 I love the EMF crowd and that's one of the reasons why. I actually got my best cheer ever at a presentation at EMF two years ago because I got to come up on stage and announced that the toilets had been unblocked. So, I like to call this slot the O-sod-the-tent. Let's get a beer and watch something slot. So, I hope you're all settled in. I hope you're all having a good time. Yeah, excellent, excellent. Right, let me ruin that for you. I'd like to tell you a bit about myself first very quickly. 100 years ago I used to write for this magazine about this computer. These days I can be found at the National Museum of Computing where I give talks on the world's first computer Colossus. Now, the reason I put this talk together was I kept getting asked the same question again and again when I gave these talks and that is we had a great computer industry. What happened to it? It all got thrown away. We were brilliant in the 80s and now we're nothing. I always give the same answer. Nothing happened to our computer industry. You just don't know about it. Now, time for the history. We have to start by asking ourselves what is a computer? Now, this is a much harder question than many people realise. It involves a lot of geeks getting around with a few pints of ale and doing some serious beard stroking while we try and work out just what is a computer. Now, I can tell you this much. These are computers. There's a lot of computers in this room but it's not the machine. It's the people. Computer was a term like bookkeeper or accountant originally. They're using mechanical calculators to do all the calculations we do with computers today. They are the computers. Is this a computer? This is around 150 maybe more BC and it's called the anti-caffera mechanism. It was discovered in 1901 but it's only recently that they've realised that it was some kind of analogue computer that may have predicted lunar cycles or eclipses. Its true purpose is still not known. But if so, if that machine is rebuilt and works the way that we think it is, that would definitely be the world's first analogue computer. Now, in answering the question what is a computer, someone came along and wrote down all the rules for us and we'll come back to him in a bit. But I'm going to start in 1801. Jackard produces punch cards for lunar machines. These punch cards are a series of instructions that the machine can read and understand to produce different types of patterns as they weave. It is code. It is the first program. It might not be for a computer, but this is the first time you can instruct a machine to do different things. Babbage takes on that idea. Now firstly, he's been working on the difference engine. The difference engine was meant to be a superb machine for doing calculations. Definitely not a computer because it's linear in operation like a calculator is, but it never gets built. Although he tried in 1822, in 1831, disputes over the elaborate machinery and the machining cost caused the project to fail. If you take a look at his close-up of the replica that's been built in the science museum, it just wasn't possible to manufacture that kind of component at reasonable cost in those times. He abandons the project, but he does start designing a second machine called the analytical engine. Now, this never got built. He was constantly redesigning it up into his death. If it had of been built, this would have been without a doubt the world's first computer. It would have been what we called shoring complete, and I'll get to that in a minute. It used those jackard punch cards to feed programs into the machine. The machine could branch and make decisions based on what the results were. This is the start of a prototype replica which should be complete by 2021, and we'll find that out once and for all. Interestingly, Babbage was in constant communication with Ada Lovelace. She was the daughter of bad and dangerous Snow Lord Byram. She was a mathematical genius, and she worked with Babbage on putting together the first algorithms that the analytical engine would have run. This is now accepted to be the world's first computer program, and this is our world's first computer programmer. Exactly. We like that. Let's skip forwards to 1936. Alan Turing. Alan Turing publishes a paper, and I hate this bit. Uncomputable numbers with an application to the Enschlangding's problem. If anyone speaks German here, I've just mangled that, haven't I? Yeah. It means decision problem, and it is this problem in computing about what makes a computer different from a calculator is that, based on one result, it can go one way or the other with the next bit of code it runs, branching. Now, this paper described a machine called the Turing machine. Here's an attempt to create one. Now, Turing machines demonstrated that you could solve any mathematical problem in the universe using ones and zeros. You just had to build a machine that could process them. This paper is accepted as founding computer science. It didn't invent the first computer, but it did