Turing: Pioneer of the Information Age

Stanford University welcome to e3 80 Stanford is a research institution for higher learning for which the facts of the matter dominate taboos and opinions considering the facts it is my opinion that without Alan Turing Stanford University Silicon Valley and the world would be a very different place the advance of our species in the past 200 years of the Industrial Revolution is marked by our increasing mastery of machines and the enslavement of the population to its cause inspired by the mechanism of perforated cards invented by Joseph jacquard and the later work of Charles Babbage and the Ada Lovelace Alan Turing formalized the concept of the universal machine and enabled the development of modern computing machinery in the lost war against tyranny Turing gave His life selflessly to an ungrateful and intolerant society he cracked the Enigma code that enabled the British Isles to overcome the German u-boat blockade that held Britain hostage his contribution shortened the war against Hitler and helped to ensure its victory it is difficult to consider Alan Turing's contributions without taking these broader matters into account during was the first computational biologist the father of what would later become artificial intelligence and he was a modern philosopher of the mind we are left to wonder what he would have made of what followed and he lived We are privileged today to have with us Jack Copeland one of the world's leading Turing scholars jack is shortly on his way to England where I shall join him with others in June to participate in the International Alan Turing Cell Grayson's Jackie's professor of philosophy at the University of Cambridge New Zealand he is currently the Royden beef Davis chair of interdisciplinary studies in the Department of Psychology at Georgetown University I'd like to welcome Jack Copeland thank you well thank you all for coming and thank you all for watching those of you who aren't here I hope you're going to enjoy it this is the talk of the book of which this is the cover and the book will be out very soon I hope I haven't quite finished it yet but I nevertheless hope it will be out very soon the publishers were expecting it at the end of last month but I fear they've been disappointed but it shouldn't be long now I blame the the irresistible attractions of Georgetown which is the most vibrant and exciting place I've ever lived in I guess no but anyway the book is on its way honest so let me show you some picture couple of pictures of Turing to get things going there aren't many photographs of Turing in existence and I'm grateful to the Turing family for permission to show these to the softer noon he was not really much photographed during his life a few stray snapshots and then six to eight photographs all taken in the same and photographic session the one in which this was taken and the previous one that I just showed that one photographic session which was probably for a passport is our best visual record of Alan Turing in fact so Turing was a shy genius studying mathematics at King's College Cambridge and in 1936 in his very early twenties he completely unexpectedly invented the fundamental principle of the modern computer Turing was working on an abstract problem in the foundations of mathematics hilda's decision problem no one could have guessed that such abstruse RK in work could have led to anything of any practical value whatsoever let alone to a machine that would change all our lives but it did as everyone who come operator PC knows the way to get the machine to do the job you want word processing so is to find the appropriate program in memory and click on it that was Turing's invention the so called stored program concept that was the crucial insight that he had in 1936 this was all on paper of course there was no actual engineering in those days that would come later Turing showed though that by inserting different programs in memory what he called his universal Turing machine could be made to carry out any tasks for which a program could be written his fabulous idea really was for a single machine a single slab of hardware that by making use of coded instructions stored in its memory could change itself chameleon-like from a machine dedicated to one particular task into a machine dedicated to a completely different task from word processor as a calculator to chess opponent so there was no need to change the wiring you just typed the instructions straight in on the tape I guess nowadays when almost everyone owns a physical realization of a universal Turing machine Turing's idea of a one-stop shop computing machine is liable to seem as obvious as the wheel or the arch but in 1936 when engineers thought in terms of building different machines for different tasks Turing's idea of a single machine that could see off all information processing tasks was revolutionary startling it wasn't until after the Second World War that Turing began to actually construct a universal Turing machine and it was here at Bletchley Park that Turing first learned of the technology that he would need in order to implement his world-changing idea the universal stored-program computer and electronics was the way to do it as he learned while he was breaking codes at this British country house where he took up residence on the first full day of the war in 1939 Neville Chamberlain declared war around midday on Germany one day and the next day during showed up at Bletchley Park and at that time he had of course a lot more on his mind than just the universal Turing machine he was working on the German Enigma codes and had been working on them for several months before war broke out there were various flavors of enigma of course airforce enigma naval enigma army enigma railway enigma SS enigma they all had their different ways of using what was fundamentally the same machine and during 1940 the first full year of the war the flavor of enigma used by the German Air Force was being read in large quantities at Bletchley Park which contributed greatly to the air battles for example decrypted enigma messages gave away the Knightly targets of the Bombers during the Blitz because the Bombers were targeted by beams they flew to the confluence of two beams and enigma messages would be sent out to the beam aiming stations on mainland Europe telling them what angles to point their beams at that night Bletchley Park was reading those messages and so the RAF knew where the Bombers were heading and were able to blast the enemy planes out of the sky cheering played an early role in breaking airforce enigma he broke a version of Air Force in dogma that they called blue of Bletchley Park they began by using the the colors to name the various indica Maki's blue green yellow red and so on they pretty soon ran out of colors and by the end of the war there were thousands of keys they went through names of flowers and names of fearsome practically every systematic naming system that exists in our languages but that was airforce enigma the much more secure form of enigma that was used by the North Atlantic German u-boats remained unbroken so let me take you back to those dark days in 1940 and 1941 when Turing on his colleagues in his unit which was known simply as hot eight were pitted against u-boat enigma the basic situation was simple all the materials that Britain needed in order to prosecute the war came from this country and the materials were carried across the North Atlantic in giant convoys 60 80 maybe even more vessels sailing at once with a few Navy escort vessels and the convoys were preyed on by the North Atlantic u-boats and the North Atlantic u-boats were so successful that by the middle of 1941 Britain was in acute danger of being simply starved into surrender everything was coming across that North Atlantic lifeline munitions petrol food and the u-boats were sending large quantities of it to the bottom of the sea and in fact Churchill's advisors told him in June 1941 that Britain could only hold out for a few more months there just weren't enough stores of essential materials to prosecute the war for any longer but if on the other hand you boat enigma could be broken then the messages from the u-boats to their bases in Europe would reveal their positions and so the convoys could very easily be routed round the u-boats the u-boats would be unable to follow them in the North Atlantic wastes just so long as the convoys knew the positions of the u-boats so that was Turing's problem and it was in June 1941 that same month that sure she'll heard from his voice advisers that the end was nigh it was in June 1941 that cheering first broke into the daily u-boat message he'd been working on the problem since September 1939 so that's about 20 months to crack u-boat enigma so I guess that's some kind of measure of how difficult that particular code was to crack the intelligence that Turing and his colleagues in Hutto were able to supply to the Admiralty had an immediate effect on the North Atlantic position the convoy we routing x' that were made on the basis of Turing's intelligence were so successful that during the first twenty three days of June not a single convoy was cited by the u-boats I think that's a remarkable statistic no sightings for 23 days you would think that the Germans would immediately suspect that their codes were being read how else could this sudden change be explained but no it seems they didn't suspect and Bletchley