IN THE history of computing few giants compare with Charles Babbage, the Victorian mathematician and polymath who died 125 years ago and who is often dubbed the "grandfather of computing".
There is no doubting the power of his vision and legacy, even though none of his machines was completed during his lifetime, but he is in truth more a granduncle than grandfather.
Strange as it might seem at first, his interest in calculating machines began as an obsession with the errors in published tables. Errors could arise during the initial manual calculation, the transcript ion or printing, and were more than idle curiosities: Babbage calculated that errors in annuity tables cost the government £2 million a year, while mistakes in navigational tables could sink a ship.
Babbage dreamed of producing an automatic error free calculating machine that would also print its answer, removing at one fell swoop all the sources of error. The same idea had actually been suggested as early as 1784 by one Johann Muller, but calculating devices then were still mere playthings. Babbage wanted something robust and industrial.
In 1832, aged 31 and with much government support, he began designing his first "difference engine". It used the mathematical principle of finite differences (reducing multiplication and division to simpler addition), had numerous toothed and numbered wheels, and was to impress its results onto strips of papier mache, which could then be used in printing the books.
The massive project needed numerous drawings, prototypes and even a complex notation for all the parts, but there were spinoffs to the developing machine tool industry, while Babbage's researches into manufacturing techniques became an important text in political economy.
After 10 years, a subsection with 2,000 parts was built. It was probably the first automatic calculator, from a workshop that was probably the first computer laboratory. It still works, in its limited form, and suggests that the complete machine might have been feasible.
But in 1833 Babbage fell out with his skilled engineer, Joseph Clement, and work stopped. Only half the 25,000 parts were made, yet it had cost the government £17,470, at a time when you could buy a steam locomotive for £875!
Undeterred, Babbage began designing an even more complex machine, his "analytical engine". A massive device, with 50,000 wheels on 1,000 vertical axes and probably needing steam to power it, it would have been a programmable calculating machine capable of the "do loops" and "IF...THEN" operations which are so familiar in computer programs today.
Other features in common with modern computers included an input card reader, a memory or "store", a processor or "mill" and a printer. But there the comparisons end, and there is in truth no direct connection with modern computers.
Ada Countess of Lovelace helped to promote the analytical engine in her writing, and Babbage's reputation rests on it, yet it too was never built. By the 1840s the government had lost patience and interest, and a Swedish father and son were selling a simpler calculating machine.
The Scheutz calculator used some of Babbage's ideas, but a cruder design, leading some to suggest that Babbage's machines were unnecessarily complex. Yet the Scheutz device and the others that followed were troublesome and of limited use.
Despite the setbacks Babbage found time to stand for parliament, travel and write prodigiously, and invent any number of things, including a black box recorder for railway tracks, coloured theatre lights, a campercarriage with cooking facilities (which he used on his European travels), and even footwear for walking on water.
A keen reformer, he proposed a decimal coinage, flat rate postage throughout the country, and life (not hereditary) titles. He published designs for diving bells and seismographs, predicted the end of coal and the need for tidal power, and his Saturday soirees were renowned affairs.
His failure to realise any of his machine designs has been attributed to many factors, from personality and personnel problems to runaway costs and the technological limitations of working in an era before mass production. And yet his engines would probably have worked.
In 1985, the London Science Museum began building his second difference engine, a simpler one which he designed in the 1840s. As much as possible, they used techniques and materials that would have been available to Babbage at the time.
It cost £300,000, weighs 3 tons, has 4,000 components, and can handle numbers of up to 31 digits. In 1991, in time for the bicentenary of Babbage's birth, the power wheel was cranked and difference engine No. 2 computed its first full scale calculation, the first 100 values in the table of powers of seven.
Today, the machine sits behind glass, the brass gleaming, a wonderful example of Victorian engineering and, in its own way, a work of art. It is a fitting tribute to the man who "taught wheels to think" and whose failures were only failures of achievement and not of vision.