Flowers, Thomas Harold 'Tommy (Oral history)

British civilian in east London during First World War, scientist working on automatic exchanges and long distance telecommunications for Post Office in GB, 1922-1965; led team which designed and produced 'Colossus' electronic valve programmable logic calculator used by scientists at Bletchley Park in decoding German coded 'Fish' messages, 1943-1944. This machine was subsequently recognised as the world's first electronic computer

REEL 1 Background in East Ham, London, 1905-19: family; education and interest in mathematics; sporting activities; eye accident; interests in modelling and motors; attending East Ham Technical College; effects of First World War, 1914-1919 including Zeppelin and aircraft raids, rationing and Armistice celebrations, 11/11/1918; nature of scientific education at EHTC. Recollections of initial period at Post Office Stores Depot, Islington, ca 1922: passing entrance examination; duties testing telephone equipment. Recollections of periods as technician and research engineer at Circuit Laboratory at King Edward's Street Post Office and Post Office Research Laboratories at Dollis Hill, ca 1922-1939. REEL 2 Continues: evening classes in telephone engineering and lack of formal training with Post Office; practical as opposed to theoretical knowledge of former army signallers; design, manufacture and testing of automatic exchange equipment; explanation of relay and ratchet electrical/mechanical switches; method of problem solving; automatic dialling system operating electrical switches; problems in equipment connecting automatically to correct exchanges; method of connecting automatic and manual exchanges; necessity of working at home; absence of concept of computer; overcoming problems and successful introduction of automatic exchanges; role studying problem of long distance dialling and background to introduction of electronic valves to replace mechanical switches. REEL 3 Continues: role studying problem of long distance dialling and background to introduction of electronic valves to replace mechanical switches; successful long distance dialling tone trials between London and Bristol and subsequent value of system during German air raids, 1939-1945; question of international co-operation on telecommunications and work of commission established under League of Nations; method of distinguishing between dialling tones and speech; pay and question of moving to private industry; question of awareness of approach of war and reserved occupation status; opinion of research heads Bertram Cohen and Gordon Radley; organisation and composition of research groups; hours worked; work at home; reaction to outbreak of war, 3/9/1939. REEL 4 Continues: emergency evacuation to GB during visit to meet telecommunications scientists in Germany, 9/1939; work on developing radar stations communications system for RAF, 1939-1942. Recollections of work leading to development of Colossus decoding machine at Dollis Hill connected with code breaking at Bletchley Park, 2/1942: meeting with Alan Turing and his explanation of technology of code breaking in connection with German Enigma coding machine; background to failure of initial machine to decode Enigma; opinion of Turing; German introduction of teleprinter 'Fish' code' and decoding significance of receiving copy of 'Fish' message duplicated in clear transmissions; Dr Neuman's role in devising mathematical process to determine setting of code wheels; taking over project after failure of team originally assigned to develop code breaking 'Robinson' mechanical machine based on teleprinter due to problems in driving paper tape sprockets; decision to abandon 'Robinson'; developing concept of using electronic valves switching to get necessary processing speed. REEL 5 Continues: principles of German Lorenz SZ42 coding machine producing 'Fish' code; developing concept of electronic Colossus machine to reverse coding procedure; testing possible speed of imput paper tape; decision not to commission Colossus machine from Dollis Park as result of questions of reliability of electronic valves and completion time required; decision to manufacture Colossus on own authority and formation of special team at Dollis Hill assisted by Post Office telephone equipment factory; method of designing and 'programming' processing of Colossus to determine Lorenz wheel positions; manufacture of prototype Colossus, 2/1943-12/1943; refinements to Colossus design to introduce quicker parallel processing and preparations to manufacture further machines despite initial lack of reaction from Bletchley Park; emergency orders from War Cabinet to prepare 12 Colossus machines by 1/6/1944; working flat out; successful production of Colossus and eradication of initial fault by 1/6/1944; second hand reports of role of Colossus machine in decoding German messages which assisted Eisenhower in making his strategic decision over D Day, 6/6/1944; assignment to produce electronic machine to use results of automatic radar tracking to lay anti-aircraft guns, 1944-1945; question of concept of Colossus as computer with memory, programme and processor; question of Turing's minimal role in development of Colossus. REEL 6 Continues: difficulty in explaining concept of Colossus to decoding staff at Bletchley Park; question of Turing's minimal role in development of Colossus; work of Professor Randle in defining of Colossus as first computer, 1975; story of German scientist stopped from working in same field, 1939-1945; Colossus attachment developed by mathematicians to break German code changes; award of honorary doctorate; state of health of Colossus team; work on memory devices for electronic machines and question of duplicated research. Post-war career at Dollis Hill, 1945-1965: ban on working on computers and destruction of Colossus machines; work on valve telephone exchange; superiority of transistors; question of effect of secrecy and delays in research work; being given Babbage award, 1998.