Park and British intelligence went to extreme lengths in order to conceal the fact that the code breakers had managed to get into u-boat enigma British intelligence kind of distributed a fairy story via German prisoners and allowed them allow the prisoners to write letters home about this very story and so on there were various mechanisms but the essence of the fairy story was that us a super radar had been invented a particularly powerful version of radar that could detect enemy submarines hundreds of miles away even when they were submerged below the surface it was a complete cock-and-bull story of course but nevertheless the cover-up was very successful Turing was assisted in his battle against u-boat enigma by the Royal Navy who were able to capture enigma materials from German vessels and this is one such vessel it's the u11 Oh which was shadowing a convoy south of Iceland for several days and then around midday one day the u11 Oh was submerged launched off two torpedoes in quick succession and within minutes of each other to merchant vessels the Navy in particular the British Corvette or Brisa were able to get a fix on the position of the u1 100 by what they called a Tzadik in those days of what we now call so now and the Aubrey show steamed to the position where the u1 were know was believed to be and dropped 10 depth charges overhead and the effect of these depth charges was to blow the u1 102 the surface it was a very unusual situation of course normally when you successfully depth charged a u-boat then it would sink without trace and I've heard what the the naval officer who led the subsequent boarding party from one Boresha on to the u1 one oh he said it was the dream of every young naval officer to see a u-boat blown to the surface during a successful depth charge raid this is the most dramatic picture I've found in connection with the battle against enigma I found it completely by accident in the archives of the Imperial War Museum in London I was looking for something completely different and I opened an envelope and inside was this and there were enough notes to enable me to figure out what it was it's the survivors from the u1 were know the German survivors who was as soon as the ship burst to the surface threw themselves into the water to save themselves so as they expected the ship to go down any moment of course and also according to the official British report they were hastened on their way by small arms fire and anyway here they are floating in the middle of the Atlantic just south of Iceland and just a few minutes before they were hauled on board a Boresha to be given steaming mugs of hot British tea possibly laced with rum totters I don't know I know what they would have done in those days but they the the German sailors were treated very well captured German sailors there were many instances where there are you know Diaries or you know what aren't accounts that survived and explain how German sailors were treated well and German materials you know died of captured sailors and so on express great surprise that they were treated so well their belief was that they would most likely be shot if they were captured but instead they were welcomed and given blankets and so on I read one diary which I found rather chilling this is from a captured German officer and you know he'd gone through this treatment and he said I wonder how it would have been had the boot been on the other foot he said it in German of course but that's what he said I found that an interesting thought but anyway a boarding party from the or Boresha quickly went on board the u11 oh and they kept they found an Enigma machine still plugged in they'd been in the middle of sending a message when they were depth-charged so sub leftenant david bohm and his boys grabbed the Enigma machine grabbed any other code books that they could find and they carried it all back to the or Boresha together with various other items of of booty there were German charts of German minefields and all sorts of highly useful bits and pieces the captain of the u11 Oh Julius Lemp was under the strictest of instructions to destroy all enigma materials if the ship was in danger of being boarded and he didn't he just left it although he must have been very certain that his u-boat was going to go down within seconds the thought in his mind must simply have been to get his men off into the sea and then the ship would be going down straightaway but it worked out quite differently and so the Enigma materials were of course carried back to Bletchley Park so this is May 1941 just a few weeks before Turing broke into the u-boat enigma this machine the bombe was of great help to Turing in breaking u-boat enigma and eventually in breaking all the flavors of enigma Turing invented the bomb in the autumn of 1939 and then the first prototype version was which they called simply Victor was delivered to Bletchley Park in the spring of 1940 and initially it with this this one bomb was reserved for use by Turing and his colleagues in hot 8 for their attack on naval enigma the bomb was about seven feet high you may just be able to make out some interesting features on this picture there are some swastikas up here on the top bar that some someone has doodled on and there's a notice here that you can just about make out a handwritten notice which says keep feet off the bomb searched for the crucial settings of the Enigma machine before the Enigma operator sent a message they would set the Machine up in various ways there were various things they could do change positions and a plug board and change the positions of wheels and so on in order to break a message the code breakers needed to know how the operator had set the Machine up at the beginning of the message and that was the information that the bomb would reveal it's searched through possible enigma settings each of these cylinders mimics one of the three wheels of the Enigma machine and so when the bomb was switched on these cylinders span rapidly for a while and then the bomb would suddenly stop and when it stopped its guess at the message start positions of the enigmas three wheels were on these three drums here the so-called indicator drums the Enigma machine had three wheels before the operator sent a particular message the wheels would be put into a particular position and the bomb could search for and guess at what the positions of those three wheels were and when the bomb stopped those positions would appear here on these three indicator drums the Enigma also had a plug board which the operator could manipulate change the settings of the plugs the plug board the bombs guess that the plug board settings could be read off from other relay equipment here on the side of the machine that you can't really see in the picture the bottom didn't always get it right when it stopped they would test those settings either on an enigma machine or a replica enigma machine they do it by hand and if the machine managed to turn the coded message into German or was well if it didn't then they press the start button on the bombe again and searched some more the bomb didn't search exhaustively that would have taken far too long to search through all the possible positions of the Enigma machine that just couldn't be done in useful amounts of time the setting space was too large it's like chess you've probably all heard the thing about if a computer were programmed to play chess by exhaustive search by examining all the possible moves that were available then it would take thousands of millions of years to play a game right out no hope in playing chess that way and it's exactly the same with enigma an exhaustive search would have taken too long have been useful the war would have already been concluded before the machine before the bomb broke its first message so what the bomb did was speed things up by using heuristic search I was astonished and extremely pleased when I first discovered that the bomb used heuristic search I found this out quite by accident again as a result of reading declassified documents about how the bomb worked in particular Turing's explanations of how the bomb worked and areas saying that the bomb used heuristic search he doesn't use those words they were invented later of course but there was the concept and as far as I know Turing's bombe was the first manifestation of what we now call heuristic search this the the the next few slides are just for people who don't know what heuristics are I'm sure probably everyone in this room knows heuristics are but just in case a heuristic is a rule that cuts down the amount of searching required and a heuristic is not necessarily guaranteed to produce the correct solution but it works often enough to be useful the bomb didn't always manage to break or break a message but it would break enough messages in order to be useful the hearest the Bob's heuristics did not always work so for example one heuristic that everyone uses from time to time as if you've lost something search in the vicinity of the place where you think you dropped it so if you lose your keys your keys could be anywhere in the universe but you can speed up the search by starting to look at the place where you thought you might have dropped them this you're a stick wide always work but it works often enough to be useful exactly the same with the bomb sorry so once Turing via the bomb had glimpsed the possibility of achieving machine intelligence by means of heuristic