The Origin and Pioneers of Digital Data Processing

We are concerned here with what has become the definition of a (digital) computer in an attempt to distinguish between computers and calculators i.e. a computer with the capability and capacity to operate on data according to any set of instructions (program) loaded into it. It is appropriate to consider the level of 'Turing Completeness' in terms of conditional processing, and the self-determination of an end-point in the processing cycle.

All the information below has been obtained from official archives - the majority has been doubled-checked by other documents and reports authored by the named individuals. The popular (American) version of computer history has been highly scrutinised. In the final analysis, whilst the concept of digital data processing was defined in rarified air of academia, it took a military need coupled with some extraordinarily talented and pragmatic Englishmen to evolve into something tangible:-

Alan Turing conceived the computer (as defined above) in 1936 whilst at Cambridge University.

Following his election as a Fellow of King's College and attendance of lectures by Max Newman, he became interested in the application of logic and produce a paper: On computable numbers, with an application to the Entscheidungsproblem, which was circulated throughout academia in the UK and USA. The Entscheidungsproblem asks for an algorithm that will take as input a description of a formal language and a mathematical statement in the language and produce as output either "True" or "False" according to whether the statement is true or false. The algorithm need not justify its answer, nor provide a proof, so long as it is always correct. To answer such a question required a definition of method which would be not only precise but compelling. This is what Turing supplied. He analysed what could be achieved by a person performing a methodical process, and seizing on the idea of something performed mechanically, expressed the analysis in terms of a theoretical machine able to perform certain precisely defined elementary operations described as symbols. He presented convincing arguments that the scope of such a machine was sufficient to encompass everything that constituted a definite method. His virtually unrivalled capacity for lateral parallel thinking led him to include an argument based on the transitions between 'states of mind' of a human being performing a mental process. Whilst the bulk of Alan Turing's paper promotes an abstract (binary-based) mechanism for computing numbers involving interlaced order code and data, section 6 entitled: The universal computing machine defines the concept of general purpose computing on which all subsequent stored-program digital computers are modelled.
At wartime Bletchley Park he made numerous fundamental contributions to cryptanalysis or code-breaking. One such was the essential (logical) design for the British BOMBE, which with an additional element (the Diagonal Board) designed by Gordon Welchman, resulted in successful deductions of some of the daily settings for the Enigma machines. Principally those used for the various German military encrypted communications networks from 1941 and 1943 onwards for the 3 and 4 rotor machines, respectively.
Postwar, whilst employed at the National Physical Laboratory, he presented a paper in February 1946, which was the first detailed design of a stored-program computer.


Tommy Flowers of the Post Office Research Station at Dollis Hill produced the digital electronic (hardware) design in 1943.

His first involvement with Bletchley Park (BP) was an investigation into an electronic version of the electro-mechanical BOMBE developed as an aid to deciphering messages encrypted using the ENIGMA machine. An electronic BOMBE didn't happen, but Tommy Flowers (on Alan Turing's recommendation) produced an electronic version of the counting elements of Heath Robinson, for Max Newman - see below, BP's first attempt at an automated (electro-mechanical) aid for the process of deciphering FISH traffic encrypted using the LORENZ teletypewriter. The LORENZ teletypewriter incorporated a far more complex encoding system than that of the Enigma machines, making the decoding of messages virtually impossible by hand, and was used for German high-command communications. The break into FISH occurred as a result of errors on the part of an operator in Aug 1941 (use of abbreviations and leaving-out words in the re-transmission), so that almost the same message was sent twice with the major error of using the same encryption. Colonel John Tiltman painstakingly deduced the plain-text of the two messages, thus exposing the nature of the encryption i.e. an additive cipher using modulo 2 addition. A few months later Bill Tutte performed a truly inspired feat by deducing the design of the LORENZ machine from these messages. The Heath Robinson machine was specified by Max Newman based on the ground-breaking work of Col John Tiltman and William (Bill) Tutte, and C.E. Wynn-Williams designed the initial counters. Tommy Flowers went on to propose an electronic version of Heath Robinson, but due to the number of electronic valves required the idea was viewed with some skepticism. Undaunted, Tommy Flowers encouraged by Max Newman invested his own time and money into the early stages of the design and build of the prototype Colossus (Colossus Mk 1 employing 1800 valves).


