David Hartley

David Hartley interviewed by Alan Macfarlane 2nd May 2017


Introduction

0:05:11 Born in Halifax, Yorkshire, in 1937; my paternal grandfather was a businessman in Yorkshire and made clothes, so did my father, whereas my other grandfather was a chartered accountant, and I wished to follow neither; one died in the late forties, the other in the late fifties, so I knew them both; their widows I knew better because they lasted longer; on the whole they were nice; I used to think of one as being the nice one and the other not so nice, but I wouldn't dare suggest which was which now; my father wanted to be an architect and his father wouldn't let him; his father had this family business making working clothes and he just wasn't allowed to go to university; he had to go and work in the business; after that, my grandfather tackled me on the subject, asking when I would be finishing school; I was actually at university at the time but he didn't really know what a university was; fortunately by then my father had intervened and said that I must do what I want; he couldn't do what he wanted but I must, so I came to Cambridge, with no intention of staying here but I haven't left yet; my mother was an actress; she went to one of the acting schools in London and got a medal, and I knew her as an actress quite closely because she used to act a lot in local theatre in Yorkshire; I used to go and help and sometimes performed with her; I was very fond of my mother; she died quite young, she was barely sixty; I think what came out of that, this love for the theatre and so on, that I ended up having three children, all of whom were somehow involved with media and the like; I loved my father very much too; he loved to make amateur films and I used to help him do that, but they both died quite young in the late seventies; I was one of the first people to put up videos of my lectures which certainly had a connection with my father; playing with video and video cameras, and later electronic ones, was great fun for me

Early Life

4:22:11 I find it difficult to say what my earliest memory was; I do remember having to go into hospital for an operation although I don't know what it was for, but that is a very faint memory; I don't know how old I was except I was under twelve; I can't remember anything of my early life; I can remember my first primary school at Hebden Bridge, and even remember my first primary teacher because she was a great personality; but that connection is all gone, it is a memory, and I have no connection with those people at all when I left Hebden Bridge and came here; after primary school I was sent off to boarding school when I was about twelve; I was sent to a prep school firstly at Conway, North Wales, called Rydal School; my father sent me there because he had been quite an important Rugby Union player; in those days there was no difference between amateur and professional and he played for Halifax; I never saw him play because that was before the war, but this was the thing he loved most, so he sent me to Rydal because that was a rugby school; I turned out to be absolutely awful at rugby, I was a complete disappointment to him, and bless him, he didn't hold it against me but accepted it; I eventually went to the main Rydal School in Colwyn Bay, and I got involved from then in technical things; I dismantled the school's telescope - a big telescope in a dome on the top of a building; I remember with a friend dismantling and cleaning it, and putting it back; it was the last thing you should ever do to a telescope but it survived; I would have been in the fifth form then, so fourteen-fifteen; there was unhappiness in both schools for me; bullying was a big problem in those days and it was for me; I don't know whether it was because I was non-sporting, it could well have been, and I had a miserable time at the main school; I even wrote home to my father saying that I was being bullied; I didn't really realize what I was doing when writing that letter, because my father came storming to the school to see the masters, and of course that can do more harm than good; but I got through it, and fortunately my last two years at school, in the second and third year sixth, were actually quite happy relatively speaking, but I always felt rather an under-dog at school; I was quite good at piano-playing, I had a lovely teacher, and I won a school competition; I was quite keen on theatre of course, usually behind the scenes as a stage-hand; I enjoyed that side of it but did not enjoy sport; I remember quite a few teachers, particularly my mathematics master who was a wonderful man by the name of Ray Blomfield; he was a bit separate from the rest of the staff as he just taught maths and nothing else; he picked me up, as it were, when I was in the fifth form, having noticed that I had some talent, and persuaded me to do it in the sixth form; he also said that I ought to go to Cambridge; he had not been there himself but wanted to get me there, for which I am very grateful to him; we got on very well; he wasn't so much a teacher as a friend actually, and he taught me some very nice maths, but when I came up here, of course, it was very different; Percy Hayward was the music teacher; he had been at the Lees here some years before, and he had written some light-hearted operettas which he got the school to perform which was great fun; we were a boys' school but there was a girls' school down the road, so when we did anything musical we did it with the girls' school; I shall never forget, halfway through a rehearsal, Percy yelled out, "Sopranos you're flat!", and this was of course a great joke; I enjoyed that, not that I sang very well