search he was hooked that was it for the rest of his life his code-breaking colleagues at Bletchley Park can remember him in his spare time you know he after a hard day's code-breaking this is how Turing would entertain himself in the evenings he'd think about his new idea of heuristic problem-solving and his colleagues can remember him circulating a typescript at Bletchley Park concerning machine intelligence and the possibilities of heuristic search this paper is now lost unfortunately but it was undoubtedly the first paper in the field of artificial intelligence not of course that the name artificial intelligence was around at that time that didn't arrive until 1956 more than a decade later in Dartmouth College with the Shannon McCarthy summer research project on artificial intelligence cheering's earlier name was machine intelligence and it was machine intelligence that he was investigating of Alexei Park and as I view the story the the bomb Turing's bombe was the first milestone on the road to modern AI if you could tell them roughly how many Enigma machines were made and how many touring Boswell's philosophy these were not 1 z 2 Z's these were massive numbers of machines they were not 1 c2 xor's you mentioned Colossus I'll get onto that later bombs by the end of the war at Bletchley Park and in the out stations around Bletchley Park they had just over 200 puring bombs operating and there were a further I believe 140 bombs in this country basically Turing bombs but changed up a bit by the by a brilliant engineer by the name of - who worked at Dayton Ohio and he designed an electronic version of Turing's bombe there were electronic versions of Turing's bombe designed by Brits as well there were various different types but dashes electronic bombs were just far superior to any of the electronic bombs they built in Britain much more reliable much faster Turing's original ball was not electronic and it was good as long as the Enigma machines only had three wheels but towards the end were later in the war a fourth wheel was added to the Enigma machine which increased the search space considerably and you really needed electronic speeds in order to search through that larger space so that was the drive to build electronic bombs the well could I just say the the the Washington bombs they were largely controlled by Bletchley Park they had such good telecommunications between Bletchley Park and this country that I've heard one of the Bletchley Park guys saying this they were able to use the Washington bombs as easily as if they'd been in an outstation just a few miles down the road from Bletchley Park which i think is quite remarkable sorry good there was another set of three wheels on the top and something on the bottom you only talk about the middle part was that another machine no that was all one bomb so the bomb has lots of cylinders they were all mimicking the three wheels of the one in the coma machine but you know at different positions and with different listen that probably not the the evidence is not clear on this point but probably not well the way of operation would be that you needed a crit in order for the bomb to run a crib I was cryptographers guess a part of the message may be a standard opening or maybe Heil Hitler at the end or something like that so they guessed that there were the these words somewhere in the message and then you would wire the bomb up to search for that particular crib to find a way of transforming the ciphertext into the letters of the German crib if he could do that then you've got the right settings so that's right so that there was a complex organization for kind of farming out the cribs to different bomb groups and also for sending cribs over the Atlantic for the US bombs to desertion Turing said at one point when he was visiting dashes outfit in Dayton he said the principle that the American bombs will use British made cribs is now settled on so and that was how it worked for a while towards you know in the later stages of the war from kind of mid nineteen forty four on Washington was kind of up to speed in the business of finding cribs which was a difficult of art and they kind of took over particularly for the shark u-boat link they took over the business of cribbing from Blish but until that point it was all you know everything was done from Bletchley Park and just kind of mechanically farmed out to the various bomb units in particular over here yeah we're used in Kerala there's an instance of kind of early parallel computing really that you split up the search base based on initial different initial guesses for the same that's no not not really they have one bomb per crib they've got lots of computing unit that's for sure so it's you know kind of parallel in some sense but and then they might sometimes have been running several cribs on several bombs but usually it was one crib per born they had so many messages to break that unless they had a a message that was particularly crucial it's unlikely that they would assign more than one bomb to it they were breaking by the middle of 1943 they were breaking two enigma messages every minute at Bletchley Park so that gives you some idea of the volume so they wouldn't really want to be you know dedicated giving you know several of their limited number of bombs to the same message they just want to keep the throughput rolling the more multi-processing I take it that each bomb was doing parallel processing oh yeah all these copies is over three reels yes yes so though we are artificial intelligence but that's kind of get it well no let me just before I stop talking about that let me say that in 1948 three years after the war ended of course Turing wrote the first manifesto of artificial intelligence he called it simply intelligent machinery so this is eight years before the Shannon McCarthy conference at Dartmouth and in intelligent machinery Turing brilliantly introduced a number of the the core concepts of artificial intelligence some of which was subsequently reinvented by other people for example there was the idea of a genetic algorithm that's their interns 1948 report there's the idea of a logical search there's the idea of a form connectionist learning sadly though Turing never published the 1948 report probably because his employer at that time was devastatingly rude about Turing's report this guy you know was he knew nothing about artificial intelligence he was running an industrial research unit the idea that Turing should be wasting his time on this stuff was just an athame to this guy and he dismissed it as a schoolboys essay and I guess faced with such negativity Turing just tucked it away in his bottom drawer and that was it the world never saw it although two years later yep Durrington quiet yeah Charles Charles Darwin grandson of the real live in the shadow of his grandfather the real John Stoll and he rang and he ran a research laboratory in London called the National Physical Laboratory which Turing joined after the war and that's where Turing built his first computer as I'll go on to explain or at any rate designed and other people built it so that was 1948 and then in 1950 of course Turing did publish a paper on artificial intelligence his famous paper and the philosophy journal mind Computing Machinery and intelligence which everyone knows about nowadays and everyone knows about the Turing test as we call it which is incorporated in that paper so the idea of Turing test is that a judge has to by question and answer decide which of the two contestants is the human being and which one is the computer and the computer cheering says is allowed to perform any tricks in order to evade identification so sensible moves for the computer would be to say no in response to the question are you a computer the computer is allowed to lie and for example if the computer has asked a multi-part multiplied to huge numbers together then the computer should pour for a considerable time and then produce a wrong answer cheering said that if a computer passed the Turing test ie managed to evade identification then he said that some people might say that that shows the computer is thinking he said that it seems better to avoid begging the question and he would call a machine that passed a grade a machine and he said the question of whether computers can pass the Turing test is not quite the same he says as whether computers can think but it seems near enough for our purposes he said Turing is widely reported in the secondary literature as having predicted that the Turing test would be passed by the end of last century and he's got some stick you know since his projection as people have been saying turned out to be false but I don't think that's fair at all because that isn't what he predicted he actually predicted something quite different this is what he said in a conversation with his friend Max Neumann in 1952 Neumann said I should like to be there when your match between a man and a machine takes place and perhaps to try my hand at making up some of the questions but that will be a long time from now if the machine is to stand any chance with no questions barred and Turing says oh yes at least a hundred years I should say so that prediction has got forty years still to run but this is to get well ahead of the code breaking story as a code breaker Turing's next task was to break the so-called Tunney machine the machine that kind of came along after enigma was in many ways more important