Mark 1 Later Machines Baby Rebuild Geoff Tootill
Gordon Bell is a Senior Researcher at Microsoft and a computer industry consultant at large, working on computer-related projects, especially parallel processing. Bell is also a director and partner of The Bell-Mason Group that provides systems for venture development. He has an SB and SM degree from MIT (1956-57) and honorary D. Eng. from WPI (1993). In 1958 he was a Fulbright Scholar to Australia and during 1959-1960 was a member of MIT's Speech Group. He spent 23 years at Digital Equipment Corporation as Vice President of Research and Development, where he was responsible for Digital's products. He was the architect of various mini- and time-sharing computers and led the development of DEC's VAX and the VAX Computing Environment. Bell has been involved in the design of about 50 computers, a dozen of which were multiprocessors, and many products at Digital, Encore, Stardent, and a score of other companies. During 1966-72 he was Professor of Computer Science and Electrical Engineering at Carnegie-Mellon University. In 1986-1987 he was the first Assistant Director of the National Science Foundation's Computing Directorate. He led the National Research Network panel, and was an author of the High Performance Computer and Communications Initiative. In 1988 he was Vice President of Engineering at Ardent Computer which built the first, graphics supercomputer.
Bell has co-authored six books and many papers about computer structures and most recently, start-up companies. He is a member of various professional organisations including the AAAS (Fellow), ACM, IEEE (Fellow), and the National Academy of Engineering. He is on the boards of Cirrus Logic, Chronologic Simulation, University Video Communications, and Visix Software, and is a technical advisor to various companies including Kendall Square Research, Microsoft, and Sun. He is a founder and director of The Computer Museum, Boston. Bell is a member of various professional organisations including the AAAS (Fellow), ACM, IEEE (Fellow and Computer Pioneer), and the National Academy of Engineering. His awards include: the Mellon Institute Medal, and IEEE's McDowell and Eckert-Mauchly Awards, and the Von Neumann Medal. President Bush awarded Bell The National Medal of Technology in 1991 "for his continuing intellectual and industrial achievements in the field of computer computer design; and for his leading role in establishing cost-effective, powerful computers which serve as a significant tool for engineering, science, and industry".
Professor Tom Kilburn was born in Dewsbury in 1921. He graduated in Mathematics from the University of Cambridge, and on his call-up in 1942 he was ordered to take a crash course in Electronics and then report to the Telecommunications Research Establishment (TRE) at Malvern. Here he was assigned to a small expert electronics group run by Freddie Williams that acted as a central think-tank and problem-solver for other groups at TRE and Air Defence establishments.
In December 1946 Freddie Williams moved to the University of Manchester as Professor of Electro-technics (soon renamed Electrical Engineering). He asked for Tom and an assistant to be seconded from TRE so that they could continue research into a new project for the storage of digital information on a Cathode Ray Tube. By the autumn of 1947, they could store 2048 bits on a CRT and decided to build a small computer round it, to test it. In December 1947 Tom wrote a progress report for TRE; this was distributed widely and was instrumental in the storage system being used on a number of computers worldwide.
In June 1948 the Small Scale Experimental Machine ran its first program, thus becoming the first stored-program computer. Work continued with an expanded team for 18 months to develop it into a full-scale computer, which in turn was used as the prototype for the Ferranti Mark 1. Tom was awarded his PhD. later in 1948, and joined the department as a Lecturer. During the period 1947 to 1950 he was the main driving force behind the development work, with Freddie Williams as the active manager. From 1951 Freddie Williams reverted full time to Electrical Engineering, with Tom, now a Senior Lecturer, leading the hardware computer group and the Computing Machine Laboratory, which housed the new Ferranti computer, the Computing Service built round it, and the software development team.
