Slashdot Mirror
A Computer Called LEO
In the early mists of computing -- pre-WWII, during, and immediately afterwards -- only a few scientists were really aware of what a computer was or could be, and no-one considered a computer to be anything other than a scientific or military tool. Except one man, John Simmons, a progressive and enthusiastic manager for the Lyons tea-shop empire in Britain, who also happened to be a brilliant mathematician and zealous proponent of the principles of scientific management.
Georgina Ferry tells the full story of how the young Simmons saw the need for automation as early as the 1930s. The monstrous task of accountants' clerks adding up copies of all the waitresses' bills for 250 tea-shops was done with mechanical calculators, and his dream was of removing this drudgery by automation.
He had seen the future of mechanical automation on a trip to the USA in the 20s, but it wasn't until after WWII that he was able to send two trusted lieutenants on an electronic fact-finding mission which included meeting Herman Goldstine, godfather of ENIAC, at Princeton. The resulting enthusiastic report, and a visit to Douglas Hartree at Cambridge, England, enabled Simmons to persuade the Board of Lyons to let him build a computer from scratch.
Post-war Britain had no dollars to buy American computers, but more tellingly, computers were viewed in the US and England by their scientific and mathematical fathers as tools of science. Simmons saw them as tools of business, and astonished them all by building one to do business processing.
The Lyons Electronic Office (LEO) was started in 1949 and entered service in 1951 with punched tape, mercury delay lines, and a program to analyze costs in the Bakery of the tea-shop business. It thus became the first purpose-designed business computer, years ahead of the first US business system (GE's 1954 UNIVAC).
It was so successful that Lyons set up a subsidiary to make and sell them to British industry. LEO spawned LEO II and eventually LEO III, which offered true multiprocessing. Sadly, British industry was slow to grasp the opportunity. Leo Computers had some notable and significant sales through the 50s and into the 60s, including winning the biggest commercial data-processing contract in Europe at the time (to the UK Post Office in 1964), but the Lyons Board eventually sold off their subsidiary, and it passed through mergers and acquisitions into ICL and oblivion, but that big PO contract was so successful that the Post Office persuaded ICL in 1969 to make five last LEO 326s which continued in service until 1981!
Ferry has managed to condense a 30-year technological saga into a thoroughly readable and hugely entertaining book without neglecting the underlying causes of Simmons' original quest to improve business efficiency. Her descriptions of the contributory threads of UK and US computer development are succinct and accurate, and they balance her careful explanations of the hugely complex world of running a large catering business manually, the complex interplay of family-business relationships, and the differences between UK and US commercial ethos in the post-war period.
At this distance in time, Ferry has been fortunate to have been able to include material verbatim from many of the people directly involved, so there is an air of immediacy which you don't get in books on earlier science. There's a full list of sources and a detailed index, and numerous photographs taken at the time. This all makes the book valuable on several levels, and it would make a great gift to anyone in business as well as computing.
Georgina Ferry is a science journalist and author, and has written accounts of scientific achievements in several fields. Recent contributions include a Life of the only woman Nobel laureate, and co-authorship of a book on the social and political aspects of work on the Genome. The BBC has a bio here. A Computer Called LEO: Lyons Tea Shops and the World's First Office Computer is available from Amazon UK. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
See http://www.picotech.com/applications/colossus.html for details of what British engineers achieved before ENIAC existed.
The real "Libtards" are the Libertarians!
Lyons being a food company apparently had many bad run ins with the UK government over rationing in WWII. This meant that hardly any computers were bought by the UK govt (apart from those purchased by the post office). By contrast IBM had a great relationship with the US govt before it got into digital/electronic computing (i.e back in the days of mechnical computing) and this fed through into a lot of demand for IBMs electronic computing offerings. Sad to say, another example of the UK govt managing to kill innovation stone dead and sabotage the economy. Willing to bet the dumb civil servant even managed a knighthood for his efforts.
The relationship of advanced calculating machines and tea was documented by this guy.
...and look up the definition for 'comeuppance'.
...wearing a skin-tight topless leather jumpsuit, with cutaway buttocks and transparent crotch panel.
LEO would have been a great success had John Carmack been around to write Quake for it. Of course the cost of the paper spewing out of the machine would have been prohibitive.