than enigma people sometimes call it Hitler's blackberry it was a machine designed by the scientists of the Third Reich specifically for use by the German military and here it is once inside its case and once outside so on the right hand side you can see the the wheels of the machine as with enigma the tönnies wheels were the heart of the action and there they are twelve of them standing side by side like plates in a dish rack the Enigma as I said had three wheels and then later versions had four wheels so that gives you an idea of the the additional complexity that there was in the Tunney machine twelve wheels yeah how many of these were manufactured Oh Lots exact numbers I don't know but thousands many were many were captured after the wars that as the Russian armies swept across Europe two hundred thousand enigmas remain so that's plausible yeah so a few thousand of these yes several phones I'm not sure that it would be possible to find out exactly now but there were there were about thirty to thirty-five tiny radio links across Europe when the network was at its height and they'd have probably four Tunney's at each end of those links and then there were large fixed exchanges in various places Berlin Koenigsberg for the eastern links through Russia Paris and so on so they'd have been you know kind of more tunnels at those places so you can form a rough estimate of how many there would have been from the number of links that there were tony was much more sophisticated than enigma enigma' was old technology by the time war broke out it was marketed commercially in the early 1920s and the German military added a few bells and whistles to it during the 30 has increased its security but it was no longer cutting edge technology by 1930 it was slow and it was clumsy to use it required the remarkably large number of six people to in cipher send and decrypt a single message you needed three people at each end of the radio link there'd be the cipher Clark who would type the the map the German message in plain German or the keyboard of the Enigma machine and then each time the cipher Clark pressed a key a different letter would light up in the so called lamp board of the Enigma machine which was an arrangement of 26 torch bulbs each one of which had a stenciled letter pasted over it so as the cipher clock typed the plaintext the keyboard successive letters would light up at the light board Lam thought they were the letters of the ciphertext and the Clarks assistant painstakingly noted those letters down as they lit up and then once the message was completely in ciphered the assistant would pass the sheet of paper with the ciphertext on to the third play of the radio operator who would then convert the ciphertext into Morse code and transmit it to the other end of the link and then at the far end of the link you needed another three operators in order to reverse the whole process with Tunney you only needed one Operator at the end of each link the Tunney machine was attached to a teleprinter keyboard the single operator at the sending link typed the plaintext at the keyboard and from then on everything was automatic the Tunney in ciphered the message automatically Morse code wasn't used at all the encrypted output of the Tunney machine went directly to air and then at the receiving end everything was automatic also under normal operating conditions neither the sending operator nor the receiving operator would ever even see the cipher text as far as they were concerned it was just plain German that when Xion and plain German that came out so how did it work well at the 12 wheels of the Tunney machine were divided into two groups two groups of five and then another group of two the so called motor wheels sitting in the middle there and and they were the the Sai wheels as they were called at Bletchley Park on the left and then the five Chi wheels on the right the motor wheels had a fairly humdrum function they just determined whether other code wheels would move or not it was the Chi wheels and the Sai wheels that did the main business and the way tani worked was essentially to take the plaintext and encrypts it by adding to it to successive layers of encryption adding two layers of obscuring letters to produce the ciphertext so if the German operator typed P for example then the teleprinter keyboard would produce a stream of ones and zeros the tele print code for the letter P teleport code always came in bunches of five the telephone code for the letter P would be some combination of five zeroes and ones and so on for all the other keyboard characters so the the pulses and no pulses from the P would flow into the adder and then the Chi wheels would produce the first obscuring letter M say the reason there are five Chi wheels was because one produced one wheel produced each bit in the teleprinter encryption of the letter so five bits so you needed five wheels so the M was added so P to produce R then R went into the second adder and the second obscuring letter from the Desai wheels was added to it addition was just exclusive disjunction essentially and so the first letter of ciphertext would pop out of the adder and that would go down the link by radio and so on for the rest of the ciphertext letters that were generated the Allies knew virtually nothing about how Tunney worked they could tell that it was a teleprinter no tiny machines had been captured as we saw in nickimja machines were captured so how they works could be determined by the engineers and by taking the machines apart but no tiny machines were captured until the end of the war in fact but amazingly this man bill Tut managed to figure out the entire logical structure of the Tunney machine without ever having set his eyes on one and even more amazingly he managed to do this from just two intercepted messages each about 4,000 characters long so he reverse engineered the Tunney machine on the basis of just two measly intercepts it was the most remarkable feat of crypt analysis during the whole of the second world war perhaps ever I don't know we don't really know what they've done since but tuss tax break must be up there they were adept yes they were there but there were plenty of depths around at that time the Germans hadn't really wised up to how dangerous depths were they dried up later but they made horrendous errors during the first months of Tunney there were loads of depths and also they sent the the 12 wheel settings unencrypted can you believe it they just use name so they were you know Anton was the AE position of the first wheel and Bertha was the B position of the second wheel and so on so each tiny message was just preceded by a string of names Gustav Dora as interim it was a complete gift they stopped that after about 10 months presumably they realized how stupid it was but by then the damage was done and the British were well into Tunney so that's what Tut did his was the first break-in he reverse engineered the Machine top was a humble quiet kind of guy and he's never really got his dues people don't know much about built art even people who worked on honey at Bletchley Park who were actually breaking the messages thanks to what Tom did even they were not told that it was touched who'd broken the machine because of the need-to-know principle they didn't need to know who'd broken the machine and so they weren't told and so tucked kind of lapsed into obscurity it's one of the sad stories of the Black Sea Park operation I think there are there are two things that you need to do in order to get into a new cipher system break the machine which TAS have done but just knowing the structure of the machine isn't enough to enable you to read the daily messages to do that you need to know the settings of the machine that the operator used when he transmitted the message so that second step remained to be done and it was Turing who took that second step I think this isn't even today this isn't terribly well unknown because the Tunney materials were were remained classified for a long time and it wasn't until 2000 the GCHQ finally let them go and the general report on taani contained details of Turing's break into Tunney so what Turing did was devised the first systematic method for breaking into the daily Tunney traffic and it was based on a process that he invented that he called Delta ring I guess it's called differencing nowadays Delta ring is kind of a process of sideways addition so if you have a stream consisting of four letters ABCD then to Delta those four characters you add a to B B to C and C to D to produce C cast three characters those three characters are the Delta of the original four characters that was Turing's invention you could be forgiven for thinking that this process of sideways addition would just make things worse surely doing that to the encoded message would just scramble it still further but remarkably Turing was able to show that Delta ring in fact would reveal information that was otherwise hidden and that was his route in to breaking the tiny messages Turing's method wasn't carried out by machine he's his method which in the laconic jargon of Bletchley Park was called simply Turing Gori his method was carried out by the code breakers using paper and pencil they attacked each individual message by Turing GUI using paper and pencil built up wasn't too impressed by that there was obviously some rivalry between tat and Turing taht said that cheering's method if you when you're applying Turing's