Computer development continued with two new machines at the same time, the Transistor Computer and Meg. These became prototypes for the MV950 and the Ferranti Mercury. In 1956 the massive (hardware and software) MUSE project was started, which required the cooperation of Ferranti Ltd. to build the first machine, the Ferranti Atlas, first running in 1962. All these machines were also at the forefront of computer development worldwide. In 1960 Tom was appointed Professor of Computer Engineering, and in 1964 the computer group broke away from Electrical Engineering to form the first UK Department of Computer Science, with Tom, now Professor of Computer Science at its head. Once the undergraduate school was launched work started on his fifth machine, MU5, which first ran in 1972. This was developed with the help of ICL, and contributed significantly to the architecture of the larger ICL2900 series machines. Thereafter the Department diversified its research effort over a far wider Computer Science field.
Tom retired in 1981. He was elected a Fellow of the Royal Society in 1965 and was subsequently awarded the Society's Gold Medal. In 1973 he was awarded a CBE.
Geoff Tootill was born in Chadderton. He obtained his Mathematics MA from Cambridge and an MSc in Electrical Engineering from the University of Manchester for work done on the `Baby' machine. After graduation from Cambridge he was a scientific officer with Fighter and Bomber Command before being transferred to TRE where he worked on airborne centimetric radar. He then joined FC Williams' group at Manchester and helped Tom Kilburn design and build the Baby. Soon after the Baby ran its first program, he left the University to work for Ferranti and so carried out an important `technology transfer' role during the period that Ferranti were beginning to build the Mark 1. Between 1949 and 1954 he was a senior lecturer at Shrivenham. He then took up a post as Senior Scientific Officer at Farnborough where he stayed until 1963.
His varied career continued with spells at the European
Space Agency, the DTI, the Executive Office of European
Informatics Network Project and the National Maritime
Institute. He has written about 20 papers, was a co-author of
an early book on electronic computers and chaired the IFIP
drafting committee for the "Vocabulary of Information
Laurie Allard was born in Gravesend, Kent, in 1923. In 1940 he joined the General Electric Company's Research Laboratories at Wembley Middlesex as a student assistant. During his work there, on cathode-ray tubes, he graduated as a part-time student at Chelsea Polytechnic, first in Physics in 1943, and then Mathematics in 1945.
The war-time work at Wembley on cathode-ray tubes covered a wide range of Service applications. The end of the war brought fresh areas of work into high frequency oscilloscope tubes, marine radar tubes, storage and switch tubes and of course the ever expanding television tube market.
He was granted Fellowship of the Institute of Physics in 1961. He was promoted to Group Leader of the Cathode-Ray Tube Group in 1965, and began research on transparent phosphors, high density cathodes and very high resolution tubes. In 1970 the Group was transferred to Hammersmith.
He retired in 1980, and has since happily pursued other areas of work, including careers guidance advisor for the Institute of Physics. In 1995, his work for the Abbeyfield Society, which provides accommodation for elderly people, was acknowledged by His Royal Highness The Prince of Wales, with the presentation of The Royal Patron's Award.
Alec Robinson was born in Scunthorpe in 1924. He graduated from Clare College, University of Cambridge, and in 1944 he began his career at English Electric Ltd where he worked until 1950, first as a student apprentice and then development engineer. Crucially during this period he took his PhD at the University of Manchester (1947-49), where he joined the Mark 1 engineering team and worked on the design of the hardware multiplier.
Along with Geoff Tootill, he then transferred to Ferranti Ltd where he worked as Senior Development Engineer. He was one of the main contributors of the many improvements introduced into the Ferranti Mark 1. For example, he redesigned his original multiplier, introducing more parallelism, so that it ran on average five times as fast.
In 1962 he moved to the University of London as General Manager of Computers (Bloomsbury) Ltd and then as Director of the University of London Computing Centre, where they had one of the three Atlas computers. He became director of the National Computing Centre in 1969 where he remained until 1974. He then worked at the University of Wales, College of Cardiff, as Director of the Computing Centre where he stayed until his retirement in 1991.
