Who had the Idea? The Manchester computer of 1948 has been reconstructed for its fiftieth anniversary on 21 June 1998. There is much information on the Manchester site about how Williams and Kilburn succeeded with their cathode-ray-tube storage and built the machine. Brian Napper, who has written the material for the Manchester site, stresses in his page on the general background that the Manchester machine was built with programs loaded in RAM --- the revolutionary idea that defined the computer. (See this Scrapbook page.) But he doesn't explain how Williams and Kilburn got this idea. It would have been possible for Williams to learn about the stored-program principle in the course of his 1946 work at TRE on the storage mechanism. It was generally in the air after the EDVAC report of 1945; and from Turing's ACE proposal. But what in fact happened, according to Williams (in the Radio and Electronic Engineer, July 1975) was that he learnt the principle from Newman after taking up the Manchester post in December 1946. With this store available, the next step was to build a computer around it. Tom Kilburn and I knew nothing about computers, but a lot about circuits. Professor Newman and Mr A. M. Turing knew a lot about computers and substantially nothing about electronics. They took us by the hand and explained how numbers could live in houses with addresses and how if they did they could be kept track of during a calculation... There is an obvious chicken-and-egg interdependence between logical function and practical engineering. Turing and Newman could not embody their ideas without engineers; the engineers would not have known what to build without the mathematicians' ideas. I feel that the latter aspect is not always given its full weight, but mathematicians are very modest people.

Carry on Computing At Bletchley Park, in building Bombes and the Colossus, the synthesis had been reasonably harmonious but it was not to be so at Manchester. There was a particularly Mancunian culture clash. Alan Turing could not 'direct' anything, but he organised the software which made the engineers' machine work. In 1950 he completed a Programming Manual. An opening segment of the second edition of this Manual is available on the Manchester site in an HTML rendition. A complete version of the first edition will appear later on-line. A preliminary draft can be found now on this MIT site, with an introduction by the editor, Robert S. Thau adding comment on Turing's programming ideas and their context. Turing's assistants on the software writing were women, Cicely Popplewell and Audrey Bates. This set-up neatly confirmed Manchester stereotypes: hard soft engineering mathematics Williams, Kilburn Newman, Turing things concepts north south Real Manchester Virtual Womanchester Has much changed? At least women in computing and gender issues are on the agenda. War Again The British government desperately wanted atomic bombs. They didn't believe that the Americans would retaliate against a Soviet nuclear attack on Britain, and they wanted to prove that Britain was still in the Big Three. The 1946 McMahon Act in the United States meant that Britain, which had given much to the wartime atomic bomb programme, was denied American co-operation thereafter. In late 1948, the Cold War began in earnest, and it became a British national priority to have computing facilities for the atomic bomb implosion calculation. A lavish new contract was rushed through to allow Ferranti to build a full-scale machine, the Ferranti Mark 1. The contract specified merely that it would be built to Williams's design. Newman's priorities for pure mathematics and science were forgotten.

Fifty Years of Hardware The prototype computer looked like this: and there are many more on the pictures on the Manchester site, which also includes a history of Manchester computers from 1948 to 1975. There are also more photographs maintained by Tommy Thomas, one of the engineers of the original project.

Parallel machines At this period the main rival to the Manchester computer development was the EDSAC computer at Cambridge, England, inaugurated in September 1949. This was the work of Maurice Wilkes, who took as his starting-point the American EDVAC proposals, but was able to beat the Americans at their own game. Turing gave a talk at its inauguration which anticipated later ideas of program proof, but, in typical disregard for his own reputation, made nothing of it. There is a description and simulator of the EDSAC by Martin Campbell-Kelly. A version of Turing's ACE design was built at the National Physical Laboratory after all. The Pilot ACE was inaugurated in November 1950. It is now in the Science Museum, London; which has a picture of it in its virtual gallery of Treasures.

Using the world's first computer F. C. Williams himself had no interest in the use of the machine he had built. Speaking in an oral history of Pioneers of Computing, Science Museum, 1976, he said: Well let's be clear right from the start, I never have been interested in computing, and I'm still not interested in computing. What I'm interested in is computers. I'm an engineer, I define the computer right from square one as a device which was designed to facilitate the performance of mathematics, the greater part of which would be very much better not done, and I've never changed that view really... Users were seen as rather a nuisance while the machine was in development, but Newman immediately found a genuine mathematical problem that could be run on the prototype Manchester computer, and thereby rescued a little of the originally intended function for the machine in pure mathematics. This was the problem of finding Mersenne primes. At that time the largest known prime was 2127 - 1, and had been so since 1876, when Lucas discovered a test for primality of numbers of this type, a test which was extremely well suited to a computer. They ran a program successfully, and then Turing coded a faster version of it, but even so did not discover the next prime, which was out of range at 2521 - 1, and was found only in 1952. The largest known prime now is again a Mersenne prime, and found by exactly the same method, only on a somewhat larger and faster computer). Luke Welsh's Mersenne page More on primes and factorisation More on recently found large primes You are welcome to use my simple Java applet for factorisation and so check the primality of 261– 1 = 2305843009213693951. The 1949 programme gained newspaper publicity for the Manchester computer, although (or because) readers of the day would have assumed prime numbers to be the epitome of pure mathematical uselessness. Nowadays these investigations are seen rather differently because of the connection between large primes and cryptographic security. As usual the mathematicians were ahead of their time. It wasn't that either Newman or Turing had a particular research interest in Mersenne primes; the problem was chosen as one which could show off the power of the computer. But in 1950 Turing used the prototype computer for a problem which derived from his pre-war research work on the Riemann Zeta-function, also associated with the properties of prime numbers. This was probably the first serious use of a computer for research in mathematics. For Turing it also illustrated very neatly the power of the universal machine concept, as it performed the work for which he had designed a special-purpose calculator in 1939, now made completely redundant.

Intelligent Machinery comes out of the Closet Although Newman made very careful statements to the press, Turing as usual announced rather incautiously that what they were interested in at Manchester was the extent to which a machine could think for itself. This provoked an immediate response from a Manchester brain surgeon, Geoffrey Jefferson, in 1949: a lecture with the title 'No Mind for Mechanical Man.' Alan Turing was stimulated by the public controversy to write a definitive paper on his views for the prospects for Artificial Intelligence, or as he called it, Intelligent Machinery. This paper, introducing the idea of the Turing Test, was his only paper on the subject to be effectively published. He took the opportunity to air a kind of wit which could hardly have been more different from the heavy macho engineering ambience at Manchester, and in which male and female role-playing enjoyed a curious part. You will find this paper described on another Scrapbook page. He also gave a talk on the BBC radio Third Programme: although the producer had strong doubts about his talents as a media performer.