method you need to make use of what you felt in your bones and tuck described Turing's method as more artistic than mathematical taught wanted a purely mathematical method that could be implemented in a machine that would make no use of human insight no use of what you've felt in your bones something that was cut and dried enough to be implemented in a machine and by making some brilliant extensions of Turing's notion of del Turing top was able to come up with such a method in the middle of 1942 but there was a snag tuts method required huge amounts of low-grade calculation so much low-grade calculation that if his method was carried out on paper and pencil as sure and guru was then it would take hundreds maybe even thousands of years for a human code breaker to break a single message using Tut's method it was a bit of an obstacle but one day Tut was in his shy quiet way he was explaining his method to this man max Newman who we met earlier in that dialogue with Turing Newman was one of Turing's teachers at Cambridge before the war it was in one of Newman's lecturers by Newman's account anyway that he first gave curing the idea that led to urine to the universal Turing machine and then in 1942 newman left cambridge joined Bletchley Park and through his considerable energies into the attack on Tunney as soon as Tut had explained his method Newman said let's use electronic counters in order to do the arithmetic it was brilliant idea Newman had seen electronic counters at work before the war in the Cavendish laboratory in Cambridge where they were using them to count radioactive emissions and in a flash of insight he saw that this technology could be transferred to the completely different problem of cracking Tunney Newman was put in charge of building an appropriate machine and a few months later it was installed at Bletchley Park it was soon called the Heath Robinson Heath Robinson was a in those days famous British cartoonist who drew overly ingenious contrivances I guess Rube Goldberg Rube Goldberg would be the nearest equivalent in this country the Heath Robinson as soon as it was switched on smoke rose and the engineers had to dive in and cool things down I don't actually have a picture of Heath Robinson I don't believe that any photograph of it survives this is a machine it was the successor machine to Heath Robinson they came to call it in the end old Robinson Heath Robinson was the prototype and this machine was a sort of smoothed over version they I gather they looked very similar except that old Robinson is not quite as rough and ready an old Robinson as you can see has machined aluminium wheels for example whereas the original it did still smoke though the tapes would get very hot the spinning toads and the the original Heath Robinson had wooden wheels instead of the machined aluminium wheels this tape that you can see this contained the Tunney message that they were going to attack punched into the tape in telephone code Heath Robinson was never really a great success but it did work well enough in order to prove that cheering's Delta rune could be done by Machine and Tut's method could be done by machine but the Robinsons only contained a handful of vacuum tubes a handful of electronic valves and they were mostly based on relays which were probably know about which what kind of pre electronic technology relays are electrically driven switches there is a moving metal rod which opens and closes an electrical circuit relays are clunky they're large they operate slowly they're unreliable because as the rod moves it generates wear and they're slow most of all they're slow for the attack on Tunney something better than relays was required the original bombs were built out of relays and for enigma that was fine but it just wasn't fast enough for Tut's method against honey so enter Tom flowers the brilliant electronic engineer who gave the Bletchley carp who gave the Bletchley Park code breakers the world's first large-scale electronic computer flowers had been working with vacuum tubes or electronic valves as he would have called them the British term he'd been working with them since the early 1930s and he knew their capabilities inside out I guess valves or vacuum tubes or such ancient technology that it's perfectly possible that some people in this room will never have seen one so I thought I should bring along some photographs of the some of the valves that flowers actually used in his computer Colossus so there are four of them and the one on the right is of not a valve at all but a photoelectric cell that they used to read the holes in the punch tape flowers was a telephone engineer he'd been working for the British post office who ran the country's telephone systems since the late 1920s and fairly early in the 1930s he had begun to build massive installations of valves that were to be used in telephone exchanges to control the connections between the exchanges London and Liverpool say and flowers built these installations consisting of three to four thousand electronic valves which for those days was enormous unthinkable unheard of because the belief at that time was that electronic valves were just too reliable to be used in large installations electronic valves contained a hot glowing filament which generated the stream of electrons and those filaments were very fragile so they could easily lead to the sudden death of a valve so if you were just using a dozen couple of dozen valves that would be reliable but if you had several hundreds or several thousands of valves in one installation then it was only going to be a short while before another valve blew and the whole installation would go offline so that was the belief amongst engineers at the time you just couldn't build large installations of valves they would not be sufficiently reliable but flowers knew better flowers had discovered that as long as you turned these large installations on and then never turned them off they would operate even more reliably than relays it was the switching on and switching off that brought about the friability so flowers was able to use valves just ad-lib he said once he simply didn't care how many valves he used he knew there was absolutely no obstacle to using valves in large numbers he told me that at the outbreak of war he was probably the only person in Britain who knew that electronic valves could be used on a large scale for high-speed computing and so when he was summoned to Bletchley Park ironically not because of his knowledge of valves at all but because of his knowledge of relays but when he got there he turned out to be the right man in the right place at the right time whoops he was called in a Turing suggestion in fact to help with the design of the logic unit of the Robinson and the design of the Robinson had been frozen by the time flowers became involved as soon as flowers saw the design of the Robinson he hated it it had only got a couple of dozen valves the rest was relays he could see that it was never going to give the codebreakers the speed and the reliability that needed so flowers made a counterproposal to newman he said i can build you an all-electronic machine consisting of about two thousand valves it will be far more successful than the robinson machine could ever be newman took advice and he heard the standard story two thousand valves never work reliable this man's an idiot so newman was unable to take up flowers on his offer newman gave flowers a certain amount of encouragement he said two flowers look if you think you can do it you build it but flowers was left without the support of Bletchley Park and he just quietly went back to his own laboratory in London and got on with building the all electronic machine that he knew the code breakers needed he took ten months to build the first large-scale electronic computer the first colossus flower said that he and his team worked until their eyes drops out night and day in order to get colossus finished and then one day early in 1944 summer flowers lads turned up at Bletchley Park with the first large-scale electronic computer on the back of a lorry and here it is flowers Colossus the name Colossus was certainly apt it was as big as a room and it weighed about a ton you can see the message tape here over on the right similar to the message tape in the Robinson flowers machine caused quite a stir when it arrived at Bletchley Park the code breakers simply couldn't believe the speed of the machine and they were also very surprised to find according to the wartime accounts that when Colossus was set the same problem twice it will produce the same answer twice so you can infer that the Robinson did not always do this so d-day of course June 1944 the first colossus and also Turing's bombs had been giving the Allies an unparalleled window on the Germans preparations for the invasion they knew the locations of every German division thanks to Turing and tart fundamentally and then by the end of the war there were ten colossi in operation in Newman's unit at Bletchley Park which was called simply the new Maneri these machines were housed in two vast steel-framed buildings and they worked around the clock ten enormous computers working around the clock it was the world's first electronic computing facility and it had things that we associate with the sixties and seventies rather than the 1940s job queues teams of operators working in shifts specialized tape punching crews engineers all was on hand to keep the