Mathematics and Cambridge

11:10:08 I started as a mathematician before university because I was persuaded by Mr Blomfield to read nothing but mathematics in the sixth form; you normally did maths, physics and chemistry in those days, and it was said that I was so good at maths that I didn't need to do physics and chemistry, which was a great mistake with hindsight; I got through the sixth form doing just maths and further maths, where it was said that just doing those two you could get to Cambridge; the school academically wasn't that good, and though my maths master knew that what I was doing was reasonably exceptional, the school didn't understand; so I did start maths early, but the problem was that when I came to Cambridge all the maths was different; for me it was too difficult; the curriculum in those days, in the early '50s, was quite different, like solving problems in geometry which you did by geometry where you ought to have been doing it by analysis; I used to love taking a picture of the lines and drawings and projecting a line to infinity; once you did that all the problems came out nicely, and you weren't allowed to do that here; so it changed very much; I did get here; I took the scholarship exam in Clare, I only did maths and further maths in the exam but you do the general paper as well and the translation paper; eventually I got a postcard from Henry Thirkill who was the Master - Henry Thirkill used to do all the admissions in those days; he wrote "Well done. You nearly got a minor scholarship" but I got an Exhibition, £40 rather than £60 or £100 which were the scales; he said that I had done very well in the maths papers "but your general papers gave you no support"; I wasn't good at writing essays, and certainly wasn't good at Latin translation, but I was forgiven that; of course I was doing Latin at school as you had to in those days to get into Oxford and Cambridge; so in I was, I was here; those exams were held at Christmas time so you had the rest of the school year to decide whether you wanted to come or not; I left after the Michaelmas term and went to work for my father; although he had said I was forgiven from coming into the family business I felt it my duty to try; so I worked for him for nine months; the company was very small- him, his brother, me, and my grandfather who was retired; during the summer my father went off on his usual trip round Scandinavia in a boat, and left me in charge of the firm; I was only eighteen, but I reckon I did that pretty well, except that one day my grandfather turned up to see what I was doing, and that was not a happy occasion for me; but I managed; another story about my grandfather was after I had come to Cambridge, and had gone in the vacation home to Yorkshire and went to the factory to help my father; I met my grandfather in the factory building one day; he told me that he had left school at twelve, and my father at fourteen, and when was I going to leave school; I was over twenty-one by then and of course is wasn't school, but he didn't understand that; he died in the late fifties and my father was marvellous; he said he would let me go to Cambridge as long as I got a scholarship; I did, so that was fine; I had said that when I had done my three years I would come home and we would then talk about what I was going to do, one of the options being the family firm; well, I never came home, that was the trouble, I stayed on, and he was quite resigned to it and was very supportive of me, so I have a happy memory of my father; he didn't have such a happy time himself because he lived in Yorkshire with this family firm which was getting boring; I started to feel slightly guilty about that; he died in his sixties - too much smoking; fortunately I didn't smoke in those days otherwise I wouldn't be here

Religion and music

18:19:00 On religion, my parents were Baptists, but not particularly religious; the grandparents were but my parents gave lip-service to it, and I give even less service to it; but when I came to Cambridge, to a Church of England College, I fell in love with my wife-to-be who was quite keen on the Church; so I followed the Church then to be with her, indeed I was confirmed, but it didn't last long; now I am pretty agnostic, I don't think about it and it doesn't play a part in my life; I sometimes wish it would because I go and listen to some very fine music; my interest in music continues but rather skirting the surface; I gave up playing the piano; with my wife we used to go to concerts a lot in London, to the Royal Festival Hall; but my wife died some time ago and I don't think I have any genuine adherence to any church except when it's convenient to do so, and music the same; I had talent at school but after that was too busy trying to be a mathematician; I had no hobbies which involved a large intellectual activity in my life; when I got into computing, of course, that completely went because computing absorbed everything, and very happy and pleasant it was too, but that was not until I had graduated