Gordon Eric Thomas was born in Cwmamman, Carmarthenshire South Wales in 1928. After graduating in Physics and Electronic Engineering at the University of Manchester in 1948, he joined the design team of Professors F.C.Williams and Tom Kilburn who conceived and designed the Ferranti Mark 1 and Mercury computers. His postgraduate qualifications of MSc. and PhD were awarded for work done at this time in Manchester. His University contemporaries assigned the name Tommy by which he prefers to be known.
He left the University in 1955 to found the Central Digital Computing facility at Imperial Chemical Industries Ltd (ICI). In 1966, he moved to the University of Edinburgh to set up and direct the first of the three Regional Computing Centres. While Director, he assisted in the development of an overall IT strategy within the first School of Information Technology of any British University.
His Australian career started in 1985 when he took up the post of Foundation Chief at the new CSIRO (Commonwealth Scientific and Research Organisation) Division of Information Technology. In 1988 he assisted Bond University to establish a School of Information and Computing Sciences and a Research Park. In 1989 he took up the permanent joint post of foundation Professor of Computing Science and Coordinator of the Research Park. He has served on numerous committees and the Boards of commercial and other bodies.
Since his retirement in 1990 he has established a consulting
partnership, TT Enterprises, which is undertaking a wide
variety of specific commissions within the Gold Coast
Region of Queensland. Recently, he has been a vigorous
advocate for wider community involvement in the use of
Internet resources. He has coordinated the development of
World Wide Web material for Bond University, the Currumbin
Sanctuary and the National Trust of Queensland.
He is a Fellow of the Royal Society of Edinburgh, a Fellow
of the British Computer Society and a Member of the Institution
of Electrical Engineers.
Professor D.B.G. (Dai) Edwards was born in Pontypridd, South Wales in 1928. He came to the University of Manchester in 1945 to read physics, specialised in electronics, studying under Professor F Williams, and graduated in 1948. He has worked continuously in the University until his retirement in 1988.
He first became a research student in the Electrical Engineering Department, working on the design of memory and computer systems which contribute to the Ferranti Mark 1. He was awarded an MSc in 1949 and joined the newly created Computer Machine Laboratory as assistant lecturer. In 1953 he was promoted to lecturer, and in 1954 he was awarded a PhD on design aspects of the Meg prototype computer, which later became the Ferranti Mercury. In 1959 he was promoted to Senior lecturer and then to Reader in 1964.
He worked with Professor Tom Kilburn both on research and on the establishment of the first Computer Science Undergraduate School in the country, set up in 1964. He was appointed to a Chair in 1966 and became ICL Chair of Computer Engineering in 1967. He was Head of Department from 1980 to 1987, and Dean of the Faculty of Science in 1982/83.
His research interests have always been hardware-oriented, with contributions to the Ferranti Atlas, to its `one level' store, the fast carry, the multiplier system, the interrupt system, the drum, disc and tape, systems, the partially switched flux system for the fast index registers, telephone line communication etc. He also had interests in early computer controlled instrumentation, and an automatic X-ray goniometer, which received a Queens Award for Technology in 1968. He was joint holder on research grants for the MU5 and MU6, making significant contributions to their design. He has published numerous papers and taken out a number of patents.
Keith Lonsdale was born in 1923, and graduated from the College of Technology, University of Manchester, in 1945. He worked at Salford Electrical Instruments as development engineer, then AC. Cossor, and joined Ferranti in September 1948 as computer development engineer. He was appointed Senior Engineer in 1950 and Chief Engineer of the newly formed Computer Department in 1953. In 1961 he was appointed Manager of the Technical Services Department and Manager of West Gorton Factories in 1963.
In September 1963 the Ferranti Computer Department was sold to ICT and Keith transferred to Letchworth where he remained as Manager of the Computer Division of the Engineering Services Organisation until 1971, when he left to move to Scotland. He started an engineering workshop and ran a Sea Angling Charter Boat until 1986 when he retired. He and his wife now enjoy regularly playing golf, with indoor bowling in the winter.
