{The following is an easily-understood explanation of how the two-digit year problem came about.]

From Connected (The weekly technology and science download from Electronic Telegraph), Text-only version, 14 January 1999

The search for 'Amazing Grace'

Grace Hopper is the woman who in 1952 devised the programming language that enabled the computer revolution to begin. Her legacy, though, included the Millennium bug that could bring much of modern life to a standstill in one year's time. Robert Matthews reports

WHEN Angela and Jeremy Perron told their friends about their plans to move up to the far north of Scotland, they spoke of wanting to leave behind noisy, over-crowded England and about giving their children a chance to live a different sort of life.

Now, six months after setting up home in a remote part of the Moray Firth, they are becoming more open about what really drove them so far from the rest of civilisation.

This time next year, they believe, the rest of civilisation is going to be in a terrible mess: electricity supplies down, lighting and heating gone, phones dead. Food and water will be scarce, and law and order pushed to the brink of collapse.

It sounds like the standard, off-the-shelf, apocalyptic vision of some barmy religious cult. The thing is, the Perrons are perfectly normal, well-educated people. Angela runs a publishing business and Jeremy is a computer programmer. Their fears are based on something once dismissed as a scare story that now looks set to be the dominant issue of this year.

[Explanation of Y2K problem's possible effects.]

Precisely what will happen, not even the experts can say. What is clear is that the action of people such as the Perrons is no longer looking so rash. Portrayed chiefly as a boringly technical glitch for bean-counters to fret about, the Millennium bug problem is now prompting experts to reveal their own personal contingency plans for coping with the likely chaos. And in their survivalist overtones, they are strikingly similar to those of the Perrons.

Dr Ross Anderson, a leading Millennium bug expert at Cambridge University's Computer Security Research Centre, said: "Personally, I plan to have three months' food, a working well, three tons of propane gas and 400 litres of diesel."

In America, Senator Robert Bennett from Utah, widely seen as the most authoritative voice on the Millennium bug on Capitol Hill, is now advising others to start building up supplies of food and water. The Federal Reserve Board is planning to inject an extra $50 billion into the US economy to avoid a liquidity crisis caused by massive withdrawals of cash towards the end of the year.

In Britain, the Government has given the task of alerting and advising industry on the Millennium bug to a body called Action 2000. On a television programme tonight, Apocalypse When? (Channel 4, 6.30), Gwynneth Flower, director of Action 2000, voices her concern about panic buying as the date approaches. She is worried that there could be a rush on essential supplies and consequent shortages next December.

Such pronouncements reflect the growing acceptance that the Y2K problem cannot be cured - just mitigated. Despite the huge sums spent by government and commercial agencies - British Telecom's Y2K effort has cost #300 million so far, and Action 2000 is spending #10 million on a publicity campaign for smaller businesses - there are just too many programs and too many microchips to be fixed: according to one estimate, 12,000 billion lines of computer code, and 30 billion microchips.

How did we ever get in such a mess? The answer lies in a headlong race that began 50 years ago, and now looks set to fly right off the rails: the race to exploit the power of the computer.

Until the late 1940s, even adding up a shopping bill with a computer required PhD mathematicians and a small army of engineers. Programming the earliest computers was, according to computing historian Stan Augarten: "A one-way ticket to the madhouse." Hundreds of wires had to be plugged in to the correct sockets, and thousands of switches set. It took technicians days just to set up the machines correctly.

And then along came Grace Murray Hopper, a computer specialist with the US Navy and one of the most colourful of the many eccentric characters in the history of computing. Born in 1906, Hopper was a native New Yorker with a PhD in mathematics and a wit and intellect to match. She worked into her eighties and well-earned the sobriquet "Amazing Grace". She claimed she always preferred to work with women because "they finish up things and men don't".

Her office featured a skull and crossbones flag and a wall-clock that ran backwards, because she thought people needed to be reminded to think flexibly. She also had an egalitarian's desire to make the power of the computer accessible to others.

In 1952, Hopper came up with a crucial invention: the "compiler", a set of instructions allowing computers to be given orders written in standard language, which were then automatically translated into the computer's own esoteric code.

Hopper's compiler was the bridge that opened up the power of the computer beyond the small society of mathematically gifted specialists. Known as Flow-matic, her compiler opened the way for the development of Cobol - "Common business-oriented language" - which allowed companies to use computers to handle routine tasks such as pay-roll and accounts. Her work also signalled the start of the scramble to exploit that power - a scramble in which the limitations of early computer hardware had to be overcome by all and any means.

And chief of these limitations was memory. Despite the giant size of the early computers, the primitive machines could not store much data in their electronic "brains", compelling the programmers to come up with clever ways of cramming quarts of data into pint pots of memory. Constantly searching for memory-saving tricks, programmers could find little scope for paring down dates: an absolute minimum of six digits - two each for the day, month and year - were needed to specify any day in the 20th century. Ideally, two more were needed to specify the century - 19 for the current one, 20 for the next. But back in the 1950s, it seemed senseless wasting valuable memory space on a theoretical problem 40 years down the line.