Cambridge Universithy life

21:28:21 I did maths for two years and then I did the diploma in computing in the third year; the diploma was a fourth year course, a one-year post-graduate course; in those days there were quite a few dropouts like me; when I arrived here I met my Director of Studies, F.C. Powell in Caius, a formidable person but I am grateful to him for what he did; he said that I was obviously clever as I had got an Exhibition, but that was all they knew about me; he said I had an option to start on second year work and do the second and third years in my first and second years, you effectively miss out Part I, so you can get on ahead and in the third year do Part III maths; it sounded a good idea and I worked my boots off for the first two years and did Part II maths; I struggled doing that as it was just too difficult for me and after two years decided to do no more maths, it was a waste of my time; I got a second, you didn't get 2:1 and 2:2s in maths which is a blessing, and I'm sure I would have been a 2:2; so I got through two years, got my second in maths, and I think that was a great achievement for me because I was stretched to my limit; I had some very good supervisors, one particular one I will never forget, that was Shaun Wylie who had come back from Bletchley Park and had also worked at Cheltenham, but he came back and retired here; a wonderful teacher; then I had to find something to do for my third year, that was the rule; I couldn't get my degree until I had done three years in something; when I look back I had only done maths in the sixth form at school, no physics, no chemistry, so was suddenly rather short of something to do; this is where my tribute comes in for the college system; you go and see your Tutor or even the Senior Tutor and present him with the problem of what to do next year, and that person will not be in my subject by definition; so I am getting advice from a good academic but not a mathematician and that seems the best way to get advice on should I do maths because he wouldn't be biased; we hunted around; I could perhaps have done engineering but I wasn't very keen on that, although I count myself now as an engineer;

Postgraduate computing

but in the end we discovered this diploma course in numerical analysis and automatic computing; they had just changed the rules to do this and add it to your degree, and I was one of the first people to do so; so I chose to do this course and I absolutely loved it; there were about five of us taking it in our third year, and about a dozen altogether with third years doing a fourth year; the University authorities had a rule that if you did this and you got your diploma in your third year you couldn't have a certificate; after that year a certificate arrived and I didn't send it back; I loved the diploma course and we had some people who have become quite well-known since; it was just after the EDSAC had been demolished which was in 1958, and we got a new machine EDSAC 2; so I started off on this brand new machine and had a marvellous time and I've enjoyed computing or programming ever since; our teachers were Swinnerton-Dyer, Jeff Miller who was really a mathematician - there was a lot of maths in the course in those days despite my feelings about mathematics - Maurice Wilke's of course, Roger Needham was still a student, about a year or two older than me; that started off a marvellous friendship with Roger because he came into the lab one year while I was already there, and we started working together on the EDSAC 2 computer; he was totally different from me in every way; he was really an intellectual while I was totally non-intellectual, but we got on extremely well, and did so until he died, which was a tragedy; we did research together throughout the sixties and then I went off at another tangent while he stayed in the maths lab, and we did a lot of teaching together and even researched together; I was very fond of Roger, a lovely man; Maurice Wilkes was of course the "old man", as he was called; I thought he was a marvellous manager; he was a bit formidable and we were all slightly frightened of him, but actually you look back after not much time and realize that he was a brilliant man; I was doing my research in the lab and got a PhD, then I went to see Maurice and said I rather liked being here and did he have any jobs; he said there were no vacancies unless somebody dies; then about a year later he came along to me and said he had a job for me, working on the Atlas 2, at a salary of £1,000 a year, so I was essentially a member of the staff but not formally; his problem was money, particularly money for staff; he could attract a lot of money for building things but the staffing was always very difficult in any place; I heard a story some time after when he had appointed me, I think I was told it by Roger Needham, that how he had managed to get money to pay me was as follows; we were ordering the bits and pieces for the Titan computer and the machine was to have seven tape drives; he had enough money to buy the seven, or six tape drives and to hire me; it is a true story and something I'm rather proud of; I liked Maurice, he was good to me, and he was my Supervisor for my PhD, although he shouldn't have been as he was Head of Department; he and David Wheeler examined me which was also irregular, but they had asked Leslie Fox who was the head of the Oxford lab if he would be my second examiner, and he refused, firstly because he was a numerical analyst not really a computer programmer like I was, and he didn't really hold with computing in that sense; he refused as a protest about how he felt which I was rather glad about actually; so I had two internal examiners, both in the department - quite improper 33:5:09 At that time I remember nothing being said about Alan Turing; Turing was not a person who belonged to the maths lab in any sense; he didn't get along well with Maurice anyway; the heart of it was that Turing was a mathematician and Maurice was an engineer, although he had been a mathematician and they both sat the Tripos at the same time; I think Maurice always thought he got a better mark that Turing did; so Turing was hardly talked about in the maths lab in my day; he is talked about now as the subject has grown and so on, but Turing wasn't part of my life or the labs life in those days; he did influence Maurice but not necessarily in a positive way; they didn't like each other, they were very different sorts of people; Maurice's interest in life was to build computers to use them for doing computation which was far from what Turing wanted; I don't know how to analyse that but it is all history of course

