THE YEAR 2000 CRISIS
12:00 AM, Sep 23, 1996 • By LAWRENCE J. SISKIND
As the Year 2000 approaches, so does an epic collision with technological shortsightedness. The impending disaster is already a hot topic in the world of high tech, but more of us should be paying attention. A coming computer crisis -- call it the Year 2000 Crisis -- says a great deal about the values and philosophy of our society in the waning days of the millennium.
Thousands of computer programs are fast approaching obsolescence. Most date- oriented software -- and that includes programs running everything from pension funds to burglar alarms to personal finances to elevators -- is designed to recognize just six digits: two each for the year, month, and day. Thanksgiving may be November 28, 1996, to you and me, but to a computer, it is 961128. The software running the computer automatically assumes that the first two digits of the year are 19.
At the stroke of midnight on Friday, December 31, 1999, the date will become 000101. Computers all over the world will continue to assume that the first two digits of the new year are 19. They will read the new date as January 1, 1900.
If you were born in 1960, a computer today would figure you as 36 years old by subtracting 60 from 96. In the Year 2000, however, that same computer will subtract 60 from 00 and come out with -60. It will then discard the minus sign as meaningless and conclude that you are 60.
A program designed to sort dates will sort the following dates in the right order: 1901, 1950, 1999. But in 2000, the program will focus on the last two digits and sort them in this order: 2000, 1901, 1950, 1999. A moment after midnight, the world's computers are programmed to start making mistakes. Telephone conversations begun just a few minutes earlier will be computed as having lasted 99 years, generating humongous bills. Hospital pharmacies will automatically lock up as their computers conclude that all prescriptions expired generations ago. Checks dated 1900 will be printed for distribution to shareholders and pension-fund recipients, checks impossible to cash. Air travel will shut down as computers try to schedule century-old flights. Employees will be locked out of security areas. Elevators will stop running. And senior citizens will receive notices from their communities announcing the time and place for their kindergarten enrollment.
Of course, the fact that these errors are programmed to occur does not mean they will actually happen. Very, very late in the game, the world is waking up and taking steps to avert it.
The steps will be very expensive and very painful. Between now and the Year 2000, the world will spend $ 600 billion to engender Year 2000 compliance among its computers systems, according to the Gartner Group, a Connecticut- based information-technology consulting firm. The U.S. government, with its thousands of programs scattered over hundreds of agencies, will spend about $ 30 billion to reprogram its computers. California's 122 agencies operate over 100,000 separate computers. The state does not yet know how much it will have to spend. Its Department of Information Technology is spending $ 4 million just to oversee the efforts of the other 121 state agencies. Even Nebraska plans to spend $ 24 million to address the problem and has earmarked 2 cents of its cigarette tax for four years to pay for it.
The cost of correction for the private sector will be enormous. According to estimates by the Orr Institute and Data Dimensions, Inc., Fortune 50 companies will spend 35 to 40 cents per line of code on their operating programs. That could amount to between $ 50 million and $ 100 million per company. Companies unprepared to meet the challenge will fail. The Gartner Group predicts that 20 percent of the world's companies will go out of business because of the Year 2000 crisis.
Law firms are gearing up for the inevitable tidal wave of millennial litigation. Software producers will be sued by software users, who will be sued by their customers, who will be sued in turn by their customers. All of them face the threat of suit by their shareholders, and all of them will consider suing their insurance companies when those insurers refuse to provide coverage for all the other lawsuits already described. Auditors and law firms will probably also be named as defendants for good measure.
The governments and major corporations of this interdependent world understand that they will sink or swim together through this trial. An entity with Year-2000-compliant software will still face ruin if its customers, suppliers, or distributors are unable to operate the software. To avert, or at least minimize, the disaster, governments and companies are pooling information on how to rework and rewrite their systems. Much of the information-sharing takes place on an Internet site named "y2k." This titanic exercise in global cooperation among so many disparate and often competing entities brings to mind the rallying of the world's surviving air forces for the final coordinated battle in the movie Independence Day.