machinery running smoothly it wasn't really until the 1960s that anything like this was seen again when the first modern large-scale computing centers started to come into existence but although flowers had built colossus nobody told the historians colossus was of course kept under the closest traps until the 1990s really some photographs were released in the 1970s but it was the 1990s before new details came out sorry where Brian Randall wrote his famous paper yes flowers bells was there yes but very little of the information a very little information about how colossus worked was released at that time and flowers rotor during flowers talking he wrote thee the truth table of the X XOR function on the board another man got up quickly and erased it but Pisa Hilson one of the one of the tiny breakers at Bletchley Park he told me that he was sent a letter by GCHQ GCHQ forbidding him to tell people that the wheels of the tiny machine were called Chi wheels and Sai wheels so they were very you know kind of capricious about the parts of the information that they would release in the parts of the information that they wouldn't release that los alamos conference the the mere skeleton emerged but all the details really were were still hidden and flowers wrote up the talk that he gave at los alamos and published it in the annals of the history of computing it came out in the early eighties if any of you have looked at it it's completely incomprehensible because flowers was never allowed to say what Colossus did and it wasn't until the the D classifications in the late 1990s and then most particularly in 2000 when General report on Tunney was finally released that the world was able to see exactly what Colossus at Donnan exactly how tani had worked and exactly what a contribution had been made to the war by these various characters that I've been talking about why did they hold on to the documents for so long very good question yu ting yes and no yes and no let me pick up the story with the ENIAC if you if you look at price if you go to the library and I'm not quite sure what time I'll stop but I'm just going to keep on talking until someone stands up and tells me to stop sorry if you if you go to the library and Paul practically any book on the history of computing off the shelves particularly ones from you know maybe 10 years ago and further back you'll find it saying I'm sure that the first electronic computer was the ENIAC and if you look at any computer science text that contains a few pages about the history of the subject you will read that Eckert and Mauchly built the first large-scale electronic computer at the Moore School in Philadelphia it was called the ENIAC and if first ran in December 1946 a full two years after the first Colossus around but that isn't mentioned in the standard histories the ENIAC was in some ways very similar to Colossus it was a a special purpose machine it was dedicated to a particular task they did eventually figure out how to turn it into a a rather more general-purpose machine but it was originally designed by Eckert and Mauchly as a special purpose machine for calculating gunnery tables and like Colossus it didn't incorporate Turing stored program concept in order to set the ENIAC up for a new job you the operators needed to monkey around with cables and switches they need to submit early rewire the Machine re-routing cables and throwing switches it could take a day to reprogram the machine if things went badly it could take as long as a week to change any acts program it was a nightmare it sounds just so primitive from a modern perspective where we simply take Turing stored-program world for granted but in those days nobody had yet built a stored-program machine flowers was quite bitter towards the end of his life about the fact that the ENIAC had got all the glory and his work was unknown and it was caught of course the the crippling secrecy that the British for whatever reasons imposed on Colossus that resulted for flowers being written out of the history of computing for all those decodes flowers said when after the war ended I was told that the secret of Colossus was to be kept indefinitely I was naturally disappointed I was in no doubt once it was a proven success that Colossus was an historic breakthrough and that publication would have made my name in scientific and engineering circles a conviction confirmed by the reception accorded to ENIAC the u.s. equivalent made public just after the war ended I had to endure all the acclaim given to that enterprise without being able to disclose that I had anticipated it it's one of the saddest stories I think in the history of computing I mean there is plenty of credit to go around and it's a shame that not enough of it went flowers why nothing nothing there was no need for him to know he talks during but Turing would never Turing was very security conscious Turing would never have told someone about Colossus unless he had been cleared by the Bletchley Park authorities to say and even after the war when they were not allowed to talk about it I can't imagine cheering ever telling for Norman about Colossus he might have talked generally about you know high-speed computing with thousands of valves but to talk about Colossus Tunney exactly what they'd done know very few people knew about Colossus in the years after the war trigger I would think one line that could probably guess most anything in the universe it's possible yeah I don't know what glasses might hit but I had the feeling that it wouldn't a multiplier yeah flowers gave the codebreakers exactly what they needed to do the job in hand you know resources were short time was short he didn't even include the stored program concept he knew about it and Newman gave him Turing's 1936 paper to read wild flowers was designed in colossus but the pressures were so great that he just couldn't consider trying to solve that technical problem so he he didn't even give them multiplication let alone as you say differential equations but of course immediately after the war if instead of breaking up eight of the colossi and taking the other two off in secrecy to GCHQ the new headquarters in London if instead of doing that Colossus had been made public turned over to public science then turing newman flowers would very quickly have had the colossi solving differential equations running store programs and there would have been this massive high-speed electronic computing facility in existence in the middle of 1945 publicly known clones of it would quickly have sprung up around the globe who knows what would have happened maybe the internet much earlier than it actually happened social networking decades earlier perhaps and all the social consequences that would have flowed from that it's boggling to imagine I set up a commercial company to make the thing after the war it was secret he'd be margin it so it so that they actually unlikely thing was was also secret I mean you see you you say you know I thought us alright I'm gonna do something completely different well I'm going to exploit the technology that I well I mean he was you know he had hardline security trees telling him that he was not to use any of the staff was not to mention any no but well you know maybe flowers have been a bit more hard-nosed things that have gone differently but also flowers was never well then he built the first large-scale electronic computer computing was not at the forefront of his interest he was a telephone engineer what he longed to do was to build electronic telephone exchanges get rid of all those nasty slow relays and replace it with digital electronic equipment and so that's what he devoted his life to and he and he was successful he were he you know was a major player in the first European electronic exchange at Highgate but eat even there there are there are sad stories because flowers couldn't tell his his colleagues and the post office what he'd done during the war they tended to take no notice of him and he told me once that he got a reputation for precociousness because he'd always be saying do it this way do it that way and they would think you know who is this guy what does he know whereas if you know even some aspects of his work had been de Klerk disclosed earlier his career would have gone quite differently and presumably they could have done that much they could have done enough to have enabled flowers to build a computer had he wanted to to have risen to the highest levels of the British post office had he wanted to and they could have eased his way to building the first or electronic telephone exchange by releasing information that did no damage at all to the core staff that they wanted to keep secret but they chose not to they were using it it's simple GCHQ does not disclose you know machinery that it is relying on the tiny machines were captured by the Russians as they moved across Europe and you know the language changed from German to Russian but then the message is still flowed and GCHQ were breaking them using to colossi they destroyed eight and another one in flowers Factory because that was far too much capacity but they thought two would be enough and they they soon combined they built a new machine that combined Colossus and Robinson they called it the collar Rob there were features of Robinson that were kind of more flexible than Colossus and so they took the best of Colossus and the best of Robinson and they built their collar odds which they