Fred Bushby graduated in mathematics from Imperial College, London, in 1944. He joined the RAF Meteorological Branch and served at several stations until 1948. He then became Scientific Officer in the Meteorological Office. In 1950 he was posted to the Forecasting Research Branch in Dunstable, launching a career in numerical weather prediction spanning 27 years.
He wrote numerical programs to compute fields and tendencies based on Sutcliffe's ideas, using the EDSAC machine, and then proceeded to develop the `2 1/2-dimensional' numerical prediction model, which produced its first forecast in 1954 using the Ferranti Mark 1 at Manchester. This led to the construction of a 3-level geostrophic model, introduced in 1959 when the Meteorological Office acquired its own Ferranti Mercury computer.
After a two year break as Chief Meteorological Officer in Aden, he returned to develop an advanced 10-level primitive-equation model. This was designed to predict the development of fronts and make the first quantitative forecast of precipitation on a fine mesh 100km horizontal grid, exploiting the greater computing power of the Ferranti ATLAS. He later extended the fine-mesh model to cover most of the northern hemisphere, and when this was combined with the Met Office acquisition of an IBM 360/195 computer, in 1971, it brought an immediate impact on the quality of forecasts, and put the UK in the world forefront of numerical weather prediction.
In 1974 his achievements were marked by promotion to Deputy Director in charge of Dynamical Research, and in 1977 he was transferred to the post of Deputy Director Forecasting Services. In 1978 he was made Director of Services (Deputy Director General). He retired in 1984.
Charlie Portman was born in 1933. He trained as an electronic engineer and joined the Ferranti Computer Department in 1954 as a circuit engineer. He went on to become a logic designer, test programmer and team leader. He worked on the design of Sirius and Orion, and later was responsible for 1900 and 2900 hardware development at West Gorton. In 1972 Charlie was appointed Manager of the newly formed Software Division with staff at Manchester, Kidsgrove, Stoke and Stevenage and was responsible for 1900 Executive Programmes, Test Programmes for both 1900 and the early 2900 machines and for an early 2900 Supervisor programme. He also managed Design Automation for 2900 logic design.
In 1984 he became Project Manager of the Alvey DHSS (Department of Health and Social Services) Large Scale Demonstrator, for the Alvey Directorate of the Department of Trade and Industry (DTI). Charlie later managed the DTI/ Science and Engineering Research Council (SERC) jointly funded 'MAKE' collaborative project with British Coal and the University of Liverpool.
In 1993 he took over management of the InformationPlus unit developing Knowledge Based and Hypertext products.
Charlie was a founder member of the ICL Technical Journal Editorial Board and was a long serving member of SERC and DTI grant awarding committees for academic and joint research. Charlie is an ICL Fellow and a Technical Consultant in ICL's Technical Directorate's Research Unit.
Professor Richard Grimsdale is chair of Electronic Engineering at the University of Sussex. He was appointed in 1967, the first person to hold the position.
He graduated in Electrical Engineering at the University of Manchester in 1950, and later that year attended the Summer School in Programming the EDSAC at the University of Cambridge. As a research student, he was given the task of writing test and diagnostic programs for the Ferranti Mark 1, which was quite difficult due to the almost total lack of circuit diagrams. He joined the staff at the University of Manchester and embarked on the design of a small machine. However the advent of the point contact transistor led to the development of the Transistor Computer. He then worked on a number of projects, including a syntax-based scheme for character recognition, a heuristic design system for an Electricity Board, the Graphical Display for the Mercury Computer and the Read-Only Memory for the Atlas. During a period of six years with AEI he was engaged on a number of industrial process control applications of computers.
After moving to Sussex in 1967, working on Computer Communication Networks, he switched his interest to Computer Graphics and has led a team developing VLSI accelerator chips for 3-D image generation and photo-realistic rendering software.
