Sunday, December 26, 1999
Y2K Programmer says he warned us
Retiree stocking food, candles
BY PHIL WAGA
Westchester Journal News
DALLAS If a meltdown of the world's technologies ignites Armageddon at the stroke of Jan. 1, or if your blender sputters, think about Robert Bemer, the father of Y2K.
The IBM Corp. worker developed a feature for a computer language in the 1950s that allowed programmers to use either a two-digit or four-digit field to record years.
To save money and computer space and in spite of his impassioned objections a two-digit field was adopted across a world that in eight days will hold its breath in hopes that computers don't erroneously read the year 2000 as 1900 and trigger chaos.
However, computer memory was so costly and precious decades ago that it would have almost been irresponsible and technologically impossible to have used a four-digit year field, many analysts say.
I never thought my work would come to this, said Mr. Bemer, now 79 and retired after a long career as a top computer programmer. He said that he and other computer experts warned of the potential pitfalls of the two-digit shortcut, but computer executives and government leaders did not give a hoot about a millennium that was then decades off.
So there's plenty of blame to go around, but I don't deserve any of it, he said.
Mr. Bemer said he doubts that he faces any great harm, but he's not taking any chances. A year ago he moved with his wife to a secluded home in Texas. The place is outside a town with fewer than 500 residents, and 100 miles from Dallas. The nearest traffic light is 14 miles away.
I just don't want kooks bothering me, he said. It's a natural phenomenon that people lash out at anyone they think is responsible for a problem, and they may think it's me with Y2K.
He predicts a host of computer glitches that will disrupt life, but not end it. He has stockpiled water and food, candles and flashlights.
Although he developed the program, he fought hard against two-digit years from the late 1950s through most of the '60s. But he ran into one wall after another.
The Defense Department, the world's biggest computer user and busy fighting the
Vietnam War, said its keyboard operators could not be bothered to type in four digits for years if two would do. Big business, mindful of government contracts, didn't say a word. His appeals twice reached the White House.
Mr. Bemer's account of his last appeal to the White House has been reported before. In 1970, he rounded up nearly 90 top scientists and prestigious technical associations to argue the case with Edward David, President Richard Nixon's science adviser.
Mr. David agreed with Mr. Bemer, and appealed the matter directly to Mr. Nixon.
Rather than acting on Mr. David's information, Mr. Nixon asked him whether he could help in repairing his TV set.
So I don't think anyone in the world did more than me to get rid of this horrible two-digit field, Mr. Bemer says with frustration. But nobody would listen.
Mr. Bemer was not just any computer programmer.
As part of his work with two and four digits, he worked alongside Grace Murray Hopper, a computing legend. Together they developed a standard programming language called COBOL, for common business-oriented language.
He also developed a technology called ASCII, for American Standard Code for Information Interchange. It opened computing by letting machines communicate with one another. He also invented a technology that enabled laser printers to function.
But as hard as Mr. Bemer argued, the two-digit, four-digit issues could not be readily resolved.
Maybe technology giants should have eased in Year 2000 solutions over many years instead of just reaping the benefits of a computer revolution. But many companies wouldn't allocate revenue far in advance to disarm a computer bug that they often theorized had been fabricated by technology firms in search of new profit sources.
Peter de Jager, one of the world's leading experts on Y2K, has spoken publicly on the problem for nine years. He said audiences of corporate executives and government and agency leaders were unreceptive as recently as two to three years ago.
Everyone felt that fixing Y2K was not going to generate revenue, so why bother, the Toronto analyst said.
The clearest lesson to be learned might just be that an embrace of a new technology should be better thought through and done with a great deal more caution.
Look, I love the Internet, but you have to remember that technology is never as simple as it looks, and what you accept today may come back at you tomorrow, said John Searle, a professor of philosophy at the University of California at Berkeley.
Mr. Bemer has reams of reports and other papers to document how often he objected to the implementation of a two-digit year field to IBM and others. This is a computer disaster that should never have been, he said.
At the time, IBM was one of the few computer powers in an infant industry. In 1964, the company introduced the System 360 mainframe, which used a two-digit field and set the industry standard.
Frederick Brooks, an IBM mainframe executive during the 360 era, confirmed that Mr. Bemer warned IBM executives about the potential problems. But he said absolutely nothing else but a two- digit field could be adopted. He said typical 360 mainframes had 16 to 32 kilobytes of memory, and the operating system ran on 12 kilobytes. Any year field larger than two digits would have eaten enough memory to make the machines almost useless, according to Mr. Brooks.
Using anything then but a two-digit field would have been irresponsible and stupid, said Mr. Brooks, now 68 and a computer science professor at the University of North Carolina.
In recent years, corporations and institutions around the world have rewritten more than 800 billion lines of computer code at a cost of more than $300 billion in preparation for Y2K.
Although the history of the Y2K bug could be traced to IBM, a half dozen legal experts interviewed said the company is unlikely to be held legally accountable in court.
Congress in July passed sweeping legislation limiting corporate responsibility for Y2K problems. The bill capped punitive damages, narrowed the scope of potential class-action lawsuits and ensured that defendants would be held responsible only for the share of damages directly traceable to them in most cases.
Legal experts also said any case against IBM would be inherently weak because the technology of the 1950s left little choice but to adopt a two-digit field.
Certainly people may file suits against IBM if big problems develop from Y2K, but I don't think any of them will stick because they'll be judged from the perspective of the time when the two-digit field was developed, said David Miranda, a lawyer in Albany specializing in technology law.
Walter Effross, a professor at Washington College of Law, said though IBM may not be legally responsible, it could not absolve itself of societal responsibilities if significant Y2K difficulties develop. If there are problems from Y2K, you can be sure that fingers will be pointed at IBM, he said.
IBM declines to speak specifically about its historical involvement with Y2K. The evolution of the Y2K issue is bigger than any one company or the industry itself, said Glen Brandow, an IBM spokesman.
IBM did act more quickly and efficiently than many other technology companies in testing its software and machines, and telling consumers and client companies what is and is not Y2K complaint, said Stephanie Moore, an analyst at Giga Information Group, a market research company in Westport, Conn. IBM has been very straightforward with Y2K, she said.