 when I had been programming for about 6 months. On the course at the same time as me from Ferranti was Chris Wilson who was later to become Managing Director of ICL for a period. I think he spent a fair amount of time trying to interest people from other companies in Ferranti Computers as he was in marketing. Little did I know that my children would eventually be graduates of Cambridge University.

Alan Turing was still alive when I started at Ferranti. I never knowingly saw him, but he was around the building at night so I might have seen him when using the computer overnight. (Turing died in June 1954).

The Ferranti factory at Moston manufactured electric fires, radios and such like. The computer design engineers were housed somewhere in the main factory, the programmers and sales staff were in the ‘tin-hut’ (a prefabricated building) at the end of the factory. A year or so after I started work Ferranti set up a London office for the marketing people and also programmers.

In early 1953, when I began, the programmers consisted of one male in charge of the programmers (he also programmed), two males sharing an office, then in one big room seven females – then me – and four males. Not that much later, two other females joined. A technical writer also shared our big office. I suspect the proportion of females in programming has never been as high since – except of course in a company which was somewhat later set up initially to specifically employ women (F International, founded by Dame Steve (Stephanie) Shirley in 1962). In fact I should imagine that in this, the 21st Century – and most of the time since the early 1950’s – a general group of programmers with more than half the group female would be so rare as to probably not exist at all. Part of the explanation for the dearth of male programmers at the time could have been that the Second World War had ended not all that long before, although long enough I should have thought. Mind you the total number of programmers in the whole country wasn’t very many at that time.

I worked for one of the men (Cyril Gradwell) in the shared office. The other was Dr Prinz, who we called DP. He was German and had been interned for a time at the beginning of the war in the Isle of Man. He had in his youth attended lectures by Einstein. One thing I learned from DP, and one thing I still always try to do, is to date even scraps of paper I write on. Later if sheets perhaps get jumbled and my ideas about something have changed I can tell what is current. Amongst the programmers already in the tin-hut when I began was Mary Lee Woods who was later to become the mother of (Sir) Tim Berners Lee. Conway Berners Lee joined the Ferranti Marketing group in London and eventually Mary Lee, along with some other programmers from the tin-hut, moved down to London and Mary Lee and Conway later got married.

The early computers always had resident maintenance engineers to get the computer up and running and to be on stand-by to deal with any fault when the computer broke down. The contents of the cathode ray stores could be selected and displayed as dots on a screen. When a bit appeared where it shouldn’t have been it was called a ‘clod dropping in’ and if a bit disappeared when it shouldn’t have it was called a ‘clod dropping out’. Once I was using the Mark 1 late one Friday afternoon, it turned out to be a bit of a hopeless time to be using it as when the Metro-Vic factory at Trafford Park closed down for the weekend it adversely affected the power supply to the computer.

In those days industrial companies didn’t seem to quite know how to treat female graduates. I believe that Sir Vincent de Ferranti (then in charge of Ferranti) said that no woman employee should be paid more than his secretary. Once when I went to an aircraft company near Hull (the Blackburn Aircraft Ltd, Brough, who installed a Pegasus in 1958) to run acceptance tests on a computer they had bought from Ferranti I had to be taken to a different canteen to the (male) Ferranti engineers for lunch.

Mark 1* was a re-design of Mark 1 and had a division instruction (?) amongst other improvements. About nine were sold, both here and abroad. I didn’t write the engineers’ test program for the main computer but I was involved with what were then more ‘exotic’ peripherals such as magnetic tape and line printers. I acted as a sort of engineers-mate (like a plumbers-mate) for the design engineers when they were attaching the peripheral to the computer, running the computer and putting in data and little programs by hand to test some aspect of the functioning of the equipment. One weekend we were working at the Atomic Energy Authority in Aldermaston; the resident Ferranti maintenance engineer enjoyed archery and during some spare time we were practicing shooting arrows, in a long corridor I think. The juxtaposition of ancient and modern weapons has stuck in my mind. I worked at Royal Dutch Shell Laboratories in Amsterdam for some weeks, helping to trying to attach magnetic tape to their Mark 1*. We were working one weekend and the Dutch engineers decided to give us a treat. The normal tea which was served from the tea-trolley was herbal, very weak and without milk. The engineers tried to make us what they thought was real ‘English’ tea – this involved them in boiling up tea leaves in some container for a long time. Just before one of the trips back from Amsterdam the waiters at the hotel we were staying in told us at breakfast about the crash (in Munich, February 1958) of the plane carrying Manchester United football club. Our own plane back was delayed whilst we were on it to sort out some problem. It was good to land safely.