Simply storing all the data used by these early behemoths gave rise to similar problems: whole warehouses were taken up with vast heaps of punch-cards carrying the information needed by a company's computer. With so much physical space taken up with data storage, it made sense 40 years ago to keep the date code in the most compact form possible.

The memory problems did indeed ease as the years passed. Crucially, however, the omission of the century digits from dates gradually hardened into an industry convention.

Warnings about the existence of this subtle time-bomb within the millions of computers throughout the world started to emerge in the late 1970s. Bob Bemer, widely acknowledged as one of the fathers of modern computing, was among the first to sound the alarm bell. "I first recognised the Year 2000 problem in 1979," he recalls. "I warned the industry that a two-digit approach would have dire consequences in the future. My efforts at advocating a four-digit approach fell on deaf ears."

Peter de Jager, a US computer consultant, recalls facing similar complacency. "In 1977 I started as an operator with IBM, and on my first day on the job I noticed the systems were using two-digit years," he recalls. "When I brought it up with management, they said don't worry about it, nothing's going to happen for another 23 years, so relax. I did."

Then de Jager heard of how millions of Americans had been plunged into darkness when a single switch failed at a power station. "It made me realise that, in 2000, these dates would have a similar impact." Once disregarded as a doomsayer, de Jager is now advising the British and American governments on how to tackle Y2K.

The biggest obstacle is not the technical difficulty of dealing with the Y2K problem, but its sheer size. The task has been likened to changing every bolt and rivet in all the world's bridges. In the face of such a mind-boggling task, many of those likely to fall foul of Y2K are acting like rabbits caught in the headlights of a juggernaut. Research by Action 2000 shows that most companies and local authorities have done nothing to deal with Y2K.

Even those who are taking action acknowledge that fixing even 99.99 per cent of the problem may not be good enough. This is because of the "weakest link effect", in which a vast network of interconnected systems can break down through the failure of one small, yet critical, component.

In February last year, the failure of a few cables caused weeks of power cuts and disruption in Auckland, New Zealand's largest city. In May, more than 80 per cent of America's pagers, as well as broadcast, information and credit-card processing services, were knocked out by the malfunction of a single computer aboard one communications satellite. And last month, 330,000 people in the San Francisco region were left without power following the failure of a single sub-station.

While few experts foresee what is being called TEOTWAWKI - The End Of The World As We Know It, the impossibility of catching every single significant Y2K problem is prompting many to predict at least several weeks of disruption.

One American estimate predicts that between five and 75 per cent of the population will be left without power for a significant amount of time during the early part of next year. A year ago, the strategy of the Perrons in their Moray Firth refuge would have seemed ludicrously alarmist. A year from now, it may prove to have been remarkably perspicacious.

Cut and pasted by

-- Old Git (anon@spamproblems.com), February 02, 1999

Answers

Old Git, thanks for posting this! Doesn't it feel great when we lunasnickers are seen as perspicacious visionaries :^J

xxxxxxx xxxxxxx xxxxxxx

-- Leska (allaha@earthlink.net), February 02, 1999.

"Lunasnicker"? Hmmm. Must be a Canadian fruitcake. Yes, it was reassuring to read the Telegraph articles and know that my fellow, unflappable Brits of legend are as lunasnickery as we are over on this side of the pond. Granted, we do eccentrics better than anyone (like the retired Admiral who lived near the Greenwich Observatory and fired off the cannon on his roof at noon every day), but the people interviewed for this and the other article sounded fairly sane and ordinary.

-- Old Git (anon@spamproblems.com), February 02, 1999.

Good one Old Git.

"... growing acceptance that the Y2K problem cannot be cured - just mitigated."

Yep. That's how global Y2K awareness IS growing.

Diane *Sigh*

-- Diane J. Squire (sacredspaces@yahoo.com), February 02, 1999.

To Old Git,

Thanks for the article I have faxed it to my in-laws in the UK.

Thank you, thank you, thank you.

-- Carol (usa-uk@email.msn.com), February 02, 1999.

A better analogy would be "checking every bolt and rivet in all the world's bridges" - many programs are okay, but since every bolt and rivet has been painted many times the past years, and since approximately 3 out of every 100 rivets (embedded chips) is made of tissue paper, and 3 out of 5 plates in the bridge (the business and government management programs) is made out of toilet paper instead of high strength steel, every rivet and plate have to be inspected. Many have to be replaced or upgraded.

If you paint a bridge made out of toilet paper, it may look good, but it is still a bridge made out of toilet paper.

-- Robert A. Cook, PE (Kennesaw, GA) (cook.r@csaatl.com), February 02, 1999.

Robert:

Hear hear...

-- Grrr (grrr@grrr.net), February 02, 1999.