Working on EDSAC2 and beyond

34:51:07 I wrote the first compiler in programming language for EDSAC 2 which was in fact my PhD; nice in those days that you get a PhD for writing a compiler, you wouldn't do so now; that brings Christopher Strachey into the discussion because Maurice had said I could stay and do research but wasn't quite clear about what I was going to do; what was normally done in those days was that you went off to another department and did their computing for them; I was assigned to high energy physics for this purpose, and I discovered two things; firstly I knew nothing about physics, and secondly they knew everything about physics they needed to know and didn't need my skills at all; they assigned me a supervisor, a guy called Hamilton, he was not part of high energy physics but at the side of it; he was a nice guy but I didn't get anywhere with that; I did this for about a month at the beginning of my three years; we then had in the department, on leave from the States a guy called Harry Husky (who died recently); a nice guy who was involved with Turing at MPL; he was interested in writing compilers - high level languages, they get the computer to turn them into machine code and off you go; I spent some time with him; he had written a compiler and I asked to look at it, and it is he who told Maurice Wilkes that I ought to be writing a compiler or something; then Christopher Strachey turned up; he was a consultant in those days doing lots of work, mainly for the Government promoted by Lord Halsbury if I remember; he came to the department one day and met me with Maurice; Maurice told him that he was thinking of giving me the job of doing an Algol compiler - Algol was a new international language and they thought it was going to be terribly important; Christopher counselled that they didn't do that; writing an Algol compiler which was completely unknown was just too much to do as he might fail; get him to start off writing a much simpler language which we called Autocode; there was one at Manchester at the time so we got hold of the Autocode compiler and I copied it; this lasted most of my three years; one particular twist I had was that in those days if you had a language and a low-level language underneath it, if you ran that program and debugged it, you had to go to the lower level and do likewise; what I did was to put features in the system so that you could stay at the higher level, you didn't have to play around with machine language, and that's what gave my thesis a bit of an edge to it; this was the first time this had been done here in Cambridge but it had happened elsewhere; I had been given it to do by Maurice, on Strachey's advice, just to get me started, but I went off at a tangent to do it well; so after three years I had done the compiler and it was in service being used by people in the University; many years later when I was a Fellow at Clare when I met a visitor from Princeton, an astronomer, at lunch; he asked me what I was doing and I told him about writing the compiler, the EDSAC Autocode; he then leant across the table and shook my hand and said it was a marvellous system; I have never forgotten that; Autocode actually survived into Titan because we made a mess of Titan on the software; we were late, very late, so we had an emergency program to write an autocode, an EDSAC autocode compiler for Titan; it was a disaster; David Barron and I were up in Manchester looking at the Manchester Atlas one day and had dinner afterwards and started discussing it; we looked into how they had done the work at Manchester; one thing they had done was to write these temporary systems to get off the ground; we thought that we ought to have a temporary autocode because CPL was going to be a long time late; Maurice Wilkes in those days had written a system which he called WISP, but he was very proud of himself; we came back from Manchester and told him we had an idea that he should write an autocode compiler in WISP because we needed it as an interim; he agreed, and a lot of people regretted it afterwards because they had to debug the thing for him; it was a bit of a disaster; so we brought in our big standby called Swinnerton-Dyer, and Peter wrote the compiler in the end; he also wrote an operating system for the Titan as well; incredibly clever was Peter; he could write programs single-handed that would take a whole time otherwise; so we had this rocky time at the beginning of Titan when we really didn't have any software which made the users rather angry, rightly so, and caused the lab to go into a period of being criticised; they said that it was not doing the right things, just playing with computers rather than providing a computing service; that had been its role when it was set up and not as a computer science department, and that was what Maurice wanted; it was called the maths lab because they wanted to put us into the mathematics faculty which took thirty-odd years to get out of; it was a service but it started to become scientific, intellectual, and it was doing both together; if peoples' intellectual research doesn't go anywhere nobody cares except the person concerned, but if you are running a computing service you have got to do it well; the period of Titan in the late sixties was a rather golden period for us on the one hand, but on the other hand being looked at very suspiciously by the University, particularly the important people like Teddy Bullard who wanted to stop us playing and to get on and provide a service; there was a lot of high-level animosity; Titan was a remarkably good system in my view but it had a rocky road coming in and people said these people are playing with computers and not providing a service; eventually they got over that when we got an IBM computer, then they said isn't Titan nice; we got to about the end of the sixties doing this and then the University said they must sort the computing service out and a report went to the General Board; they decided to reorganise the lab and have two labs in one rather than splitting them up; they would be kept in one department of which Maurice Wilkes would be Head but he would not be head of the service;