After Mark 1* Ferranti had two lines of computers going. Mercury was designed and manufactured in the north – by this time the Ferranti Computer Department in the north had moved to Thomas Street, West Gorton, (Manchester) into what had been Brookes and Doxey’s textile machine manufacturing factory. The factory was situated in an area of back-to-back houses. I believe it was thought locally that a furniture shop had opened up, but it was actually the reception area to the offices.

Pegasus was designed in the south (London) and manufactured in the north (West Gorton). I wrote the engineering test programs for Mercury and most of the peripheral tests for Pegasus. Mercury was a floating point machine, more for scientific use; it had a core store. Pegasus was a smaller capacity computer with nickel delay line storage, more for commercial (?) use. When the first Mercury was being built and was still in the design engineers’ hands I would sometimes get a phone call which consisted of one word – help. I would know to trot down from the office to the factory area to write or run some little program to try and help them sort out some problem they were having difficulty with. Sometimes, at lunch-time the engineers would say they were “off to the church” – not what it seemed, but a local pub.

In those years, I was really more interested in a computer in its early stages when it was not working than when it was. There was great satisfaction however when a faulty computer decided to give in and work – at least it did feel like a battle of wills sometimes.

Before a computer left the factory it had to run ‘Acceptance Tests’. These consisted of running test programs on the central computer (CPU) and the peripherals for a certain length of time, then running some general operational tests (GOTs). Once on the customer’s site and up and running again the Acceptance Tests had to be repeated. I ran quite a few Pegasus Acceptance Tests and the odd Mercury Acceptance Test when the lady who usually ran them couldn’t. One Mercury computer went to the Belgium Atomic Energy Authority in Mol (in September 1959), which was outside Brussels. I just missed out on a helicopter flight from Brussels to Mol as they had stopped flying for the winter; instead I was subjected to a hair-raising drive by army personnel.

After Pegasus the design engineers in the south built Perseus in Lily Hill. Lily Hill was a big old house in Bracknell, Berkshire, to where the Ferranti design engineers who used to work in London had moved. It had a lovely garden with lawns and big old trees and an area of many specimens of rhododendrons. Perseus was built in the library. It was a multi-radix machine for commercial use. I wrote the engineers’ tests for the CPU. I was seconded from West Gorton to Lily Hill to help get the computer working by writing and running test programs as required. I also wielded the odd Oscilloscope probe when I was there and another engineer was not available to assist the engineer doing the testing. Someone else wrote the General Operational Tests, but the chief engineer and I had to get them working as the programmer didn’t have a clue as to what was going on. Once Perseus seemed to be obeying all its instructions correctly and its peripherals were working it wasn’t very reliable. Eventually we had to run the Acceptance Tests at Lily Hill. There was much rejoicing when they passed. Soon after that there was a big dinner given at Hurley (Berkshire?). I got a verbal invitation and it wasn’t until after the dinner I found out that initially I had not been invited as it was thought that a female would inhibit the conviviality of some of the senior men. Some of the design engineers said that they would not go if I was not invited, so I duly was. I did not think it was strange that I was the only female amongst many males – I had started getting used to that at school – and apart from when I had first started programming – I had been used to that ever since. In any case I had put in a great deal of time and hard work to help get the computer up and running, being there most evenings. At one stage I had had to stop working after midnight as the chap in charge of administration of the building did not like the idea of a female working with males in the early hours as “the ‘facilities’ were not suitable”. Years later when I was having driving lessons in Crowthorne I discovered that this chap had been having an affair with his secretary and had tarred everyone else with the same brush. The Perseus was shipped to Sweden and eventually I went over to Stockholm in the final stages of commissioning (spring 1959). We were working one weekend and none of the Swedish engineers were around. Suddenly firemen in yellow helmets burst into the computer room. They looked around then pointed at Dudley Berry’s pipe. There were smoke detectors and the alarms were connected to the fire station. Dudley had to stop smoking his pipe in the computer room after that. The incident was duly reported in the computer log by me.