start the branch of computer science. The reason you can't build a Turing machine is to solve every mathematical problem, the tapes on the two sides of it have to be of infinite length. So it's a false experiment. But what Turing did next is a lot more interesting. He went here. Anyone ever been to Bletchley Park? Oh, excellent. Okay. Right. I used to be a tour guide there, and I used to love showing off this machine. So I'm sure many of you are all familiar with this. Bletchley Park was the home of the code breakers during World War II, and it's principally most famous for breaking the enigma machine. A German siphon machine so complicated it had 150 million, million different combinations. Turing is one of a team who successfully broke that machine on a regular basis. His idea was don't try and find what the settings are, try and find out what they are not, and reduce the number. To do that, he designed with Gordon Welshman a machine called the bomb, with an E on the end, bomb. If any of you have seen the imitation games might look familiar, although they referred to it as Christopher in the film. It's actually called the bomb, and that took the process of breaking enigma down from something that took weeks to around a few hours. But it's not a computer. This is an electromechanical device. It doesn't have any code that it runs as such. Another siphon machine that Bletchley Park were up against in World War II was this one, the Lorenz SZ42. Now this was the machine used by Hitler and his high command. Much more powerful than enigma, much harder to break. But that intrigued me. Now it wasn't until a German military operator made a huge mistake and sent two messages using the same crypto key that Bletchley Park broke into it, and a man called John Tiltman broke it. Bill Tutt took that information away and created an algorithm that could break these Lorenz messages, and this bearing in mind, without ever having seen the machine or having an ounce of technical data on it, they just reversed engineered it from the brakes. He was able to build an entirely working replica of the machine and also had this algorithm to break it. Tommy Flowers took that algorithm and had a bit of a breakthrough. He realised just like Turing said, it was all to do ones and zeros, and things are great at ones and zeros on and off, true and false, of valves. I don't know if you can see that. So Tommy Flowers, actually working for the post office building phone exchanges, went on and with his valves designed the Colossus Mark 1. This is an almost Turing complete computer, and it took the breaking of Lorenz down to six to eight weeks per message to around 40 minutes, 300 messages a day. It broke thousands of messages towards the end of World War II, including one that was a direct order from Hitler to his generals in Normandy to move to Calais and fortify it because he thought the Normandy invasion was going to be a trick. That information set the day for D-Day. So successful was this machine. A second version was built. This is 2,500 valves, and you can go and see this one. This is a rebuild of one of the 10 Colossus that were built, and that's at the National Museum of Computing, quick plug there, and that's a wonderful thing to go and say. If you're very lucky, I'll give you a talk on it, no. Right, now the problem was that when they went away after World War II, all of this was covered by the Secret Act. You couldn't breathe a word of it to anyone, and in fact, the existence of Colossus was not public knowledge until around 2001. We knew there was a machine there, but we didn't know what it was for, and it wasn't until 2001 that secret was revealed. So a lot of people say the World's First Computer is actually Americans, and the Americans come and ruin it for us all, and they say, no, 1946, ENIAC, University of Pennsylvania, it's a monster, 17,500 valves, five million solder joints, and it was built particularly for military purposes, wasn't very successful, was not reliable, it's more of a sort of ongoing development of machines. But actually, Colossus was there a few years earlier. But all the people who were working on the Colossus project, Alan Shorring was on the periphery, Max Newman, who headed up the project, they all went off to universities, and suddenly out of nowhere, these universities post World War II got very good at building computers, but they couldn't say why. So you got suddenly a whole rash of computers coming out like the Manchester Baby or Manchester Mark I. Alan Shorring designed the pilot ace, but sadly he left before the project was completed, and sadly died shortly afterwards. Now another computer of note is EDSAC, the Electronic Delay Storage Automated Calculator, built by Remick and Wix in Cambridge. Now this, you can say, is the first reliable computer, the first useful general purpose computer, and it was the first computer to have RAM, because Colossus and all these ones before it had zero bytes of RAM, they couldn't store anything. They had a bizarre idea, they used mercury