An interesting perspective by Alistair Cooke on Y2k attitudes in the States just before the new year. This is for the archives.

http://news.bbc.co.uk/hi/english/world/letter_from_america/newsid_597000/597280.stm

LINK

Monday, 10 January, 2000, 10:55 GMT

On Y2K alert

Notice from the landlord to all tenants: "In anticipation of any potential Y2K problems the four elevators in your building will be brought down to the lobby and kept out of service for 30 minutes beginning at 11.45pm on December 31st, 1999."

If ever there was a talk that ought to be heard the moment it's being spoken this is the one.

I have a problem which has never come up in 53 years and plainly will never come up again.

I'd better begin explaining it by saying that I'm recording these words on Wednesday December 29 in deference to - how shall I put this delicately? - in deference to the fairly recent European custom of stretching what used to be a two-day holiday into something more like a two-week break.

As I understand it many millions of people in Europe, including Britain, will be taking, or have taken things a little easier than usual on Friday - New Year's Eve - perhaps even leaving work early.

When I mentioned this to an American friend who happens to be, as we say, in broadcasting, he reeled visibly in disbelief: "You've got to be kidding."

I think that would have been the reaction of most other Americans simply because on Friday millions of Americans stayed at work or came on to an all-night shift, from the whole federal government down to every kind of office worker - people who work in hospitals, airports, banks, factories and companies that have never before been open through New Year's Eve into the dawn of New Year's Day - all of them on, of course, a Y2K alert.

The suspense that built up in the weeks before New Year's Eve was nerve shattering, scary, also often funny, and I had meant to recite the drama of it - not to mention the nationwide fear of terrorism - until I realised that it would be like retelling a gripping whodunnit when you already knew who dunnit - you being the majority everywhere that hears this talk on Sunday or Monday.

By now you'll know the worst and the best. Either chaos has come to Russia and millions of small businesses in Latin America and the Third World - possibly even to little Piddletrenthide and Tombstone, Arizona - or nothing at all happened and the wise guys - that's to say the knowalls who know absolutely nothing about the history of the computer - are going to be able to chant: "I told you so. Tremendous fuss about nothing."

Well as I must remind you I'm talking more than 48 hours before the great revelation.

If all is well and a whirlwind sigh of relief passes throughout our world there is one human being more than another to whom we should give the credit. He will deserve the Nobel Prize for Peace, for Physics and perhaps a newly created one - the Nobel Prize for Survival.

It was a United States senator, one Robert Bennett, from whom we first learned of a technical glitch that any computer mechanic could fix but which, after deeper thought, turned into what one expert called "the most ominous threat to human life since the discovery that e=mc2."

Before we come to our blessed watchdog I'd like to go back to the experts' first awareness of the problem and the recollections which have only now been told.

The problem first occurred to a young expert on systems development about, he recalls, 1978. In a jokey moment with some colleagues at a time when there was a subdued but general fear about the possibility of a nuclear war one man said: "Just imagine: the world is going to end when somebody presses a button."

The systems development man, one Richard Reagan, said in response: "No, the world's going to end when the clock ticks over to 2000 and all the computers will erase all the data."

Now this was not shouted aloud to the multitude as a threat to anybody it was more or less of an in joke - shop talk.

Mr Reagan says: "We could have fixed it then but it would have meant re-creating all our customers' existing programs. And at that time no responsible programmer would have been anticipating a problem 22 years away. "Also at that time there were not too many computers and they had a large number of uses."

Another systems developer touched the crux of the matter: "The problem of Y2K isn't an internal computer problem, it's in the connection between business computers and the reliance on data transfers."

That's about as simply as it can be put.

I think the point for lay persons to appreciate is that even way back in the 70s to fix Y2K would have meant recalling all the private computers or junking them and creating new ones.

"The main point," says another expert, "is none of us was so arrogant as to think any of the code we were writing then was going to be around in 20, 30 years."