One was not supposed to stay in the YWCA at Altrincham for more than 5 years, but I did stay a bit longer. One of the girls who had stayed in the YWCA left to share a flat with three other girls, and then her sister joined her. When one of the other girls left the flat I moved in. The flat was in Brighton Grove, Rushholme in Manchester. It was the first floor of the right-hand half of a semi-detached house, with the landlord in the ground-floor flat and a couple in the flat upstairs from us and a family in the left-hand house. There was also a semi-basement cellar where the coal was kept and where I kept my bicycle. It would have been an elegant house in its day. The floor of the sitting-come-dining room sloped very much towards the centre of the room. There was a very large ornate sideboard – we called it the aircraft carrier – which stood right against one of the inner walls of the room. Just up the road from us was a large, for then, luxurious block of flats where we understood Sir John Barbirolli lived when he was in Manchester. He conducted the Halle Orchestra in the Free Trade Hall in Manchester. Unless it was pouring with rain I cycled to work from Brighton Grove to West Gorton, passing what had been Belle Vue Zoological Gardens and was then an Amusement Park.

During the time I was working at Lily Hill on Perseus I started going out with John Travis who was involved with the magnetic tape decks. He had first invited me to ‘go to the dogs’ (greyhound racing) when he was up in Manchester working on magnetic tape for a Pegasus computer at ICI Blakeley. I did not accept his invitation then. After I had finished my secondment to Lily Hill I was back again working in West Gorton and John was working in Lily Hill. John would sometimes hitch-hike up north to see me and we would correspond by letter. Eventually in early July 1960 we got fed-up with being in different places. John had already booked leave for the middle of August so we decided to get married then. I made my own wedding dress with white satin that I got from the sales in Manchester. We were married in Denby Church on August 13th. The reception was in Denby Primary School. Big adults had to sit on little chairs. In the evening we went to collect the caravan which we were to tow on honeymoon. We got up to Scotland and parked on the edge of the road (you were not restricted as to where you could park in those days); next morning we found we were almost hanging over a loch. As well as the countryside we visited Edinburgh, Glasgow and Inverness – all in the space of a week. John already rented a council flat in Bracknell so I moved there to live. I left Ferranti at West Gorton and moved to work in Newman Street, London, where the London based Ferranti programmers were now housed (after moving out of Portland Place which was near Broadcasting House).

(John Travis was born in Brentford in 1931. After leaving school in 1947 or 1948 he joined the Borehamwood Laboratory of Elliott Brothers, working firstly on digital fire-control for naval gunnery and then on the Elliott 401 computer. He did his National Service in the RAF, training as a wireless operator. John left the RAF in 1956 to join Ferranti Ltd in London as a computer commissioning and maintenance engineer. He stayed with the company and its successors (ICT, ICL, Fujitsu) until he was made redundant in 1989. During his time with Ferranti, one of John’s specialities was magnetic tape systems).

John and I eventually started to look for somewhere to buy, and in 1961 we moved into a bungalow called Merrywood on Sandhurst Road in Crowthorne. Some time in that year I was asked if I would like to work on a military defence project at Ferranti, Lily Hill. One of their programmers had unexpectedly left. There was no overlap period, so I had to find my way round what she had already written. The project was called ‘Green Ginger’ (a cold war mobile air defence initiative) and the special purpose computer, Cerberus, on which the project was based was in the process of being designed and built. It was not long before I found out that I was pregnant, I told the project leader early on so that they could get someone-else to eventually take over. Before I left I did see the computer start to work. For some reason which I found very strange they had not got anyone to write test programs to test the functioning of the computer – at least I knew of no one – but I did manage to prove to the engineers one problem they had which was ‘cross-talk’ whereby writing to an area of store corrupted another area of store. I left in March 1962 not knowing if I would ever be able to work as a programmer again. The idea of part-time working in such jobs was not really a concept then.