Director of the Computing Service at Cambridge and CPL

there was to be a new post of Director of the Computing Service and rather remarkably I got the job in 1970; that I enjoyed because I wasn't really developing into a computer scientist, I wasn't going that way, I didn't have the intellect, whereas people like Roger Needham were brilliant people, and the computer lab here is full of brilliant people
45:00:16 On CPL. We were still together then and David Barron and I were on the staff and looking after the software side, my job was not permanent but his was; we set about doing two projects, we wanted an operating system for the Titan and we needed a compiler in their language; we were still in the days where everyone thought they had an idea for a language and we should produce one; they didn't want to copy anyone else's but to innovate all the time - actually it was a great mistake - so David Barron and I designed CPL; David Wheeler joined in a bit and then Christopher Strachey joined us and took us off into quite an intellectual direction; the problem with that was that we nearly finished the language but we didn't finish the compiler, so we couldn't use it; that was another disaster because at a meeting of users chaired by Maurice in the Cockcroft lecture theatre, people got up and asked where CPL was; they brought out old scruffy bits of paper with memos from the lab that CPL will be this, that, by this date, and so on; poor Maurice was very embarrassed because we had set him up with this very ambitious project which we never finished; anyway, Christopher Strachey took it over but actually fell out with Maurice about that and he left and went to the States for a year; he came back to Oxford and set up his own group there and CPL went with him; so we were involved with CPL in the lab, but not for long; what happened then was that CPL was taken on by Martin Richards who wrote a simplified version called Basic CPL - BCPL; he did an excellent job and BCPL became popular with people writing programs; then people at Bell Labs saw it and developed their own language too and called in B after BCPL; the next stage in Bell Labs was that they then invented C; what BCPL and B had done was taken the complexity of the types of the variables that you were writing programs for and said there should just be one variable, just a half word or something, some 24 bits; then when it came to C they put the types back; I pride myself as being one of the parents of C at three levels removed, a great-grandparent; but because of all this we were still short of a compiler, a programming language for which to write programs on Titan and we were going to get into dead trouble there; so again we persuaded Maurice Wilkes to write an autocode compiler for Titan which was a bit of a disaster because Maurice was not the world's best programmer - he was best at all sorts of things, but not as a programmer; David Barron debugged this compiler for him but it never really worked properly; so we then set up another emergency project to write an autocode compiler which lasted into the Titan days; we realized that you don't invent languages for new machines but pick existing languages; so in the Titan period the programming languages side of the business wasn't really successful; what was successful was the operating system; that was due to Maurice because he had been over to the States in the mid-sixties and seen what they had done at MIT; they had devised what we called a time-sharing system where instead of having a computer with punched paper-tape or punch cards going in at one end and results from the other end you had terminals connected, several at once, so that the users could interact directly with the computer, all in parallel with each other; that is one of the things I was involved with, with Roger Needham and Barry Landy; that was really the success of Titan, the writing of an operating system, and it made a complicated machine very easy to use; so that takes us to about 1970

