 Bae gwaith yma… Maen nhw'n bwysigion yn ddweud dros cwyntlyg. Rwy'n bwysigion yn y pub ar y cwrnau. Rwy'n bwysigio'n ddwy yma ac oedden nhw'n gyfriffeithio'r pethau. Rwy'n bwysigion, ac yn y bwysigion ar y ddweud, sy'n ymmellig a'r pethau yn y ddweud yn ei wneud. Mae'r pethau yn y ddweud. Yn y ddweud, y pethau wedyn yw'r mwyaf, yn y gallu cwmau, Mae'r wneud i ddweud yn ymddangos, ond mae'r gwaith fel yw'r gweithio yn y bwrdd. Ond oedd y dyma'r lle, mae'r wneud yma'n gweld i ddweud yn mewn erbyn. Felly mae'r ddechrau a'r ddefnyddio ar y gweithio cyffredinol, yn ymddangos. I was reading The Guardian and actually in Prince Ford too, although I don't normally read it on my iPad. So it's good to be here back of course. I left this place to go to university in the 1950s and promising my father that I would then come back and we would talk about me joining the family business. This place was full of family businesses in those days making clothes. And we had a deal which said that if I got a distinction, if I got a scholarship I could go to Cambridge and I did get a scholarship actually, if you will nicely, don't know how. And then I do my degree three years of course and then we would talk about what I was going to do afterwards. I'd come back home and we would talk about my future. The trouble is I never left Cambridge. I didn't quite sort of say that was a deliberate ploy on my part but my father was very good about it. He saw the sense in it all. So I left here, went to Cambridge and say I'm still there, I love the university. Well I want to talk a bit about Hebding Bridge first, some of which you will know about although you may not. Hebding Bridge used to be a small industrial town and it flooded from time to time. As you know, he's been very popular about that. The town is at the confluence of the River Hebden and the River Calder. The River Hebden is fed by six reservoirs up in the hills, I know them all. They were there to control water flow and so on. And of course if it rained too much, there was too much in the reservoirs and they had to let the water out. And that's what caused the floods down here. So they discovered flood prevention, going to stop all that and that's not going to happen again. But it does because it's a very difficult task. 1947, I remember well actually, I was a young lad in those days. But somewhere in the archives or in my family is a cine film. It was a cine film taken of my father with his trouser leg rolled up to his knees, shoes and socks off, going down Market Street to Martin's Bank to get the wages for the stuff. And he liked all these local businesses. They cared very much for their stuff and I'm serious about that. But these floods were really not really nice. It took a long time to sort out and I said they kept coming back anyway. The last flood in 2015 got this place rather thoroughly. And my stepmother lived just across the bridge there, the old back of the bridge. And she had to be taken out of there and they got mountain rescuing to lift her out of the place. I don't know how they lifted it but they took her straight to a nursing home in Heland. And she'd been quite happy with her sense fortunately in rebuilding the flat. But that's what it's like in Hebdingbridge from time to time. It doesn't happen too often but it does happen. I found a quotation in something to do with the Hebdingbridge floods. I'll read it out. It's a well-known journalist of a newspaper which has been mentioned before. It's quoted by saying, the government pays grouse more owners to drain their glorified outdoor chicken runs. Meanwhile, it's the proles downstream who get the soaking. I had a fairly bitter comment that but it's a ride. Anyway, so anyway I stayed here for a bit and got my scholarship. But there's something else about the history I want to tell you about which you may not know so much about. You may have heard of the problems that they had in the 18th century here of the crag coiners. They lived in a house or farm up just past Mythenroyd, which you'll grow up. I used to actually live in Mythenroyd. And they made money by scraping the edges of the gold coins and using the bits they scraped off to make other coins. And they reshaped the coin that they mutilated. And this was a rather serious business because Hebdingbridge was quite isolated in those days. It was a big country and not as much as it is. And it became a problem for the government of the country because the pound or whatever it was they had in those days was getting a hammering rather like now actually. But this gang of people who did this and the coiners were led by one David Hartley. When you got someone who was an ancestor that long ago, however nasty he was, he was a hero. Or so called. Anyway, David Hartley led this band and he was hanged eventually at York. And he's buried in Heptonstall Church up. And he's still there. And you can go and have a look around there and you'll find David Hartley all long forgotten. But he was an awfully nasty man. Rumour has it that I'm a descendant of David Hartley. I don't think I am. I had a cousin, John Hartley, who as a young student did the research and managed to prove that he was a descendant of David Hartley, but not me. Well, so much for that. But