Frank Sumner was born in Manchester in 1931, and graduated from the Department of Chemistry, University of Manchester, in 1951. As a postgraduate he worked in Theoretical Chemistry under Professor H.C.Longuet-Higgins. This involved a great deal of computation using a Marchant electro-mechanical calculator. Early in 1952 his supervisor suggested that he should talk to Alan Turing, who "had an electronic computer". Turing gave him a copy of the programming manual and told him to come back when he had written a program.
That was more or less the end of Chemistry and after getting a PhD in 1954 he held a post doctoral post in the Computer Laboratory until going to Aldermaston in 1956. He returned to the University of Manchester in 1957 and used the Mark 1 (until it was removed) and then the Mercury, for pattern recognition and artificial intelligence.
He simulated the Atlas one-level store on the Mercury and used this to develop the Drum Transfer Learning Program for the Atlas. He also worked on the logical design of the Atlas Central Processor. Later, he worked on the design of the MU5.
He helped to set up the Undergraduate Course in Computer Science and became the first admissions tutor in 1964. He was promoted to Professor in 1967. Since the early 1970's he has been involved in many national and international organisations, including the Computer Board, SERC, AFRC, IFIP and the British Computer Society, of which he was president in 1978/79. In 1983 he became Director of the Manchester Computing Centre. He retired in 1995 and returned part-time to the Department of Computer Science.
Tony Brooker graduated in Mathematics from Imperial College in 1945. It was here in 1947 that he first became involved with computers by building from electro-mechanical relays a multiplier unit which served as the arithmetical basis of the Imperial College Computing Engine based on the same technology. In 1949 he moved to the Computing Laboratory at Cambridge to work first on the differential analyser and subsequently on software development for the EDSAC under the influence of pioneers such as David Wheeler and Stan Gill. In 1951 he moved to Manchester to relieve Turing of the day to day running of the Manchester Mark 1. Here, he established the Mark 1 as a viable system for a whole range of users from different disciplines. More importantly he started his work on high level languages with Mark 1 Autocode, a boon to many early users. This was followed by Mercury Autocode, a contemporary of the first Fortran and finally Atlas Autocode and a compiler-compiler.
In 1967 he moved to Essex University as founding professor of the Computing Centre where, with the late Professor Keith Bowden, he initiated (just two years after Manchester) one of the first undergraduate schemes in Computer Science that encompassed both software and hardware. Although not a personal contributor, he created the environment for research and teaching in artificial intelligence at Essex, now realised in hardware by a Robotics Laboratory, opened by Professor Mike Brady in 1995. After serving many years as faculty Dean and as a Pro-Vice-Chancellor he retired in 1988. Subsequently he has published a book on a data based programming language.
David Howarth's first encounter with computers was as a research student at Imperial College, University of London, using the Ferranti Mark 1 at Manchester to handle band structure calculations in solid state physics. He then spent a year at MIT using the Whirlwind system for similar calculations, followed by a period at the Royal Radar Establishment (Malvern), using the TREAC machine for problems in mathematical physics.
In 1960 he switched from physicist to systems engineer moving to Ferranti in London, which led to involvement in the Atlas project and the challenge and excitement of leading the team working on the Atlas Supervisor, one of the world's first operating systems. He then continued this work in parallel with work on designing the ICL `New Range': the ICL 2900 series of computers.
In 1970 he moved from industry back into academia, first in the form of a Chair at the Institute of Computer Science at the University of London teaching on the MSc course, and then as a professor in the Department of Computing at Imperial College, teaching at undergraduate and postgraduate levels. He has also been extensively involved in other institutions as an external examiner and an assessor in the Quality Assurance exercise.
Professor Jeff Rohl is with the Department of Computer Science in the University of Western Australia. His current challenge is to to teach Java to first year students.