The IBM 370/165 and Big Computing

51:20:15 Titan was replaced in 1973 by the IBM 165 which was a political move; it has to be remembered that Teddy Bullard who was a geophysicist was then a director of IBM UK and he put a lot of pressure on the University to get an IBM machine, for good reasons; in those days if you are a big University doing big computing you wanted to have machines like the rest of them, a family of machines so that you can swap programs and so on; that had to be an IBM machine because all the other big universities had them; getting it was a bit difficult because it was against government policy which said we had to buy an ICL machine - buy British; we wriggled out of that and got this IBM machine; that did keep a lot of scientists who were used to swapping programs with people in the States really quite happy because we changed the operating system but we didn't change the interface between the machine and user-end programs; nothing was changed at all there, it was only the operating side where we were going to a terminal and doing this and that; we did something really quite remarkable in those days because it meant that the machine was so much easier to use, and we were one of the first in the country to do that; Phoenix comes out of that (by the way, calling things by bird names is my fault); without the Phoenix operating system you had to use the standard IBM one, JCL (Job Control Language) which is terrible; so we set about modifying the system so that we had a new layer on top; that was mainly done by Barry Landy, he is still around and was my chief systems man; throughout the seventies we improved the machine that way so that using a big IBM mainframe went on and on into the nineties, very unusually
53:55:08 The IBM started life as a 1 megabyte machine when we bought it and it had only been designed to have a maximum of 3 MB so we were going to enhance it sometime; we got to 2 MB and it turned out that there was a company selling IBM compatible memory which made IBM very cross; towards the end of the seventies we ordered these extra 2 MB so went to a 4 MB machine; that still doesn't give you an idea about speed, but it did make a difference to how much you could put through the machine; IBM once took me on a trip to Germany to some of their manufacturing facilities; while I was there we were in a restaurant one evening with the IBM people and I got really set-upon about how appalling it was that I had arranged for the University to buy something that was not IBM but connected to an IBM machine; I then tried to make out that this wasn't allowed, that they had done the design work for it, but this soon fell away as they were too much of a monopoly to let them do that; in general terms the modern computers etc. are far more powerful than anything we had in those days [see Table]; these things got bigger and better without us looking; there was no plan for saying now we want a machine with this and that, we just wanted the best that we could get; it is very difficult to compare machines; part of the Table showed how many users each computer had; EDSAC had fifty, EDSAC 2, probably 500; Titan had several thousand; then at the end of the chart I had my Apple machine and said that it had one user, me, it's mine; that is the big difference; it is very difficult to compare speeds although they can be compared; in my business as Director of the Computing Service I was concerned with providing a service to a rather variable number of people rather than what it is like with one machine; the computing service - the Phoenix system - was kept running into the early nineties, which was quite unusual, but it was running so well that they didn't want to throw it away; by that time, of course, everyone had their PCs as well

The Cambridge Granta Backbone Network

59:24:18 The Cambridge Granta Backbone Network didn't really happen until the late eighties; throughout the eighties networking was becoming more and more important and we were renting lines from British Telecom, whatever they were called, to link departments to the computer lab, the idea being and we had the big computer and they could connect their smaller computers to our big one and get some benefit from that; this we discovered was not what networks were about; networks were to enable people with their own computers to communicate with people, it was not to use a big mainframe, and we got that wrong; as we went into the eighties this networking business became more important and the fact that British Telecom by Government regulation controlled all networking, controlled us and what we could do with it; it was terrible, and becoming plainly obvious that something had to be done; I found myself on an advisory panel for the Prime Minister - there were six of us and we met at the Cabinet Office once a month and used to have brainstorming sessions; sometimes with Kenneth Baker chairing as he was the Minister concerned, though Mrs Thatcher turned up at some of our meetings; we set about worrying politically about this; JANET [Joint Academic Network] became the apple in people's eyes as it were through the seventies; in the early seventies we had networks but they were only for computing services, also the technology was very much in its infancy and a lot of research had to be done; by the end of the seventies there was a body called the Computer Board which was a DES [Department of Education and Science] body whose job was to give grants to universities for their computers; they did more than that because they gave grants for people to run computers as well; we benefited in the seventies very well from this policy of computing specialists, and they got into the networking business and set up a body called the Joint Network Team whose job was originally to learn about the technology and lay some standards down and build some networks for the academic community; I was sort of involved with that as I was a member of the Computer Board; Granta came in on the back of that as it was Cambridge's interest in that; networking was between universities and within universities; they were slightly different technologies as well; when we got into the Apple Mackintosh when they did a hard-sell for us, I was able to say that there was no point trying to sell them to me; I don't buy them, go and sound out the users, they buy them with their departmental money; Cambridge was unusual in that respect as we are such a big university that was the only way to do it, whereas in smaller universities the computer service would buy all the machines and hand them out; it is an odd thing to do because it constricted people; computing services were very good at constricting people to keep the job within limits so all were to have the same computer; absolute nonsense, we didn't do so in Cambridge because we had no choice; in about 1987 we had one of our five-year reviews with the Computer Board where they reviewed what we were doing in Cambridge and would give us a grant for the next seven years; we were therefore forming a strategy for Cambridge by which we would follow other universities and give it a fancy name, and it was called the Granta Project, Network was only just part of it; that went very well until we got to about the end of the eighties, about 1987, when I and my deputy director, Mike Sayers, a Fellow of Emmanuel, decided to design a network for ourselves; at that point in time you were not allowed to do that by law, you had to get it from BT; we started a project and put someone in charge of it; first we had to find the money; the Computer Board would not give us money for this, the University wasn't keen, no one was keen because what we were asking for money for were cables under the ground; we realized it was going to be very difficult to persuade anyone; we went to the Nat West Bank and they got quite keen for a time until they realized they would get nothing out of it whatsoever and lost interest immediately; as it happened, the Chairman of Nat West Bank was Eddie Nixon, the Chairman of IBM-UK; I worked on the University and kept putting the idea to them, I wasn't rejected but didn't get anywhere; the Treasurer, Michael Halstead, wanted to be up with the modern stuff and said he was interested; then I had to get hold of 31 Colleges asking if they wanted to join in; to cut a long story short, once you got seven to join, the rest could not resist; we got to about 1991 when we had a plan and decided it was going to cost £3,000,000 to dig up Cambridge; we made a design for it and it went as far as Addenbrookes in the south, Girton in the north; it took a couple of years to build and we had some interesting problems on the way, but we got it; we called it the Granta Backbone Network, the GBN; quite frankly, that was the best thing I ever did as Director of the Computing Service, much more important in hindsight than buying big computers - the networking was far more important; of course it is there now; these cables are underground, they are not doing any harm to anyone, they are not deteriorating, and they will work forever; you might have to replace the cable sometimes, but it's the ducting that costs; the Computing Service now has one or two staff looking after it; every time they find a builder building somewhere in Cambridge they go and talk to them indicating where not to dig; the World Wide Web came along at the same time; it wasn't a particular innovation of Cambridge but we were all following it through but it had nothing to do with building the network; it was rather an application of the network than the network itself; to get the maximum benefit from the World Wide Web we wanted everyone to have access to everything and for that you need a network; you do have a network today because you buy it but we have had a network in Cambridge since 1990; it was certainly one of the first and was one of the most spectacular because Cambridge is a big university; the trouble with Cambridge is that you have a city in which you wanted to build a network, whereas most universities are on a campus which is easier to build on because it's neat and tidy; in Cambridge you had to dig it all up to get what you want; the biggest problem was British Rail, trying to get across the tracks; it was a fantastic project; to me the warmest accolade I ever had out of that was when David Williams who was then Vice-Chancellor had come to the end of his year and was writing his speech for Senate House; in that speech he thanked Dr Hartley for his courage and foresight, and the word courage was a good word, and I was very pleased he said it; we did persuade all the colleges to do it, and it is growing all the time as legislation is now much lest strict; then you have got JANET which interlinks all the networks in every university now; on a campus university you can afford to have wider pipes so faster networks, whereas going from campus to campus you have to go through the JANET network which is pretty good as well; I was involved in the JANET stuff right from when they started in the seventies
Running JANET