he also had an initial R. He was JR Hartley. And the poor guy became a headmaster. Anyway, he's retired now like me. Anyway, off to Cambridge. My father was a very keen rugby player. He played for Halifax. In those days of course there was no professional or amateurs. He was downer, of course, but he was very, very keen on rugby football. And he wanted to go to rugby school. And his father wouldn't let him. His father says, you're going to work in the family business. So he said to me, he said, I'm not going to do that to you. You can do that or not. But he did send me to rugby school in Wales of all places. It turned out I was actually awful at rugby. And rather disappointment for Dad. But anyway, we made this fact about what I would do. I said, I've got an open mind about going to work in the business. All small companies around here. And he was agreed that if I got a scholarship, I could go to Cambridge. And if I went to Cambridge, I could go there for three years, then come back and talk about it. And as I said before, I never left Cambridge. I got this scholarship in Cambridge in mathematics. And that was all right. My school teacher thought I was wonderful at mathematics. And so I was, but the standards of the school weren't very high. But I went there and met my tutor. And my tutor said, ah, you've got a scholarship. Therefore you're one of the clever ones. And what we do with the clever ones is we miss out the first year's study. If you know Cambridge, it has its examinations and triposes. And there's a part one and a part two. And part two is a two-year tripose. And there's part three for the really bright ones. So I said, right, you miss out part one and do part two in two years. I said, oh great, sounds good to me. This is at the very beginning, the first day at Cambridge. And I did that and found mathematics at Cambridge impossibly hard. If you meet any mathematician at Cambridge, they didn't all say that as well, except that they really are bright people. And I turned out I wasn't that bright at all at Cambridge standards. But I worked like mad for two years. And I got a second class in those two years. And I was rather proud of that. But no more mathematics in my third year. Absolutely none. But you can't get a degree at Cambridge unless you're there for three years. It doesn't matter what you do, because you're there for three years. So we had to find a course for me. And again, the Cambridge tutors come in and they're very helpful people. And they said, well, we've got this and that. There's engineering, there's economics, and it didn't sound good at all. Eventually they found an odd course, which was really a fourth year course. Not for a degree, but for a diploma. And he said, you might like this. He was called the diploma in numerical analysis and automatic computing. Brand new. We're talking about the late 1950s now. And so I did that for something to do. And I had a ball. It was wonderful. I introduced to computing. And I thoroughly enjoyed my third year. And that, of course, turned me round. And after the third year, I went to the head of department, Professor Wilkes. And he said, you've got a good degree in your good examination result in your diploma. You can stay on and do a PhD. I said, what's that? Well, you tell me what happened to having a PhD and all that. It sounded nice. They called you a doctor after that. You know, it sounds very grand. And there's also another reason as well, which was that in those days there was something called national service. Now, probably very few of you here are old enough to go about national service. But you had to go into the army for two years before university or after university. I hadn't been here before. But if you stayed on to do a PhD, you're exempt from national service. So that's why I did national service. I did the PhD. It was to avoid national service. So that's all right without that. So that was very good. I thoroughly enjoyed that course. We spent part of the time turning handles on Brunsveegers, if you know about what Brunsveegers are, and part of programming computers. In those days we had Edgesack 2. You've heard about Edgesack 1. You've got to hear more about it this week. And it was one of the very first computers and so on. But the problem about Edgesack 1 was that we went and built Edgesack 2. And we knew how to build a computer by then, and we thought we did. So Edgesack 2 was what streets ahead of Edgesack 1. Edgesack 1 had about a few hundred words of memory. Edgesack 2 had a few thousand. Edgesack 2 was made out of transistors, not valves. Its memory was made out of magnetic cores. Hence the name Edgesack didn't really apply. And it was so much better than Edgesack 1 in every way. No-one wanted to use it. I mean the new one. She was much better. Also it was taking up valuable space. So you only want what you can do. You throw it away. That's what you did in those days. You never thought of saying this is historic relic, anything like that. And so they threw Edgesack 1 away, literally. And in July of 1958. And I arrived in the department to do my fourth year course about two months later. So I didn't really ever see Edgesack 1. So don't ask me about Edgesack 1. I'm not that old. But Edgesack 