Max Newman's input was the system design in terms of digital Boolean algorithms, but most importantly he got it built and working at the Government Code & Cipher School, Bletchley Park, in 1944.

He devised a way of carrying forward the work of Tiltman and Tutte by use of specially designed machines, and for this purpose was given charge of a new section at Bletchley Park, which became known as the Newmanry.
Max Newman was a pure mathematician, but the war had given him, like Alan Turing, a vision of what an electronic computer could do. He was fully aware of the power of Turing's universal machine concept having been the first to read Turing's 1936 paper in which he describes the universal machine and postulates its operations on numbers with its order code also represented by numbers i.e. a stored program, general purpose computer. Unlike Turing, however, he had no interest in getting personally involved in electronic engineering, he was after-all a facilitator of the first order. 

Newman acted swiftly. In February 1946 Newman wrote to von Neumann (see below) that he was... 
hoping to embark on a computing machine section here, having got very interested in electronic devices of this kind during the last two or three years. By about 18 months ago [i.e. soon after D-Day, and a year before von Neumann's EDVAC report - see later] I had decided to try my hand at starting up a machine unit when I got out. It was indeed one of my reasons for coming to Manchester that the set-up here is favourable in several ways... I am of course in close touch with Turing...

The construction of a digital computer by Newman's department at Manchester University was a key element in the establishment of the UK computer industry. Ironically, Ferranti Computer Systems Ltd involvement with the department ensured the UK had militarised Minicomputers; these, for example, where at the heart of RN ship and submarine Action Information Organisation and Fire Controls systems.


John Von Neumann put computers into the public domain in 1945.

This occurred by virtue of his incomplete but widely circulated paper: First Draft of a Report on the EDVAC. Had he not done so the U.S. Army and the University of Pennsylvania would have ensured nobody else got to know about them, for years; the (initial) development of the digital computer in the UK i.e. Colossus being surrounded by the Official Secrets Acts.


Timeline

July 1943

Design of Electronic Numerical Integrator And Computer (ENIAC) begins at the University of Pennsylvania's Moore School of Electrical Engineering. ENIAC was conceived and designed by John Mauchly and J. Presper Eckert of the University of Pennsylvania. Note the ENIAC was decimal-based and was not a stored program machine i.e. it does not satisfied the above definition of a digital computer. Until the release of information on Colossus it was claimed to be the first digital computer, having been announced to the public in Feb 1946. It could more precisely be said to be the first electronic calculator.

 

Jan 1944 

Colossus operational at Bletchley Park; clearly an earlier date could be declared i.e. when the prototype (Mk I) first performed to the system requirements at Post Office Research Station, Dollis Hill i.e. late November 1943.

 

About June 1944 

John Von Neumann virtually takes control of the development of the ENIAC.

 

August 1944

ENIAC inventors John Mauchly and J. Presper Eckert proposed the construction of an Electronic Discrete Variable Automatic Computer (EDVAC), and design work commenced before the ENIAC was fully operational. Unlike its predecessor the ENIAC, it was binary rather than decimal, and was a stored program machine.

 

About April 1945 

(Enhanced programmability) Colossus Mk III - this is still subject to the Official Secrets Act.

 

30th June 1945

John Von Neumann's report on the EDVAC issued publicly. It contains the first published description of the logical design of a computer using the stored-program concept, which has controversially come to be known as the von Neumann architecture. Some on the EDVAC design team contended that the stored-program concept had evolved out of meetings at the University of Pennsylvania's Moore School of Electrical Engineering predating von Neumann's activity as a consultant there. Is this argument conveniently ignoring Alan Turing's 1936 paper?

