Dead At 92, Business Computing Pioneer David Caminer

Brooklyn Bob points out this fascinating obituary of David Caminer, the first systems analyst. "The tea company he worked for developed their own hardware and software — in 1951! Quoting New Scientist: 'In today's terms it would be like hearing that Pizza Hut had developed a new generation of microprocessor, or McDonald's had invented the Internet.'"

Comments

Daily Telegraph - same story, no registration reqd

[Anonymous Coward]

http://www.telegraph.co.uk/news/obituaries/2188963/David-Caminer.html

Re:McDonalds?

[Free the Cowards]

I don't think you actually understood that comparison.

They're not saying that it's like McDonald's inventing the McNugget. They're saying that it would be like McDonald's, the fast food company, inventing a computer from scratch.

This would be an American article then...

[The Famous Druid]

From TFA: So it was only natural it would look at the electronic brains that scientists in the United States were developing for scientific and military purposes as a way to streamline its own empire

Why do Americans have this urge to claim the credit for everything?

The Germans built a computer during WWII, and the brits built Colossus computers to break German codes. The University of Manchester built their first computer in 1948, and another in 1949, even the aussies had built CSIRAC in 1949, two years before LEO, and yet the NY times has to claim the LEO was based on what 'American Scientists' were doing.

There's a whole big world out there, and America doesn't have a monopoly on innovation.

Deal with it.

Re:This would be an American article then...

[Anonymous Coward]

As far as I recall the history, they didn't ask Americans anything. They were examining business methods round the world, and had discussions with other businessmen - both in America and Europe - as a matter of course. Computers (or Electronic Brains!) were being thought about at the time, and Lyons staff wrote a report saying that they should be investigated.

So a meeting was held with Maurice Wilkes of Cambridge, and the upshot was that Lyons sponsored the manufacture of the first commercially designed computer (and, more importantly, the first Business and System Analysts). There was no particular pressure or direction from any other company or country.

Oh, and another error - Lyons was NOT a tea company. It was a chain of restaurants, placed in city centres; they were called 'Lyons Corner Houses' because Joe Lyons, the owner, figured that a corner position got trade from two streets simultaneously. They typically served the office lunchtime trade - their waitresses were known as 'Nippies', because of their fabled speed of service. Tea would have been served, or coffee, and cakes, sandwiches or light meals. It's like calling McDonalds a Dairy Farmer because they serve milk shakes....

   

Re:This would be an American article then...

[Rostin]

Point taken, but FYI:

"J. Lyons and Co., one of the UK's leading catering and food manufacturing companies in the first half of the 20th century, sent two of its senior managers to the USA in 1947 to look at new business methods developed during the Second World War. During their visit they came across digital computers then used exclusively for engineering and mathematical computations. They saw the potential of computers to help solve the problem of administering a major business enterprise."

http://en.wikipedia.org/wiki/LEO_(computer)

The NY Times claim is stronger and more arrogant than is really warranted, but (assuming Wikipedia is accurate) it does seem to have some basis in reality.

Re:This would be an American article then...

[mangu]

Why do Americans have this urge to claim the credit for everything?

People in most nations seem to have this urge. Brazilians claim the airplane was invented by a Brazilian and Italians claim the telephone was invented by an Italian.

When you consider a "computer" as a generic machine capable of performing calculations, maybe it could be claimed the Greeks did it, but if you limit your definition to an electronic equipment doing calculations by binary logic, then it's true, an American has the earliest claim.

Re:This would be an American article then...

[aproposofwhat]

As a matter of fact, Lyons sent Oliver Standingford and Raymond Thompson over to the US in the summer of 1947, to meet with Herman Goldstine at Princeton - it was Goldstine who recommended that they visit Professor Hartree (Wilkes' boss) at Cambridge.

There's a damn fine history of the LEO computers, written with input from Caminer himself: A Computer Called Leo, by Georgina Ferry.

I just dug out my copy to get all the names right :o)

Re:This would be an American article then...

[Anonymous Coward]

An 'electronic machine doing binary logic' is the WORST definition of a computer I ever heard of. You might almost think it was specified simply in order to shoe-horn and American (/Bulgarian!) into the frame!