2 was wonderful. And it was a paper-tape machine, of course. It was very, very slow by today's standards, but very, very fast by those standards. And that's because of the history of computing. As we go along, the machines get bigger and faster and so on. And even smaller. And it's difficult to say, well, why has it gone that way? Because technology makes you able to do things you couldn't otherwise do. So I was allowed to do research on Edgesack 2. And that was fine. I wrote a compiler for Edgesack 2. I invented a programming language. And you can't do that nowadays. It's too easy. But in the 1950s, designing a language and writing compiles was a wonderful thing. And it wasn't really. But anyway, I was very, very lucky there. I got my PhD. And I went to the head of the department, Maurice Wilkes. By then, they were looking at the third computer. The third computer was an Atlas. And it wasn't built by ourselves. It was built by Ferrante and Manchester University. And Atlas was much bigger, much faster, and so on. Then Edgesack 2. So we set about doing that through Edgesack 2 away and all the rest is history. I went to see Maurice at one stage. When we were in the early stage of designing the machine and I was enjoying that. And I got my PhD. And I said, Maurice, any chance of staying on? Have you got any jobs in the department? And we said, no, I'm not sure. Not unless somebody dies. These are very words. And encouraging words in a way. But I was sort of kept on in a funny way. Until suddenly he said, ah, I've got your job for you. And he told me how you're going to pay me £1,000 a year. And I said, good, thank you. And I didn't know to many years later how he got the money. Apparently, when he was getting the Atlas designed and buying bits and pieces for it, he had a choice. He could either buy it with seven tape drives. These big things that walk around. Or he bought it with six tape drives and hired me. And that's how I work. I'm a tape drive. I was very grateful for that, too. So that went on. And we got to 1970 when the university decided to reorganise the department. They had to stop having computer scientists running a computing service because they kept meddling with it. And so they split the department into two and created a separate computing service. And they had a new job of directing a computing service. And I got the job. Everyone said, you don't want that job. Better nails. Everyone's going to hate you. And I said, no, no, I enjoyed it. So I'd stayed doing that for 23 and a half years. And it was quite fun. But what happened in that period, as time went on, the machines got bigger and faster and so on and we got them upgraded and we got more and more users on board. So go back to Ed Sec 1. There were probably about 50 users, which was quite a achievement those days. Three of whom were Nobel Prize winners. And then when Ed Sec 2 came, we had the odd couple of hundred users. And then when Atlas came into operation, we had the odd thousand users. And by the end of the 1970s, we had 8,000 users on that machine. Ridiculous. Think of all sharing a Macintosh between 799 users. Because then suddenly it changed. It was all very different. And we had a problem, of course, what to do about that. A computing service with people sharing out a computer amongst 6,000 users. When all you had to do was get a bit of money and go out to the local store and buy yourself one. And of course, they were getting bigger and faster and so on. And everything was shifting up. Anyway, we did all that. And we had, eventually, well, billions of words of memory, a very fast machine and all the rest of it. And so it went on. And then we got it in the 1980s, which was another critical time, because personal computers were arriving and users used to go and buy them. Now, if you're in many universities, you were not allowed to do that because the computing service said, I have the same sort so we can look after you all and so on. We were having more enlightened and we decided first of all that they were using their own money to buy these things. And who are we to tell them how to spend their money? So we let them. And I said to my staff, your job is to help users with their computers, which they buy with their money. And no matter what they buy, you've got to try and help them. And that worked really pretty well. Far from being the better nails job, we got quite popular. So that went on into the 80s. And we come to, again, a big change of technology and that's called networking. One of the problems of Cambridge, Oxford for that matter, we don't talk about Oxford very much, part of David, Cambridge is not a campus university. Most universities are a campus university. You've got a spatula land, a pile of buildings on them, and that's the university. And he wants to wire up the campus. You can just do it because it's all of your land. You've probably got ducks underneath the grass anyway for central heating. So it's quite easy for them to build their own network and get free of the big enemy BT. And we can do that. One day, the end of the 80s, I and my deputy director were having a brainstorming session. We said, let's do something daff today. And what would we like to do if we really could do it? We