Our first son, Kim, was born in April 1962. Whilst Kim was a baby I did some technical writing at home for a Ferranti navy project, I wrote the programmers manual for the ‘conventional’ part of the Poseidon computer. Our second son, Christopher, was born seventeen and a half months after Kim.

In autumn 1966 when Kim was about four and a half we found out that the local primary school in Crowthorne would not be able to take him until September 1967, by which time he would have been nearly five and a half. He was well ready for school and so we decided to send him for the mornings to a private nursery school in Bracknell. John would be able to take him in the morning on his way to work and collect him at lunch-time. Christopher was just three so we didn’t intend to send him to nursery school at that time. However when John took Kim to see if he could enroll him he also had Christopher with him and the lady who ran the nursery said why didn’t we try Christopher as well, so we did.

This started quite a hectic period in our lives. With some clear time on my hands I decided to get John to ask at work if he could borrow some programming manuals for the 1900 which was then the computer being manufactured by ICT. John got some manuals and I tried to familiarize myself with the 1900. In January 1967, I think, out of the blue, I was asked if I would like to apply for a job to work part-time on 1900 programming. As far as I was concerned I could recognize the 1900 as if it were a Pegasus gone to Canada and come back again. Of course it had advanced considerably, but it was not as steep a learning curve as it might have been to start programming again after nearly 5 years. I started work at the ICT Offices in the centre of Bracknell. In March 1967 the nursery which the children attended had to close. A lady at the nursery took them on for a time in her own home. A month or so later ICT announced that they were closing Lily Hill and they expected people either to transfer to West Gorton, Manchester, or Stevenage in Hertfordshire or leave. No one wanted to go up north, many left and a few opted for Stevenage. I had only been in my new job for a few months. John had a job with a small group who opted to go to Stevenage. Fortunately I was offered a programming job with the same group. My job was to program for John Iliffe’s ‘Basic Language Machine’ at Stevenage. (This project ran from 1963 to 1868. It introduced techniques such as descriptors, memory management and memory segmentation, all aimed both at the efficient protection of concurrently-executing programs and the reliable implementation of high-level languages. By 1969 ICL had decided not to support the BLM project).

Postscript by Simon Lavington

Joan Travis wrote an account of her life for her family in about 2005. The above text is an edited version produced in December 2022 by her son Kim for me, in connection with the task of cataloguing all Joan’s computing documents and artefacts. Joan’s original text does not cover her programming activities from 1967 to her retirement in 1997 so these years are summarised below, based on evidence revealed in her collection of computing documents.

Finally, an assessment is given of Joan’s extroardinary life as a pioneer female programmer in a male-dominated world. The assessment draws on information provided by Joan in a series of e-mails to me in the autumn of 2015, in connection with researching the book: Early computing in Britain: Ferranti Ltd. and government funding, 1948 – 1958.

Joan’s programming activities subsequent to 1967

During the period 1969 to 1985, Joan and John Travis continued to be involved in various Ferranti/ICT/ICL projects, mostly within the Research & Advanced Development Organisation. Amongst projects on which Joan worked were the Variable Computer system, VCSL, and the Content-Addressable File Store (CAFS).

From 1984 – 1988 Joan used MUSL (Manchester University System programming Language) running under MUSS on a 16-bit micro for two separate projects: (a) development of tree file structures for speech annotation of documents; (b) work on a Human Interface Monitoring System for analysis of information flow across a communications interface. From 1988 – 89 Joan worked on enhancing the code of an ICL Logic Programming Language (PLL) under an Alvey project. By 1985 John and Joan had moved back to Bracknell. Joan was by then a Senior Programmer in ICL’s Research Department. In 1989 both John and Joan were made redundant by Fujitsu, the successor company to ICL.

From February 1991 to February 1995 Joan worked as an Analyst/Programmer for Compass Peripheral Systems Ltd in Newbury, Berkshire. During this time, John acted as a House-husband. Joan retired in 1997. John died in October 2022. At the time of writing (March 2023) Joan is a frail 90-year-old, living in a nursing home after an amazing life in computing.