 

Autumn 1945 

Max Newman accepted the appointment of Professor of Pure Mathematics at Manchester University, and had (ambitious) plans to build up a powerful department there.

 

Late 1945

Alan Turing having accepted an appointment at the National Physical Laboratory (NPL), Teddington produced his proposal for a stored program digital computer, the Automatic Computing Engine (ACE) and included detailed logical circuit diagrams and a cost estimate of ₤11,200. He felt that speed and size of memory were crucial and he proposed a high-speed memory with an access speed of 1 MHz. The ACE implemented subroutine calls, whereas the EDVAC did not. The ACE & the EDVAC differed greatly, as the ACE employed Abbreviated Computer Instructions, an early form of programming language.

 

12th February 1946 

ENIAC unveiled at the University of Pennsylvania - was it actually working then? It was seriously modified in November 1946 and re-commissioned (finally got working!?) on 29th July 1947. Some claim that this is the true date of first operation as a general purpose computer.

 

19th Feb 1946

Turing presented a detailed paper to the National Physical Laboratory (NPL) Executive Committee, giving the first reasonably complete design of a stored-program computer. However, because of the strict and long-lasting secrecy around the Bletchley Park work, he was prohibited (because of the Official Secrets Act) from explaining that he knew that his ideas could be implemented in an electronic device. Although sanctioned straightaway at high level, the project met with all sorts of obstacles; initially there was conflict with the Post Office (PO) over use of its staff for the electronic design and construction. The PO did eventually build some mercury delay lines for ACE. 
Turing's colleagues at the NPL, not knowing about Colossus, thought that the engineering work to build a complete ACE was too ambitious, so the first version of the ACE that was built was the Pilot Model ACE, a smaller version of Turing's original design.

 

Early 1946 

Max Newman's proposal to build a computer (at Manchester) is accepted by the Royal Society who grant a budget of 35000. At that stage Newman expected that the American Iconoscope would become available as the storage system - as postulated by Von Neumann in his report, but it didn't work.

 

April 1946

A contract to build the new computer was signed in April 1946 with an initial budget of US$100,000. The contract named the device the Electronic Discrete Variable Automatic Calculator (EDVAC). The final cost of EDVAC, however, was similar to the ENIAC's, at just under $500,000.

 

July 1946 

ENIAC formally accepted by the U.S. Army Ordnance Corps. ENIAC was initially designed to compute artillery firing tables, but its first use was in calculations for the hydrogen bomb.

 

November 1946 

FC Williams (out of a job, ex TRE) was appointed to the chair of electrical engineering at Manchester. Newman's idea was that it would be advantageous to exploit Williams' work on cathode-ray-tube storage, even if, as it then appeared likely, an on-site development would take longer than the Americans. Newman had no rigid ideas about hardware, and simply wanted a computer built by the most effective means possible. 

 

29th July 1947 

ENIAC operational in its final form i.e. working.

 

Mar 1948 

Newman offers Turing a post at Manchester - Turing having been side-lined at NPL, see note about security clearance - below.

 

21st June 1948

First successful run of a program on the Small-Scale Experimental Machine, known as SSEM or Manchester Baby. It could store 1024 bits on a cathode-ray-tube, enough to demonstrate the stored-program principle in working electronics, the first in the world to do so. 

 

October 1948 

Alan Turing joins Max Newman at Manchester. (I believe it's fairly significant that Turing lost his security clearance in 1948 - this was an MI5 versus MI6 thing.) The salary for Turing's post came from the Royal Society grant. He was formally 'Deputy Director' of the Royal Society Computing Machine Laboratory. The grounds for appointing him to this post, as minuted on 15 October 1948, were:- 
It was in his paper on 'Computable Numbers' (1936) that the idea of a truly universal machine was first clearly set out. This paper was written for purely theoretical and logical purposes, but Mr Turing has had over two years of practical experience since the war, as designer of the ACE machine which is now being constructed at the National Physical Laboratory.
Thus at the time of his appointment, the character of the Manchester machine as a practical version of the Universal Turing Machine was made clear. It was soon totally forgotten.