used until you know general-purpose store program machines were available cryptography until 1990 amazing is enough yeah and the East German Stasi were using enigma until I don't know well into the 60s the Washington bombs were kept in mothballs for a long time and they were brought they were brought out and readied for use in the 1950s don't quite know why but there was something going on a chicane was like the iliac yeah some cables you know time's running out could you finish your story with our IDs questions tell these guys to shut up let me tell you what happened after the war flowers ideas lived on via Turing and Newman although the rest of the world didn't know about Colossus they did and flowers have said that as soon as Turing and Newman saw his racks of electronic equipment they realized that this was the way to build a turing universal Turing machine in hardware a stored program computer so immediately after the war Newman went to the University of Manchester where he became chair of mathematics and he lost no time in setting up his computing machine laboratory and Turing in the meanwhile went to London where he was hired by the National Physical Laboratory in order to build an electronic stored program computer so a kind of race developed between Newman in Manchester and Turing in London this is bushy house part of the campus of the National Physical Laboratory this is where Turing worked designing his automatic computing engine as as it was called the name was kind of a homage to Charles Babbage with his Difference Engine and his analytical engine cheering's stored-program computer was called the automatic computing engine and a pilot model of the automatic computing engine first ran in bushy house in what was previously the the butler's pantry when this was a you know upstairs downstairs kind of establishment and Victorian tones cheering's ideas for the automatic computing entrant caught the attention of the British press and the British public the the u.s. model mentioned here is of course the Aniak and the the journalist was quite right to mention that the ace had a vastly bigger memory than the ENIAC but the the article misses out the most important difference namely that the the ace was designed to be a stored-program computer whereas the ENIAC wasn't they did manage to get it to run in a kind of read-only stored program mode in a very Khiladi kind of way a few years later but it was never designed as a stored-program machine sadly though Turing lost the race I think it would have been entirely appropriate historically if the ACE had been the first stored-program computer to run but it wasn't Newman had the edge Newman was fortunate in hiring these two extraordinary engineers Tom Kilburn Freddie Williams they'd worked in radar during the war and they were able to get the first stored-program computer working in Manchester in 1948 June 1948 it ran its first programme it was a machine that owed a lot to Alan Turing Williams was completely frank that he and Kilburn as he put it didn't know the first thing about computers when they first entered the field in 1946 and killburn learned how to build a computer through attending a series of lectures that Turing gave in London in 1946 and 1947 lectures on the ACE so Turing made two contributions to two fundamental contributions to the Manchester computer the basic idea was his the idea of a stored-program Universal machine and he also taught Kilburn the essentials of computer architecture cheering went to Manchester as soon as they got the the Manchester baby as they called it working Turing couldn't wait to get his hands on a universal Turing machine in hardware he fled from the National Physical Laboratory leaving his project behind him left his colleagues to build the pilot model of the automatic computing engine and here it is at first round in 1950 it was what they built what the engine is built was only a shadow of the machine that Turing had originally envisaged in 1945 this machine had only 5% of the memory that Turing had specified in 1945 but they stayed with cheering's clock speed one megahertz and so the pilot ace although it was physically quite a lot smaller than some of the other pioneering computers it was able to blow every single one of them off the stage because it was so fast thanks to Turing's foresight and his confidence that one megahertz engineering was possible many didn't believe that that could be done at that time you can see that there's there's really not much equipment in Turing's machine not much hardware if you've seen photographs of some of the other other early machines the pilot ace is quite a lean machine by comparison this this is in accordance with Turing's design philosophy which was to minimize Hardware in favor of software after all he'd been writing computer programs since 1936 as far as he was concerned programming was pretty solved he knew that was the easy part of the equation it was the engineering problems that were hard and so keeping the hardware to a minimum and throwing the burden onto the software was the obvious thing to do as far as cheering was concerned so while other people were trying to build hardware multipliers and additional hardware to do floating-point arithmetic cheering did all that by programming we know what Turing's viewers of people who didn't subscribe to his design philosophy from a memo that he wrote around the end of 1946 he said I've read wilks's proposal for a pilot machine Wilkes was a Cambridge engineer who was building a machine that followed the echo more Cleavon Lyman edvac principles edvac being the successor machine to the ENIAC in Philadelphia Turing says the code or the operating system as we would now call it which he suggests is however very contrary to the line of development here and much more in the American tradition of solving ones difficulties by means of much equipment rather than thought a wonderfully ac2 comment by Turing so here is the the deuce the engineered commercially marketed version of the pilot model ice it was one of the major workhorses of the early decades of the the computer age there were many juicers in operation and they didn't I think the last one went out of service in the nineteen seventies so this this is another machine spawned by the ACE design this is the Bendix g15 designed by the American engineer Harry husky and built by the Bendix corporation in Detroit it's the first personal computer somewhat larger than the personal computers that were used to nowadays the Bendix was about the size of a large domestic refrigerator but nevertheless it was a single user machine Huskie worked with cheering at the National Physical Laboratory for a year in 1947 and when he came back here he applied Turing's design philosophy minimizing Hardware throwing the load onto software in order to design a machine that was small enough and cheap enough to be marketed as a single user machine so the idea was that engineers would have a Bendix g15 for their sole use parked by their desk and thanks to Turing's timing Trix husky was able to make the Bendix as fast as mainframe machines many times at saunas so that's pretty much the end of the story I find it very satisfying that there is this direct line of development from the universal Turing machine of 1936 through flowers Colossus to the first stored-program computer at Manchester in 1948 and then on to the deuce which was one as I say was one of the major workhorses of the first years of the Information Age and then the line extends on to to the first person or compete I think that's not a bad little bundle of machines to have flowed from cheering's work in 1936 and of course he gave us the store program world thank you I'm happy to carry on answering questions if anyone wants to ask well I don't have any personal knowledge about you yeah I'd say on Stanford campus the role of Alan Turing and computer science doesn't be very well reprieved well appreciated and computer science computer science department here is much more interested in computer engineering than it is in the foundational questions that answering address that would be my explanation at this point the whole story is on challenging turns oh you're a great author on provocate I'm sorry you had your hand off oh yeah I just wanted to mention something that is important to me that I think is significant but it is it's very little-known just that at Manchester Turing wrote a several hundred page manual on how to use the Manchester and and she chose him to be a well-rounded computer scientist even more I mean just this Mike winter sent me a copy this manual about 30 years ago I tried to fight what I was in Manchester last year and tried to because I won I wanted to check something for lunch I was bidding you nobody there could find it yeah they keep it hidden away it's on the web flow now what it's on the web okay well yo volunteering no but I imagine that when everybody who thinks about drinking about 1936 paper me and know nothing about how how he would he'll say you can debug programs by listening to the speaker that yes yes well that was that was pretty Turing's third contribution to the Manchester confusion you know the basic idea then he taught the engineers how to build it and then he went on got the machine working before he arrived programming was done by hand they had a row of switches and they had to enter the program digit by digit into memory an output the only output facility was there was a TV