Atanasoff simply re-created Babbage's Difference Engine in electronics. So I would still go with the British as the first 'computer' inventors - Babbage and Turing between them defined the concept of the first general purpose machine which worked numerically. Adding electronics is not really an 'invention', it's a development...

As an aside, I don't think that anyone thinks that the Americans 'invented' the first airplane, either. That would be Sir George Cayley. What the Wrights did, by a short head, was BUILD the first CONTROLLABLE airplane. The principles of heavier-than-air flight had been known for 100 years by 1905. The key things they did were use a light frame, a powerful engine and a 3-D control system. Other people across the world were doing exactly the same independently, but they (just) got in the air first...

Interestingly, their control system, though functional, was not practical. It was the devil to use, and would not scale. So the Wrights were first, but a dead end - much like John Logie Baird with television. And similarly, they tried to keep their 'lead' by legal means - so successfully that in the US they suppressed all aircraft development, and when WW1 came we had to buy our fighter aircraft from the French, because we had no air industry of our own!

Santos Dumont, the Brazillian, is a much more attractive inventor of the aircraft. He was the first to fly an aircraft which needed no ground-based take-off assistance, (the basis of the Brazillian claim) his aircraft did scale, and the world's aircraft industry has developed from his designs rather than the Wrights, in part becuse he made them freely available for the benefit of mankind. He was Linus to the Wright's Bill Gates....

Re:What on earth would they do with this computer?

[jd]

The tea industry was so big at one point that it was profitable to build an entire class of ship specifically for tea and nothing else. Lyons deals with all kinds of commodities, many perishable, so high-power optimization was viable. As for "glacially slow", Colossus may have been slow per calculation but performed thousands of calculations in parallel and in benchtests compared favourably with a Pentium doing the same work. Early computers could, if built well, be damn fast and there are still problems where an analogue computer will outperform a digital computer at the same task.

Re:What on earth would they do with this computer?

[Scrameustache]

What sort of calculations could possibly be worth the expense of building an early computer to do them with? That's one thing I have wondered about : these machines had about as much memory as a sheet of notebook paper, and were glacially slow at calculations. What kind of tasks could be worth the expense of building one?

FTFA: millions of daily transactions

Tea and bombs

[Dynamoo]

It's not about tea - but as the New Scientist says, the exact equivalent to Lyons is something like Pizza hut. Lyons were the absolute masters of logistics in their time - they ran a huge network of outlets to a consistent quality with a very large turnover. So, they were really an ideal company to experiment with this new technology. Lyon's logistical expertise was such that during the Second World War they ran one of the largest bomb making factories in the world, just a couple of miles from where I live. One in seven bombs dropped on Germany came from the Lyons factory at Elstow.

Drool Britannia!

[fm6]

And never managed to maintain the loyalty of their colonies and ended up losing them all.

Another nitpick: LEOs were not exactly mini. See the pictures on this enthusiasts web site.

And we've been here before.

Re:Nah ah!

[carlzum]

Um, pretty much every source of economic data. Take a look at the US Census data since 1980. Total manufacturing output in the 2000's is several times greater. As population grows the number of things made follows. It's not of an indicator of economic health, but the US definitely makes more crap today than it did in 1980.

Re:What on earth would they do with this computer?

[Tablizer]

Here's some links to the IBM mechanical business machines:

http://www.columbia.edu/acis/history/tabulator.html

http://www.columbia.edu/acis/history/407.html

With successive stages of punched-card processing, fairly complex calculations could be made. One could roughly think of each stage as an SQL clause: SELECT (filter columns), then WHERE (filter cards, or "rows"), then maybe a GROUP BY, then a SORT BY, and then perhaps feed those back to another set of SELECT and WHERE cycles again if needed. Still, a human operator usually had to store, load, and monitor the various card stacks over each stage.
       

Re:What on earth would they do with this computer?

[aproposofwhat]

LEO was actually used to calculate and print the income tax tables for the British Inland Revenue in 1955 - the task was completed overnight, as opposed to taking several weeks if done manually.

Not only did Lyons build the first industrial computer, they even had a bureau service running as soon as the machine was ready to take on the extra work.