That prospect of making all the private computers obsolete at a stroke reminds me of a trip I took about 50 years ago into the Adirondack mountains in upstate New York.

In an old phonograph store - in England, a gramophone shop - the owner - an old man - beckoned me to a phonograph and played me a record. It was a Beethoven symphony.

The recording was tinny but the orchestra was on pitch but the record instead of spinning around at the usual 78 revolutions a minute was crawling.

"We thought," said the old man, "everybody would soon have one of these at the time."

It was what would come to be called 'a long playing record'.

Now the point of this harmless story is the machine the old man played to me had two switches and a fine grooved record. They had been manufactured in 1906 but because everybody would have to buy a new player - the idea of a cheap attachment didn't occur - and somehow come by these fine grooved records, the manufacture of two switch machines was abandoned and the patents or patents on long playing records were frozen for 40 years.

The old man was right. By the early 1950s only the standard universal record playing 78 rpms was dead.

It is a crude analogy but I hope it's a useful one and stresses the extreme unlikelihood of any computer expert in 1978 guessing that the code of 1999 computers would not be able to cope with that almost comical hypothesis of the clock rollover problem of 2000AD.

All this about the first gleanings of the problem has come out only in the past few weeks.

But when did we wake up to the imminence of the rollover as a menace to the whole world?

That takes us to the summer of 1997 when Senator Robert Bennett of the State of Utah called into session a senate sub-committee of which he was the chairman: the senate banking sub-committee on technology and financial services. Senator Bennett brought up, for the first time in public, the strange puzzle of the year 2000.

I don't believe any of us had heard of a disturbing episode that happened that spring in the warehouse of a frozen food factory.

A supervisor going his rounds happened to notice something he'd never seen before: the expiration date of a shipment of food that was just going out said that the food was already 40 years old - an absurdity since 40 years ago the firm didn't exist.

The supervisor consulted his boss and it was decided, considering the one chance in a hundred, say, that the printed note might be right, they dumped the whole shipment.

Senator Bennett had heard about this and other freaks and he knew then why this had happened. Mainframe computers more than 10 years old, which most around the world were, had not been programmed to handle a four-digit year - as in 2000 AD.

To make an immense saving in storage space in all the memory banks, two digits only had been allocated to any given year, so 1978 was 78, 1993, 93, so on. Same with chips and software.

So when 1999 was to turn over to 2000 the chips and the software would go back to 1900.

Senator Bennett presented to his sub-committee a necessarily blunt and graphic picture of what this could mean.

Government services everywhere could be partially paralysed, the electric grid of towns and cities out, no light, no power, no water supply, telephones, railroad traffic is computerised, food transport stopped, hospitals back to the most primitive care.

And - especially troublesome to this banking committee - the international banking system crippled. Maybe you'd wake up, in the dark, to find among other annoyances that your bank deposits registered zero.

Senator Bennett went on holding hearings and he woke up the world in June of last year.

"It is not a computer problem simply," he said, "it is the responsibility of management of businesses big and small around the world to reprogram their product to fix things now.

"When people say to me is the world coming to an end I say I don't know whether this will be a bad bump in the road - the most optimistic assessment - or whether it will trigger a worldwide recession with absolutely devastating economic and social consequences."

After that committee rose the government of the United States, beginning with the White House and then on to both houses of Congress, redoubled their work on fixing, as we put it, the Y2K bug.

All the allies were consulted to see how far along they were with the necessary work. Some weren't along at all.

Russia was the most notable laggard. And in no time the United States lent a couple of billion dollars to Mr Yeltsin to get busy killing the bug.

As of New Year's Eve the president assured us that as for the federal government anyway everything was A-OK. The work that had to be done in every country around the world was prodigious if January 1st 2000 was to be only a bump in the road.

If you are listening to me anywhere in comfort I can only say God bless your government, your businessmen and computer doctors, most of all Senator Bennett - and happy, happy New Year.

-- (for@the.archives), July 21, 2000