The progression of the calendar is as regular and incorrigible as an undoctored odometer. It doesn't take a village to teach a child that Tuesday follows Monday, that July follows June, and that the Year 2000 follows the Year 1999.
That being so, why all this frenzy? How did so much software containing such patent shortcomings get into so many hands so near to the date of obsolescence?
When the first primitive computer programs went into operation to handle payroll in the 1950s, the date problem seemed understandable. No one expected these programs to be running in 2000. But as the years progressed, and as the turn of the century drew nearer, why did code writers continue to incorporate the twodigit date field? Why, in the face of imminent inevitable failure, did the computer industry continue for so long its march of folly?
A number of explanations are circulating. One is that the industry is long on code writers and short on code designers. The writers, sometimes called " techno-twits," tend to focus on the narrow problem at hand, without regard for the overall structure of the system. One expert has analogized the industry to pre1600 Europe, when bridges and even cathedrals were constructed by craftsmen working without plans. Some of those bridges and cathedrals stood for a long time. Most did not.
Another explanation arises from the special time pressures on the computer business. There is usually an immediate need to get a system up and running, coupled with an understanding that there will be plenty of time down the road for refinement and corrections.
A third explanation deals with the pace of computer-technology innovation. Hardware and software have improved so rapidly over the past generation that no one expects his product to last long. Even as the window of time has narrowed, programmers still could not believe that their handiwork would be around in 2000. Adapting Lord Keynes's axiom to the computer industry, they believed that in the long run we'd all be updated.
All these explanations share a common element: remarkable shortsightedness. The crisis has not developed in secrecy. The inherent flaws of the two-digit date field were public knowledge from the beginning. But all the parties involved -- and that includes not only the designers and writers, but also their corporate and government customers -- seemed constitutionally incapable of viewing the situation beyond the range of the moment. Behaviorists are familiar with the phenomenon known as mass hysteria. The Year 2000 crisis has followed a period of mass myopia.
This myopia has other consequences in the industry. One program designer contends that the Year 2000 problem is not unique. It is one of many widespread program flaws, charmingly known in the trade as "cluster-f--s." He points to the upcoming exhaustion of available Internet Protocol addresses (we will run out by 2009) as another example.
More significantly, the myopia is not confined to the computer industry. We are approaching other collisions with shortsightedness in Social Security, Medicare, education, and other public matters. They may not occur as soon. The Social Security trust fund will not run out until 2029, according to the bipartisan commission that "fixed" the problem. But these collisions, these social cluster-f--s, will occur. We approach them with our eyes wide open, knowing that disaster is inevitable if we do not change course -- and yet we do not change course.
The Year 2000 crisis is not so much technological as it is cultural. It is the product of a generation incapable of vision beyond the range of the moment, a generation endlessly encouraged to "Just Do It." A generation whose object is immediate gratification, and whose slogan is: The Future is Not.
It was not always so. On the eve of the 20th century the world had a different view of the future. In The Great Illusion, Oron Hale wrote of the long-range optimism pervading society: "The sense of a boundless future was strong among all Western leaders in 1900, and it brought zest and assurance to the cities and cultures of men. It was a time of incredible innocence: War was unlikely; social reform was every man's duty; and progress was inevitable." In such an age, men thought and created for the long run.
In Boston, site of the nation's oldest subway system, there is a tunnel connecting downtown to East Boston. It was designed in 1897 and opened for traffic in the spring of 1900. The tunnel was constructed by immigrant laborers from Ireland, Italy, and Eastern Europe. The men worked hard for low wages so that their descendants could have a better life. Some of those descendants got into Harvard and MIT across the Charles River. Some became computer programmers, and they ride through the tunnel today on their way to Logan Airport. More than likely, it brings the builders no comfort in heaven to know that their handiwork will outlast that of their great-grandchildren.
Lawrence J. Siskind is a San Francisco attorney specializing in intellectual property law.