 

Late 1948

With the onset of the Cold War 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.

 

September 1949

Electronic Delay Storage Automatic Calculator (EDSAC computer) fully operational (with Bootstrap Loader) at University of Cambridge Mathematical Laboratory, designed by Maurice Wilkes. Initial programs (hand loaded) ran in February 1949.
The EDSAC group was the most influential of the early British computing teams in terms of setting high standards for the development of software and the organisation of a computing service to scientists and engineers.
The success of EDSAC caught the attention of catering firm J Lyons which funded further development of the machine and led to the creation of the Lyons Electronic Office (Leo). One of its first roles was to calculate how much each worker at the hundreds of Lyons tearooms was to be paid. 

 

November 1950 

A version of Turing's ACE design known as Pilot ACE operational - fastest in the world.

A second implementation of the ACE design was the MOSAIC (Ministry of Supply Automatic Integrator and Computer). This was built by Allen Coombs and William Chandler of  PO (Dollis Hill) who had worked with Tommy Flowers on building the ten Colossus computers. It was installed at the Telecommunications Research Establishment (TRE) which soon became the Royal Radar Establishment (RRE) at Malvern and ran its first program in late 1952 or early 1953. It was used to calculate aircraft trajectories from radar data.
The principles of the ACE design were used in the Bendix Corporation's G-15 computer. The engineering design was done by Harry Huskey who had spent 1947 in the ACE section at the NPL. He later contributed to the hardware designs for the EDVAC. The first G-15 ran in 1954 and, as a relatively small single user machine, some consider it to be the first personal computer.
The first production versions of the Pilot ACE, the English Electric DEUCE, of which 31 were sold, were delivered in the spring of 1955.

 

February 1951 

Ferranti Mk 1 operational at Manchester University - the first production machine. The Ferranti Mk 1 was based on the Manchester Mk 1 (operational in April 1949), which was designed at the University of Manchester by Freddie Williams and Tom Kilburn; the Manchester Mk 1 being a development of the 'Baby' that they together with Geoff Tootill built.

 

October 1951 

RA (Tony) Brooker, from Cambridge where he worked with Wilkes on EDSAC, replaces Alan Turing in charge of the software side at Manchester. Whilst at Manchester, Brooker (having already developed in 1954 the world's first publicly available high level programming language - Mk 1 Autocode) developed the concept of a Compiler Compiler; this revolutionary idea was presented at the British Computer Society's 1960 conference, and paved the way for many high-level language developments.

 

1951 

EDVAC operational finally, as design rights dispute rattled-on between the designers of ENIAC and University of Pennsylvania.

 

Replicas of the early computers that can be seen today


A working replica of the Manchester Baby was commissioned in 1998 and may be seen in Manchester's Museum of Science & Industry.

A working replica of a Colossus has been built at The National Museum of Computing (TNMOC), by a team inspired and led by Dr Sale*. Inaugurated in 1994, its completion was marked by the 2007 Cipher Challenge in which Colossus was pitted against modern computers (principally PCs & Laptops) in an authentic code-breaking task including the radio reception of the enciphered messages. The Cipher Challenge was also used to mark the start of a major fund-raising drive for the fledgling National Museum of Computing. The Museum is based in Bletchley Park's Block H (known during wartime as the Newmanry) and Colossus forms the centre-piece of its exhibits.

A working replica of EDSAC is to be built at TNMOC, BP; this project was announced in Jan 2011.


*Tony Sale, perhaps best known for leading the team that built a (as near as possible) replica Colossus computer, helped establish the Computer Conservation Society, co-founded The National Museum of Computing and was a key figure in starting the campaign to save Bletchley Park in the early 1990s.