Diversity assessment

Ferranti’s Mark I computer programming team at Moston had a unique size and mix, compared with other UK computer manufacturers in the early 1950s. At the start of 1951 the Ferranti’s team numbered four, of whom one (Audrey Bates) was female. A year later there were seven programmers, of whom four were female. By April 1953 there were 14 programmers, of whom eight (including Joan) were female. By the end of that year Audrey Bates, who had been Alan Turing’s research student at Manchester University, had left to join the Ferranti Mark I computer delivered to Toronto University There she worked with Beatrix Worsley, whose Ph.D. thesis had been partly supervised by Alan Turing. Worsley has rightly been called “a pioneering computer scientist and the first female in Canada to make significant contributions to the field”

After 1954 the Ferranti programming group split, some remaining in Manchester whilst others went south to build Ferranti’s software presence in London. By then, Ferranti had agreed to pay female programmers the same salary as male programmers – a very enlightened decision for a large manufacturer at the time.

How do the Ferranti statistics compare with other UK computer manufacturers of the time? By 1952 the Lyons company, developing their in-house computer LEO, had six programmers, one of whom was female. By 1953 the team numbered nine, of whom one was female. By the end of 1955 the Lyons team had grown to 14 programmers, but still only one of these was female.

Elliott Brothers Ltd of Borehamwood was another early entrant with their Nicholas and 401 computers. In 1952 the Elliott team included four active programmers (one male and three female). By the end of 1954 the three females had left to start families and five new males and one new female had been recruited. This remained the story for the next two or three years. The star of the team was Dina St Johnston (neé Vaughan), who left Elliotts in 1958 to start Vaughan Programming Services (VPS), the UK’s first ‘software house’.

The English Electric Company, building the DEUCE computer at Kidsgrove, had only six programmers when the software team was set up in 1954. None of these was female. The final contemporary British manufacturer in the early 1950s was, BTM (later becoming ICT); they had a small team, exact size unknown, in 1954. It is believed that BTM had no female programmers at that stage.

In summary

During her active life with Ferranti Ltd and its successors, Joan had used the first stored-program computer to appear on the market anywhere. She retired forty years later, just as early-release mobile ‘phones and laptops were waiting in the wings. In her chosen role of, as she put it, Engineers’ Mate, she had got to grips with computer hardware at its lowest level and written low-level test programs and applications software for a variety of interesting machines. She was a unique and highly-respected pioneer in a largely male-dominated world, often the only female in a group. This was certainly the case when she made four visits to Shell’s Amsterdam Laboratories in the period 1955 to 1958, mostly each of about a week but one of 38 days’ duration. The purpose was to attempt to connect a novel 35mm magnetic tape deck and then a half-inch magnetic tape deck and a Bull lineprinter to Shell’s Ferranti Mark I* computer. In a similar exercise, Joan made one visit to Rome in 1956 of 17 days to connect new equipment to the Instituto Nazionale per le Applicazioni Calcolo (INAC)‘s Ferranti Mark I* computer. Joan was indeed a valued and trusted colleague in a male-dominated world.

Simon Lavington, 24/2/2023, lavis@essex.ac.uk

Photographs

• 

  Part of Ferranti’s Moston factory, north Manchester. The Tin Hut is circled in red. Permission from Fujitsu Services Limited, with captioning help from the Ferranti Archive at the Science & Industry Museum, Manchester and the School of Computer Science, University of Manchester.

• 

  The Ferranti Mark I computer In 1951. John Bennett, who had a desk in the Tin Hut, is on the right. The lady seated at the console is believed to be Joan Hart, a secretary in Manchester University’s Computing Machine Laboratory. Permission from Fujitsu Services Limited, with captioning help from the Ferranti Archive at the Science & Industry Museum, Manchester and the School of Computer Science, University of Manchester.

• 

  Joan Kaye working at a desk in Ferranti’s West Gorton factory, Manchester, in about 1960. Permission from Joan Travis.

• 

  Joan at a West Gorton reunion in about 1975. She is in the front row, the only female in a group of 65 former Ferranti computer engineers and programmers. Permission from Joan Travis.

• 

  A Tin Hut reunion in 1993. Joan is in a dark dress, standing in the centre of the front row. Permission from Joan Travis, with captioning help from Olaf Chedzoy.