Trolling is a art,
...drink their tea straight. More of a caffeine hit, and it tastes better than coffee. And, you don't get that nasty twitch that Red Bull gives you.
Dude, where's my packet?
Oh for God's sake. Two words. Bletchley Park.
...
Still, if we're worthless, you won't want us to join in any of your military adventures, even if only to lend a respectable gloss to what you're up to. We actually had more of our boys killed by your lot than the Iraqis
The Colossus was an important achievement. But, like many inventions, it was not without predecessors in some form or another.
Ita erat quando hic adveni.
The LEO veterans have a web site with some interesting pics here.
Y'know, that Leo 1 desk looks awfully like the computer that terrorized Emma Peel in The House that Jack Built. Spooky!
I guess you never read enough history that the poles and brits have a huge head start on enigma decrypts using their mechanical computers (bombes) and developed the Colossus to break "fish" code. The americans and pre-NSA were children still playing with paper and pencil in those days. It wasn't until the brits and poles gave them everything on a silver platter that they even started to get a clue...
"It thus became the first purpose-designed business computer, years ahead of the first US business system (GE's 1954 UNIVAC)."
General Electric did not create the Univac, it was Mauchley and Eckert for Remington Rand.
BTW - My dad used to work for GE computers. I had a Multics terminal in my house as a kid. Learned how program that way.
Dang, I'm old...
Such stories are always a good read, and put things into perspective nicely. Here are a bunch of guys doing payroll on a machine that probably has less computing power than the battery in my laptop!
And here I sit, watching the secretary on her 2,4GHz P4... playing freakin' freecell.
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
It's somehow gratifying to read about how the product I'm using (my computer) and the product I'm drinking (an indian black tea) are connected in history in a way which neither coffee nor red bull can compare.
To set the record straight for those who call tea girly, no product which for its aquisition entailed the conquest and enslavement of an entire (sub)continent can be called girly... Even if I drink it out of cute little porcelain cups with my cute teapot. It's still manly. Arrrrghhh (manly grunt)
So. . does it say bad things about me that the second I see "LEO" I think. . .
Law Enforcement Officer?
Find out about my new childrens book: SS Death Camp Criminal Batallion Go To Monte Carlo For The Massacre
The Burroughs 5500 was a very similar stack machine from the late 1950s, but was more successful commercially. It was produced in quantity, had a good OS (the Master Control Program), and did the back-office work of many banks for a decade. It was even a shared-memory multiprocessor.
Somewhere in all the mess that surrounds LEO is George. George I, II & III. I think (usual provisos about senility apply), George I was the OS on later LEO's. ICL ploughed on with this OS for years. In 1981 I was coding Cobol at ICL on a rare 2966, on VME-B, being hosted on George III - a feat common on IBM machines but less so on ICL.
George-III had some multi-user features, but mostly they used it just for multi-processing.
VME-K and VME-B which were meant to be new OS's inherited quite a bit of George and were still in use till at least the late-80's on 2900 and 3900 series ICL kit.
Imagine MS-DOS at 40 years old? Eyyyyhhhh!
My father (who was a LEO programmer in the 60s) told me of the 'drum drive', precursor of the disk drive. The principle was the same, but the 'platter' was a cylinder, which had the benefit of constant track length, but the drawback of being similar in form factor, noise level and maufacturing tolerances to a washing machine.
If I recall correctly, they managed to cram ana amazing 50k of data onto a drum.
A pizza of radius z and thickness a has a volume of pi z z a
I have a mercury delay line transmitter and receiver (but not the mercory-filled tube which used to fit between them) from a LEO mark 2, although I don't know which machine they came out of. I have a power supply unit (one of many in the original machine) from The Corby Steelworks LEO Mark III machine.
I'm old enough to remember when discussions on Slashdot were well informed.
Magnetic drums weren't used for quite the same purposes as disks, though. Disks were for file storage, but drums were more often used as a low speed high-density working memory, similar to a modern virtual memory swap device. Running programs were swapped between core memory and the drum.
Before magnetic core memory became cheap enough to use in low-cost computers, some people even built low-end computer systems with a magnetic drum as the only working memory. Such computers were much slower than computers with electronic memory, since you might have to wait an entire rotational delay for a desired memory word to come under the read heads. Clever programmers arranged the instructions and data on the drum memory to minimize this delay.