T. William Olle

Dr T. William Olle FBCS, CITP
Born	1933 (age 91–92)
Education	Boston Grammar School (1943-50); University of Manchester MSc 1954, PhD 1957
Occupation	computer scientist and consultant
Employer	NATO, Control Data Corporation, RCA Corporation, T. William Olle Associates, retired
Organization	T. William Olle Associates

T. William (Bill) Olle (born 1933) is a British computer scientist and consultant, and President of T. William Olle Associates.

Biography

Bill Olle was educated at Boston Grammar School (1943-1950). He received an M.Sc. degree in 1954 and a Ph.D. degree in 1957, both in Astrophysics at the University of Manchester, which involved extensive programming work on the Manchester University Electronic Computer.

In 1957, he moved to the Netherlands where he worked in computing for a NATO organisation. In 1964, he moved to the United States where he was employed by Control Data Corporation in Palo Alto until 1966. From 1967 to 1971, he was employed by the RCA Corporation in Cherry Hill. In 1972, after a year in Norway, he returned to the UK to establish his own consultancy firm T. William Olle Associates, specialising in database management applications and information systems methodologies. He consulted clients in Europe, Australia and Canada, and presented lectures on database topics around the world. He retired in 1993.

Beginning in the 1970s, Olle became active in the CODASYL organisation as Chairman of its Systems Committee and spearheaded the preparation of two early analytical reports on "Generalized Database Management Systems". He has represented the British Computer Society on IFIP TC8 since its inception in 1977. He was also active in database standards work in ISO and was chairman of the BSI standards committee for many years.

Bill Olle was awarded an honorary doctorate by Middlesex University in 2001.Template:Citation needed

Work

Bill's research interest in the field of computing started in 1953 at the University of Manchester. In the 1960s, he became interested in database applications, and after his retirement in the 1990s focused on the history of computing, and on "professionalism in the computer field".

Electronic Brain

This article written by Bill Olle was originally published in the 1954 Year Book of The Old Bostonian Association. The author was at the time a research student at the University of Manchester and his research involved extensive use of the Manchester University computer.

Having chosen the title for this article, I must point out that the name "Electronic Brain" is perhaps misleading; the correct name of the machine I wish to describe is the "Manchester University Electronic Computer."

Most people will have heard of this machine through the popular press. The general impression which seems to prevail is of a robot with flashing eyes, which does multiplication very rapidly, and on occasions stoops to playing draughts, chess, or noughts-and-crosses. If there is any similarity between the outward appearance of the Manchester computer and the human form, I have failed to see it. The computer consists of two bays sixteen feet long, eight feet high, and four feet wide - rather like four very large glorified kitchen cabinets placed back to back in pairs, with doors which open to reveal thousands of valves and other electronic devices. There is also a control desk about the size of a church organ control, with numerous switches and cathode ray tubes that enable the operator to observe what is going on inside the two large bays. These two bays contain about 4,000 valves, 2,500 capacitors, 15,000 resistors, 100,000 soldered joints, six miles of wire, and several cathode ray tubes. If you have ever had any trouble with your wireless, you will appreciate that the scope for things going wrong is tremendous.

To consider the abilities of the machine: It can multiply two twelve-figure numbers together in one five-hundredth of a second. How long would it take you to multiply 127,865,439,608 by 659,413,780,046 ? (Incidentally it takes the machine as long to multiply 2 by 2). The machine can do in an hour as much numerical calculation as a competent mathematician aided by a desk adding machine can do in 300 hours. It can do three of the basic arithmetical operations - add, subtract, and multiply. It can also divide, but in the machine this is not a basic operation but a combination of the other three. It can "remember" numbers, and it can make decisions. The machine does this by testing the size of a number, and then doing one thing or another depending on the result of the test.

How do we tell the machine what to do? Every problem has to be broken down into the basic operations already described before the computer can solve it. This "coding" of problems is by no means simple; I should say it would take the average mathematician about two months' intensive work to learn how to code-up instructions for a computing machine. The everyday system of counting uses ten digits, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9: the machine uses only two digits,0 and 1, working in the scale of two, or binary scale. When we have coded-up our instructions, they are punched on to a tape about three-quarters of an inch wide and varying in length up to several yards. After we have fed our coded instructions into the machine, we switch on and it starts work. This is what I mean when I say the machine "remembers"; it stores the information until it is required for use.

When the machine has finished the calculation, it prints the answer on a teleprinter on the control desk. If the answer is wrong, it is probable that a mistake has been made in the instructions given to the machine and these must be carefully checked. If we give the mechanical slave a wrong instruction, it will obey, and the fault is ours, not the computer's.

The machine has great importance in mathematical research. Many problems which have frightened mathematicians, in view of the arithmetic involved, are now being tackled with the machine "doing the dirty work." Research problems in astronomy, nuclear physics, and biology necessitate much calculation, and the machine is of great help in such projects. Some of them are pure research, seeking after knowledge for the sake of it: the machine is also being applied to problems of statistics, engineering, commerce, and economics. In economics, machines might be used to answer questions such as - "If there is less electricity available, what will be the effect on all other industries and on economics as a whole?"

Again, experiments are being made in the use of the machine in meteorology. The atmosphere is believed to obey certain physical laws and in order to make a weather forecast one has to solve a system of complicated mathematical equations: by the time they are solved by normal methods, the results are out of date.

There is a lighter side to computing-machine operating. First, consider the ability to play tunes. If the operator includes in his programme of instructions one certain instruction, when the machine impulse will be applied to the diaphragm of a loud speaker. By doing this repeatedly and rhythmically, a steady note, rich in harmony, can be produced; and by suitably varying the frequency of the note, it is possible to get a reasonable tune from the machine. But musicians need have no fear; computing machines are expensive to maintain!

The Manchester University Machine has been made to play a fair game of chess and to solve simple chess problems of the "mate in two" type; but it is much more successful at the game of draughts, for which a complete mathematical theory exists. Of course the machine does not "know" this theory, but is told it and makes its moves accordingly.

On one occasion the Society for Physical Research tried to see if the operations of a machine could be influenced by concentrated thought on the part of their research workers, most of whom were elderly ladies. They also tried to discover whether they, in turn, could be affected by vibrations from the machine. The experiments were, however, a complete failure; electronic computers are evidently less co-operative than human beings in experiments in telepathy.

T. W. Olle, B.Sc.

Publications

Bill Olle published numerous books and articles. The following is a selection:

• 1971. Feature Analysis of Generalized Data Base Management Systems: technical report Conference on Data Systems Languages Systems Committee.
• 1978. Codasyl approach to data base management.
• 1983. Information Systems Design Methodologies: Improving the Practice. IFIP WG 8.1 Working Conference on Feature Analysis of Information Systems Design Methodologies 1983: York, England. Edited with Henk G. Sol and C. J. Tully.
• 1982. Information Systems Design Methodologies: A Comparative Review. IFIP WG 8.1 Working Conference. Edited with H.G. Sol and A.A. Verrijn Stuart.
• 1988. Computerized Assistance During the Information Systems Life Cycle. Proceedings of the IFIP WG 8.1. Working Conference on Computerized Assistance During the Information Systems Life Cycle, Cris 88, Egham, England, 19–22 September 1988. Edited with A. A. Verrijn Stuart and L. Bhabuta.
• 1988. Information systems methodologies: a framework for understanding. North-Holland.
• 2008. Eight Significant Events in the 50 Year History of Computing. British Computer Society Kingston and Croydon Branch joint event with Central London Branch, Davidson Building, 5 Southampton Street, London, England, 2 December 2008