delay lines. These are tubes of mercury that you send ultrasonic waves down. The mercury slows the wave down, and then you've got a piece of information in stasis and you have to loop it round and round and round. Very bizarre, but it worked. Now this is all very high tech stuff and very expensive to build. The university couldn't fund it, so how did they pay for it? Was it some scientist wanting to get an early start on the space race? Was it the military looking for the calculations to build a super bomb? Was it some evil geneticist candidate some kind of monster? Oh, the political joke worked, nice. Who paid for it? Well of course this was Britain, and if you're going to solve a computer, use a computer to solve a problem, you're going to need to do something about tea and cakes. Now for any foreign visitors we have with us, firstly welcome, secondly we need to impress on you how important this is to the British people. This is the queen's 90th birthday celebrations. This is the most British photo I have ever seen. They are happy for two reasons. Firstly, they're getting a cup of tea. Secondly, they're going to be cute to do it. It's fantastic, and it's raining, so we can all complain about the weather, three reasons. So based on that, who paid for that computer to be built? Well actually it was Jay Lyons and Co, the tea shop. Now this wasn't any ordinary tea shop, this was a Starbucks of its day. It was everywhere. Every high street had one, they had big restaurants in London, they invented the ready meal and used to sell them in the shops. They were a huge operation and they were having a problem with bakery valuations. Here's the one of the five bakeries, get an idea of scale. Every day they had to have the fresh ingredients and the right amounts with least amount of waste to save expense, to get all those cakes they had to get the orders in and process them. It was a massive operation and they had the foresight to think these newfangled computer things could help us. They bankrolled Edsak and then used the information they were given at the end to build the world's first ever computer for business purposes, the Leo, the Lyons Electronic Office. This did the bakery valuations with daily orders from the shops and the payroll and that went live in 1951. That was so successful they thought, maybe other people were going to buy them too and they formed Leo Computers Limited. This is important. This British company was the first ever company solely designed for building computers. Don't set up for that job. Other companies like IBM fell into it as they evolved. This was the first ever commercial computer company. They built Behemoths like the Leo Mark II, did very well, had sales of 11. Now thankfully they were so expensive that made them a fortune. Slightly more successful Leo Mark III which sold a massive 62 and one of those was still in use by BT in 1981. I'm not joking. They were revolutionary because they updated the storage to these things called core stores. These are made up of a horizontal and vertical axes. Each crossover point has a little magnet in them. I'm putting current down there. You can charge the magnet one way or the other. There's your one and zero. They were also non-volatile which meant when you took the power off, they carried on storing the memory. That was the world's first SD card. That is a whopping 1,024 bytes of memory. Right there, 1K. Now there were other companies coming up at this time who had evolved into making computers. They were English Electric and they had just merged with Marconi, who you may have heard of. Eventually as things consolidated as they always do, Leo joined them as well. You've got these three companies come together, this new powerhouse of British computing. What are we going to call them? They called themselves English Electric Leo Marconi and kept the rather questionable logos. Another company came along at the same time called Elliott. They've been making scientific instruments since the 1800s but in 1967 they joined the company as well. They've been building things like the 405, the 803 and the 903. Again, that's our one in National Museum of Computing, still fully running. Then some other companies came along. Now these companies didn't have bad logos. They just had names and they weren't happy with that so they wanted to merge and be a competitor to English Electric. They formed together and got themselves a nice nasty logo, which is International Computers and Tabulators. Then the Americans turned up and ruined everything. They launched the PDP-8. I've got a replica of one here. I admire the blinking lights. These were the first personal computer. They sold 50,000 of them because they were useful for smaller tasks. That's what mini-computer means. It doesn't mean the size means the task. If you had a reinforced steel desk, you could put it on your desk and use it as well. This caused a seismic shift to the industry. The Labour Government at the time, led by Wilson, got Tony Ben to actually merge ICT and English Electric