screen with dots on it and you had to read the output off the screen in that way it was Turing who turned it into a working machine he provided them with input and output and he used the technology that he'd been involved with at Bletchley tape he used punched tape sorry of Bletchley Park he used punch tape technology and he built them a tape reader for for input and various output facilities so the the manchester computer is indebted on all sides to Turek although up in manchester they they try and sideline Turek and it's heartbreaking to hear that they have hidden away his programming manual won't show it to visitors such as yourself yes documentation is really really really fascinating I think the importance of documentation is sort of unappreciated in the computer science world and significant at significant documentation like like Touring's manchester manual or small talk ad manual or just great documentation is is really hugely influential more so than people imagine so I did question your story sort of ended what did Tory actually do at Manchester and and how did he finish his career well he wrote the first artificial intelligence which was a chess playing program called Toro champ he wrote it with David Champ a noun a mathematical economist he was friendly with they they never managed code euro champ for the machine they had they had the program written and they could hand work it it played games it beat human players cheering wanted to code it for the Manchester machine but he and Kilburn never got on Kilburn had built the machine you know that the machine was Turing's idea or and you know Turing must have thought of it as his baby and he in one place you can get a feeling for what his attitude to Kilburn and perhaps also williams might have been from a phrase that he uses not it not in connection with them in particular but he talks about the mechanics who built the machine so Kilburn was just the mechanic but Kilburn had actually wired the thing together with the sweat of his own brow as far as he was concerned it was his machine and there was this guy Turing who do you know arrived late in the piece from London from you know what Kilburn would have seen as a failed project down there Kilburn knew nothing about what's urine had done he you know Kilburn would never read during the 1936 paper he didn't know that all these ideas were were Turing's he knew that he'd been to Turing's lectures and had learnt computer architecture from turing but he tried to you know hush that up in later life he didn't like admitting any indebtedness tutoring and if you asked him well you know how did how did you you know learn how to build a computer back then and you know in 1946 1947 he would say well i don't know somehow at that time i just knew how to build a computer but actually it's no mystery how he found out hearing told him but anyway there was this considerable hostility between the two of them Kilburn just didn't know how to handle a man like touring and touring sorry versa vice versa but anyway Kilburn prevented he restricted your exact cess to the machine he didn't want to run using it particularly and he forbade Turek from implementing his chess playing program i could almost hear Kilburn saying you know what is this nonsense we don't want people writing chess programs on my precious computer he never was particularly favorably disposed to artificial intelligence even in later life I think it's been revived I don't think it was ever I don't think it ever ran on a machine at the time sorry manuscripts of its surviving the source code um no I think not but Turing's no no the the the details of the turo trap program don't survive all that exists is a note written much later by champ a noun and he describes a bit about the program but really he leaves too much too vague but cheering rose a paper on chess programming really the the pioneering paper on chess programming although it was published without Turing's name on it and so it's rarely associated with him but there is much detail in there about how the successor version of the Tarot champ worked and there's there's plenty of enough staff there to be able to recreate Turing's Pro Yosi Universal machine is very similar to girdles a definition of inference in terms of arithmetic in terms of its conceptual apparatus and I wonder if Turing ever met butyl at Princeton and the interactive because if it is in his in his famous paper and when she introduces the Turing machine he has this little sentence and and saying of course what I'm doing here is quite different than what's been done by professor McCrystal before me yes yes well Turing certainly knew something about girdle yes well before though he attended Newman's lectures on the foundations of maths and although the lecture notes for the year that Turing attended have been lost the lecture notes for the preceding year have survived so you know academic life being what it is presumably the course was pretty much the same for the following year and that the course ends with an exposition of girdles and completeness results and as you know a lot of stuff about Hilbert and the enshrined ins problem and so on as well I have seen a letter by Richard Braithwaite who was a fellow at Kings who high table dinner one night managed to get Turing talking about what he was working on in 1935 and Braithwaite kind of said hmm I can see interesting connections with what girdle over in Vienna has proved he didn't get a very good response from cheering cheering didn't really want to know about girdles work you know he liked working alone he liked inventing everything for himself I think it's quite possible he didn't pay too much attention to curdles well but of course girdle was quite the reverse you know he was very interested in what other people were doing he read Turing's 1936 paper and there there are many very complementary quotes about cheering scattered through girdles work and it was Turing's 1936 paper really that enabled girdle to generalize the incompleteness result beyond principia mathematica to you know formal systems in general containing but of arithmetic yes but describing a few photographs you could survive are there any excellent video images like home movies or audio at once no there were audio recordings cheering made a number of radio broadcasts with the BBC but unfortunately the BBC thrilled by the tapes I'd like to line them up and shoot of myself the one hope is that some Commonwealth country that would have been receiving material from London has archived the tapes and over the years I've written to a few likely candidates but I've never managed to get anywhere but I still hope that there might be something but no no it's certainly no video and as far as we know no voice in Australia yes besides Harry husky who's stood alone in your thoughts oh well probably not many people um Harry yes I can't think of anyone else that I mentioned who's still alive sorry they may have some informations but who is this people you mentioned that still alive well yes Harry hustle you know knows a considerable amount there there are a few of the Bletchley Park people who asked a lot Gary Roberts for example who worked with cheering who's still sharp as a razor and full of energy and will be it here's a plug and Gerry Roberts will be giving the annual Turing lecture the King's College Turing lecture at Alan Turing's hundredth birthday party at King's College Cambridge which I'm organising and it will be on the 16th of June just a few days before Turing 100th birth two days many star speakers including sorry yes but if you fancy coming to King's College for Turing 100th birthday party you'll be very welcome we've got Turing's three nieces coming to talk about Turing and also Turing's nephew we'll be talking about his memories of terror and a number of pioneers of computer and various interesting characters yes I was yes well flowers died I think he died in May 1997 or maybe a party oh yes so he knew he knew something about it he missed the declassification of the Tonya report but I'd managed to get hold of a you know kind of a bootleg coffee which I gave so he had seen it and he wrote a couple of chapters for the book that I subsequently brought out on Colossus but his and although he devoted great energy and enthusiasm to those chapters his his whole attitude was it's too late you know if this had been 10 years ago 20 years ago it could have made a difference to me you know he had a quiet pride and was he darling he was glad that the story was just about to break but he knew it was too late for him there's quite a jolly yes I remember him as jolly full of stories yeah very modest had to be faster than a mechanical one but but not that large were they anywhere near comfortable well there were various different types of electronic ball that the cheap and dirty version had an electronic attachment they called it the Cobra because of the great snake of wires that went from the basic curing Bowl to the electronic attachment that just looked at the fourth wheel but dashes bombs were much more integrated devices as to the comparative speeds I don't have the figures at my fingertips for more please visit us at stanford.edu

Channel: Stanford

Views: 157,209

Rating: 4.8395524 out of 5

Keywords: computer science, technology, engineering, machine learning, artificial intelligence, Turing, Heuristic Search, nazi code

Id: p7Lv9GxigYU

Length: 96min 38sec (5798 seconds)

Published: Mon Sep 10 2012