He came to Manchester from Brisbane in 1960 as one of the first cohort of British Commonwealth Scholars to do a PhD under Tony Brooker and, later, Derrick Morris. He worked initially on the (Brooker-Morris) Compiler Compiler and subsequently on languages for Atlas --- his PhD thesis title "The Compiler Compiler and Atlas Autocode" says it all. He subsequently joined the staff and worked on the design of the order code of MU5.
With the introduction of the BSc course in 1967 he was able to indulge his interest in teaching algorithms and programming languages. In 1972 he became Professor of Computation within the Faculty of Technology.
In 1976 he returned to Western Australia (half-way home as Derrick Morris put it) to found the Department of Computer Science in The University of Western Australia, which he headed for 15 years.
Roland Ibbett has been Professor in the Department of Computer Science at the University of Edinburgh since 1985 and was Head of Department from 1987 to 1995. In 1994, he became Vice-Principal of the University. He graduated in Physics at the University of Manchester in 1962 and went on to obtain an MSc at Manchester and a PhD at the University of Hull. In 1967 he was appointed Lecturer in Computer Science at the University of Manchester and was subsequently promoted to Senior Lecturer and then Reader. He was a major contributor to both the design and production of the MU5 and was responsible for day to day liaison with the ICL West Gorton factory whilst the hardware for MU5 was being manufactured. In 1976 he spent a semester at Carnegie-Mellon University, Pittsburgh.
His research interests continue to centre on computer architecture. He is currently involved in the development and use of a Hierarchical computer Architecture design and Simulation Environment (HASE). HASE is a virtual laboratory for research in computer architecture, and is in use both in teaching and research.
Ian Watson is Professor of Computer Science at the University of Manchester. He graduated in Physics and Electronic Engineering from the University of Manchester in 1969. He obtained a PhD in Computer Science in 1973 for work on the MU5 computer system. This was followed by a period working on the structure of special purpose machines for both graphics and signal processing.
His major work during the last twenty years has been on the architecture of parallel computers. This started in 1978 when, together with John Gurd, he initiated the project which built the prototype Dataflow Computer at Manchester University. He later led the Manchester University team which was responsible for the architectural design of the Flagship machine within the ALVEY Flagship project. This collaborative work continued as part of the ESPRIT EDS project, leading to the ICL Goldrush parallel database server.
More recently he has been involved in projects concerned with the architecture of Virtual Shared Memory and the possibility of `single chip' multi-processors.
Brian Procter is chief architect of ICL High Performance Systems in Manchester responsible for the technical aspects of the "Millennium" multi-server programme.
In 1958, he started work with the Computer Division of EMI Electronics as a commissioning engineer on the EMI-DEC 1100 prototype. In 1962, he moved to ICT, Stevenage, with the merging of EMI's computer interests. As logic designer and later manager of the processor design unit at Stevenage, he worked on most of the smaller ICT/ICL 1900 range machines, 2903/4, and subsequently the smaller 2900 range machines. He moved to ICL Manchester in 1976 and was part of the team which developed the ICL Series 39 nodal architecture. He worked on the introduction of VLSI into ICL mainframes and the technology source agreement with Fujitsu. He was technical manager on the Alvey ‘Flagship’ project researching parallel processing, and the subsequent Esprit EDS project which led to the ICL Goldrush parallel database machine. He was appointed as an ICL Fellow in 1991.
Chris Burton graduated in Electrical Engineering at the University of Birmingham in 1955. He worked on computer developments in Ferranti Ltd and then ICT and ICL from 1957 until his retirement in 1989. He is a Chartered Engineer, Fellow of the Institution of Electrical Engineers, Fellow of the British Computer Society and is a Visiting Lecturer in the Department of Computer Science at the University of Manchester.
He is a member of the Computer Conservation Society and of course, most significantly at this Golden Anniversary Conference, he has been Project Manager of the Small-Scale Experimental Machine Rebuild Project. In addition to pioneering the re-build effort, he has taken time to inform and entertain numerous audiences with his lively presentations covering the trials and tribulations of undertaking the rebuild.