1:12:14:07 Anything that makes use of the Granta Backbone Network is of interest to me and now it is big business within the University and between universities; in fact I went off and ran JANET for three years; once we'd finished the GBN I decided that it was time for a change; I had been Director of the Computing Service for twenty-three and a half years, I had been in the computer lab for fifteen years before that, I wanted a change; getting a new job at fifty something is not easy, but I got one; that was to run the JANET network because the Government wanted to privatize the JANET organization, get it out of the Science Research Council, and make an independent company; so I was hired to do that which I did for three years; they then threw me out as I was then sixty and Civil Service rules still ruled; I came back to Cambridge and got another job running the Cambridge Crystallographic Data Centre which Olga Kennard had founded; she was an incredible lady; she persuaded everyone to do what she wanted to do; she raised money by selling this crystal structures database with all the small organic molecules in it, and they were constantly adding to it making it bigger and bigger, and were renting it out to academics, pharmaceutical companies and so on; pharmaceutical companies were encouraged to pay $23,000 a year for this database, so the CCDC, as it was called, really made a lot of money; Olga persuaded the University to let her have a piece of land by the Chemistry Lab in Union Road; the organization went a bit pear-shaped just at the time I was leaving the JANET network and Olga contacted me; she said she was retiring though she didn't say why; in fact, Jack Lewis who was the chairman of the governors of her board had told her she should do so as she was then in her seventies and too old for the job; she asked me if I would take on the job; I said I would be interested, went to see her, and got the job; I did it for five years; it was a fantastic job because one thing we were not short of was money, it came out of our ears, and we were a charity, we didn't pay tax, an ideal organization; I am not a chemist, but I got it in order as an organization, and made sure that all the staff got decent pensions and that sort of stuff; I enjoyed it although I had no idea what they were doing; after five years I went to see the Master of Clare, Bob Hepple, and said that I was retiring and was looking for some work; in those days Clare had one Bursar and they realized that you couldn't let one person run the College, it was too big a job; they had decided to hire a Steward as they thought it a nicer title than Junior Bursar, and they asked me to do it for two years and then write a report to the Governing Body on what to do thereafter; I did that for three years in the end; that was quite fun though I knew nothing about catering or anything like that, but you don't need to if you are a manager;