said, we'd like to wire up Cambridge. You know, once we got our own wires down there, to hell with BT, we could run as fast as we liked. We wouldn't cost anything once the doctor was down there. And we set about it in three years, mostly for the politics. We had to get government approval because in those days, BT was still a monopoly. This is about 1989. But we got there and we persuaded the university and the collegeers to dip in the £3 million that we needed. Costs three and a half million. And we dug up Cambridge and we put down the things. And of course now, therefore, we are like a campus university and we can put fiber down where we like and do. And that makes an enormous difference. And if you're having to go to BT, who always give you something less than standard, because they don't want you to benefit too much until you pay some more money. And that was quite an achievement we had. If you go to Cambridge and walk around the streets, you will see a lot of manhole covers. And they say BT on them, or something, or post office. There's another set there. It says university on it. And that's ours. And it's very, very valuable. And we enjoy that enormously. When I left the university to go to run Janet, the Vice Chancellor of the Day, who always gives a speech at the end of the year saying how good a year it's been for the university and who's left and who's died and the rest of it, you get mentioned if you're important enough. I got mentioned. The Vice Chancellor said, we're grateful to Dr Hartley for his courage and foresight. And the word courage was a good one. It really was. We had quite a battle, but we got it. And networking to us now is now the most important thing for computing service runs, not the computers. Computers belong to users, but networking is needed to glue them together. So, finally, I want to say something about it's second one. I haven't talked about it's second one. I don't know much about it's second one. I never saw it's second one, except on a trip with a student once. And this was only five years ago or so that I was in a group of people talking to Herman Hauser. Now, you all know Herman Hauser, the father of the BBC Micro and Acorn and so on. A good friend to all of us. And I met Herman at this party and he said, oh, David, what are you doing these days? Well, in those days, I was just getting involved in computer conservation and being a member of the Computer Conservation Society. And he said, oh, that sounds interesting. Has anyone ever rebuilt EDZAC? I said, of course not. Impossible thing to do, we can't do that. Oh, he said. Oh, how much would it cost? I said, I've got no idea. Absolutely no idea. Oh, he said. Why didn't you find out and let me know? I never said that to you, you know. It's rather a nice feeling. It's true. And six months later, we got together and we wrote a report. Cabinet's one of them. And we presented Herman with the report. With the answers, your question, Herman, a quarter of a million pounds and take three years. It has cost a quarter of a million pounds and it's taken a bit more than three years. But EDZAC is now, it's built a replica. It looks like EDZAC. It runs like EDZAC. It's a bit different in size. It doesn't have mercury delay lines, for example, because it's far too dangerous these days. So we have got delay storage. And it's been built, first of all, in the volunteers' homes, in kitchens and what more, all around Cambridge and anywhere else in the country. And now it's been all bought together at Bletchley Park in the museum where it's going to live. And it's getting very close now to working. So you've not seen EDZAC, go and see it. Go and have a look at the thing. But back to Herman and his money and all that. I think we were very pleased. And it was an enormous job to find out how did EDZAC work. People used to ask Maurice Wilkes about EDZAC. And he wouldn't tell you much. He said, you want to find out about it, go to the archives, go to the university archives, it's all in there. And when we started this project, we went to the university archives and he was wrong. Maurice Wilkes was very possessive about EDZAC. He was his machine, he designed it. He had the vision about it and so on, which was wonderful. He wanted to have a computer that people used in the university. He didn't want people to have a computer to sit in them and making it faster and faster. He didn't want them to be able to play with it. He wanted a tool for research workers and so on. As I said before, he did create at least three Nobel Prize winners. For which we were very grateful for what Maurice did. He died in 2010. We were about to announce the project to rebuild EDZAC when he died, which in a way was perhaps useful because we knew he'd be against it. And I have worked with that man so much in my life that I know what he's like. He looks you in the eye and says, you can't do that, can you? And challenged you. I never very seldom won those arguments with him. I was trying to work out how I was going to handle that one. He was 97, so sadly he died. So he didn't know we were going to do that. So there we are, a bit of a potted history of EDZACs, EDZAC 1s, 2s and so on, and what it's like to run a computing service. Thank you very much.