Leo McHoney, take all their terrible logos and make one truly awful one, which is International Computers Limited. In 1968, another little company set up then as well. I'm not sure what happened to them. Right. A short break. Thank you. I just want to be the first ever presentation to have an ad break in it. I'm already running horribly behind and they're going to start throwing things at me. 1973, Clive Sinclair sets up Sinclair Research. He's been working along with smaller things like amplifiers, record players, that sort of thing, but he's convinced by his partner Chris Corry to start on a computer project. Chris Corry heads up the team that gives us the MK14. This was the Raspberry Pi of its day, an amazing price point of £39.99, about £260 now. But you've got a little computer. This is the time when computers filled rooms. This was unbelievable. Still a very nice product, but they thought they were going to have huge success with it. But the Americans turned up and ruined everything because they had the Commodore Pet. Although it was very expensive, this was a one-plug deal. You plugged it in, switched it on, you were computing £32K of rock and memory. Chris Corry wants to fight back against this. Clive Sinclair isn't that interested. So Chris Corry decides to go it alone. So in 1978, Chris Corry leaves Sinclair. He tells Clive in the barren a beef pub over lunch. Clive proceeds to chase him around the pub with a rolled up newspaper. And that's how a pub fight revolutionized the computer industry. And I'll tell you why in a minute. Because Chris goes off with Herman Hauser and forms Cambridge Processor Unit, not the catchiest name, so they decide to trade under Acorn so they go above Apple and the phone book. The 8-bit wars of the 80s begin. Sinclair's ZX80 is already out. They react with the Acorn atom. Much more expensive but more powerful. One of my favourites. ZX81 comes out. First sub £100 computer if you're handy with a soldering iron, which I wasn't at it turned out. Amazing, sorry. And that was an amazing thing. But at that time, the BBC, British Warcraft Casting Corporation wanted to have an educational computer and they were going to pay a lot of money for it. Sinclair and Acorn both went for it. I think most of us know who won. It was Acorn with a masterpiece, the BBC Model B. But Sinclair sort of had the last laugh. It was an amazing computer. What really succeeded in the homes was his competitor, the ZX Spectrum. Yes, I've got one of those as well. Wonderful computer. And they tried to react because the BBC was very, very expensive by comparison. So they came up with the electron, which is a bit smaller. Didn't do that well, really. But the Americans were trying to ruin it all at the time. And many of you who fought in the eight-bit wars in the 80s will know about these fierce Sinclair Commodore rivalry that took place at that time. I'll just give you my opinion on that. There we go. But something else seismic happened as well. The birth of the video game industry. This was truly remarkable. All these kids up in bedrooms, starting to code away, trying to learn the machine code. Suddenly they were driving Ferraris around. There was this massive explosion. Some of these companies still exist of us today. And suddenly you had all these millionaires and this new industry was born in the UK. But what about the business sector? What was going on there? Well, the Americans ruined it all. Because in 1981 they invented this. The IBM PC came out and it totally dominated the business industry from there on in. Because our industry at the time was building behemoths like this, the 2966. This is a small one. It used to run McAlpine's payroll. And there's one belief to still be running in Russia that is four floors high. But why have one of these? Well, you could have one of those. Another little thing happened around the early 80s as well. A new chip was designed by ACUM. And it was called the ARM one and it was designed for the R communities machines. Berlar in mind will come back to it. Also in 1985, we did have a decent good stab back at the PC with this remarkable little device. Alan Sinclair introduced the PCW. £199. You got your monitor, your software, your keyboard and your printer. A complete all to go business machine. It sold 8 million units. The biggest selling UK computer of all time. However, consolidation happens and things go wrong and the industry actually started to collapse in 86. ACUM was purchased by Olavetti. And Sir Clive Sinclair sold his business to Alan Sugar. And what did we do during the 90s? That. Horrible little beige boxes. That was the computer industry in the 90s. So much so that in 1990, ICL was sold by a two Fujitsu who still owned the intellectual rights to this day. And the actual brand was discontinued in 2002. Gone forever. Also ACUM was split into two companies. One was called Element 14, dealt with broadband. And the other was ARM, continuing the beautiful tradition of really dull and terrible British computer logos. ARM was a venture