Bletchley Park Museum and re-building EDSAC

then I retired finally; the museum bit came about very much as an afterthought; there are two bodies in community conservation in this country one is the Special Interest Group of the British Computer Society, enthusiasts who want to restore things, and then separately there is another group at Bletchley Park who run a museum, and we are very close together as complementary; I was asked by one of the trustees if I would take over the management of the museum on a part-time basis; I agreed, but then I found myself in a mire of difficulties mainly because Bletchley Park proper wouldn't talk to the Museum and vice versa; I had to sort them out; there was an ex-Officer from the Army running the other one - he was alright; the first job I was given was to take a draft agreement between us where we would share ticketing because we had separate ticketing which was nonsense; we were both keen on it so I put a draft agreement before lawyers and when I got back I showed the draft to the acting chairman of my trustees who said we couldn't trust the people at Bletchley Park and I was told not to talk to the head; then my acting chairman decided to get rid of me as they wanted a full-time rather than part-time director; I did not want a full-time appointment so agreed to leave but no date was set for my leaving; months went by and nothing happened and after a year I received my pay cheque and a P45, and left; I then went back to the Computer Conservation Society; I am involved in a project with them to rebuild EDSAC but it is also happening at Bletchley Park Museum; I managed to get hold of a very good project manager called Andrew Herbert who used to work for Microsoft, was head of the Microsoft lab here; Roger Needham and Karen died and had left a legacy of £1,000,000 to be distributed, and it was my job to do that; Wolfson got £250.000, so did Newnham; the EDSAC project started because Roger and Karen, posthumously, had been appointed members of the Guild of Benefactors in the University, apparently if you give £1,000,000 or more you join this group and they were so honoured, but of course they weren't there; I was invited with one or two others to attend the ceremony at Senate House with the Chancellor and have a little party afterwards, and I was very honoured to be their representatives, as it were; at the party Herman Hauser came up to me and I told him about the Computer Conservation Society; he asked if anyone had ever rebuilt the EDSAC; it had been mentioned in front of Maurice Wilkes long before and he was against it; I told Herman that I didn't think it could be done; he asked me to find out what it would cost to do so and to let him know; I discussed it with my colleagues in the conservation society, we did the research, and looked at the documentation in the University Library; Maurice had always said that if you wanted to find out about EDSAC it was all there; he was wrong as we went and looked at it and it wasn't all there; we did the research and wrote a report for Herman six months later; the problem was that we were conscious that Maurice would have been against it and he was still alive; he had expressed himself really forcefully about this; I hung on and didn't tell him until I had to because we were going public in October 2010; on one of my visits to his house I told him I had exciting news; he said I'd avoid that if I was you, said it again, then fell asleep - he was 97 by now; fortunately for the project he died in November that year so we didn't have to battle that through him; we went back to Herman, showed him the report, and told him it was cost £250,000 and take three years to do; Andrew Herbert is a brilliant manager and has about twenty volunteers, not all from Cambridge but quite a few are, who turn up on schedule at Bletchley Park where it is being done; it is due to be finished this year, we hope; we won't know it is finished until it is finished; it is actually in it's museum position already, so a lot of people have seen it, but no one has seen it running yet; that has been very exciting

Living through exponential change

1:27:08:04 My realization of what an amazing period this has been was certainly gradual as it goes back to 1958 when I joined the lab; then EDSAC had been thrown away and EDSAC 2 had come along and was the wonder of the age; then 1965 we threw away the EDSAC 2 and got the Titan, and that was the wonder of the age; so it's not that there was one big revolution, it was a gradual process; one could not have foreseen the world as it is now apart from a few people, and as time went on more and more people had the idea; it has been a gradual process because it's been controlled by electronics which have become cheaper and cheaper, smaller and smaller; in my mind it is an evolution which has gone on a long time and will probably go on, why not? each time it has moved forward, what computers can be used for has kept changing, so it's an incredible development, but so different; I have nearly always taken the line that cyborgs will not happen, there will be a gradual evolution and we won't start doing fantastic things we never thought of; I may be wrong of course, but I've tended to be on the cautious side; you can see the way things are going today but you can't predict the years ahead; some do, including Turing, but he was more sensible than most of them