between what was left of ACUM and Apple and VLSI. And it was to design and make chips for this, the original iPhone. It didn't go so well, that machine. The idea was that ARM would only design the chips, they would not manufacture them. So it was an intellectual property. They sort of went along in the background doing quite well with themselves, growing steadily, but in 2007 they exploded. Every iPhone ever made has an ARM designed processor in it. They may call it the Apple A5 or the A7 or whatever, it's an ARM processor and it was designed in Cambridge. And just to show this is a bipartisan presentation, there's a lot in these ones as well. All our Android friends have lots of these wonderful ARM processors in them. In fact, 14 billion ARM-designed processors were manufactured around the world last year. And on that point, I was going to wrap this presentation up. I was going to think, you know, makes you proud to be British, that doesn't it? Oh, wonderful, wonderful ARM on the computer industry. Ars. It hasn't happened yet. Softbank have offered a heck of a lot of money. Herbert Hauser is still involved and the company doesn't want to sell. And they've said they will keep on Cambridge, they will double the employees, but as it sort of demonstrates with this talk, that doesn't always happen. So what are we left with? Well, actually, it's not all bad news. Imagination. Ever heard of them? Only, yeah, not that many, but you will find their design processes in every iPhone as well. They design GPUs. They're based in Kings Langley, just outside Hemel Hempstead. Apple are currently looking into acquiring them for billions. British company. And of course, we've got our wonderful video games industry. The video games industry is now bigger than Hollywood. And a lot of it is in the UK. Let's look at some of those names there. So at the top, you've got Rockstar, Grand Theft Auto series, say no more. They're based in the UK. You've got Frontier, David Braverham's company. Originally back in the 80s, they did Elite. Now we have Elite Dangerous and the new theme park simulators coming out as well. Rare came from the ashes of Ultimate Play the Game. They're now owned by Sony, but still based in the UK. Us 2 did Monument Valley, one of the most beautiful mobile games I've ever played. Team 17, Lemmings of course, and that series of games. And Hello Games. Now on Tuesday, Wednesday, depending on what you believe in the papers, they are going to publish No Man's Sky, which looked to be one of the most astounding sandbox games ever released. Strouder than secrecy about what it's actually all about, but it's all stuff that's coming out of the UK. And of course, it would be wrong for me to leave out these guys. David Braverham and Eben Upton formed the Raspberry Pi Foundation to perform a new kind of computer. This is the MK14 of its day. Back again. History is repeating itself. This wonderful, cheap, 30-odd pound computer came out and completely took the industry by surprise. If that wasn't enough, they decided in 2016, this year, to make the Pi Zero. A computer so cheap and so small, they decided to give it away on the front of a magazine. And also, we must mention, of course, our friend's BBC Microbit. That's a wonderful little device and that's being given to, I think it's Keytage 2 children. All free of charge. Every child in the country qualifies for it. And unlike the model B, the original BBC, which is more about having your screen and writing your code, this is about interacting with the world around you. Very appropriate for the sort of people we find here. So, let me leave you with a final thought. This photo was taken in 1957. This is an Elliott computer being delivered to Norwich Treasury to help with the budget calculations for the council. The day after the Pi Zero came out, someone went and took this photo. The question we're all in charge of answering is, what's that photo going to look like in 60 years' time? Thanks, everyone. We've probably got time for a couple of questions, if anyone's got any. I've got one, actually. You mentioned machines being Turing Complete. Yes. Or partially Turing Complete. What does Turing Complete actually mean? I did want to get into more detail on that. Turing Complete is a set of definitions of what a computer must do. The one that gets argued about all the time is it must branch, which means it has to make a decision based on a result it's calculated. So, people often say that Carl Zeus's machine, the Z3, which predated Colossus by a couple of years, is the first computer, but it could not branch, so we don't count it. Any other questions? Well, I was going to say, I'm happy to take questions down here. I've also had some interest and people want to have a look at the PDP-8 replica, so I'm going to be over in the robot arms because I'm parched. If anyone wants to come and play on it, you're more than welcome to do so. If you've got any questions, I'll be down there. Thank you very much, everyone. Okay. Thank you very much, PGA Evans. Thank you.