On one sunny morning in the middle of June, the residents of Van Nuys, Calif., awoke to intimations that the shit was starting to hit the fan -- or in this case, the streets.
Four million gallons of raw sewage had spewed forth overnight from the Donald C. Tillman Water Reclamation plant, spilling onto a local thoroughfare and covering a large, wooded portion of the city's Lake Balboa Park. The problem? A computer undergoing Y2K tests had closed a gate and blocked a major sewage line from the San Fernando Valley, without notifying plant personnel at their control consoles what it had done. A couple of weeks later, on the opposite coast, the Washington Post featured a page 1 story in which city officials conceded that the District of Columbia was hopelessly behind in its Y2K preparations. The D.C. officials unveiled extraordinary contingency plans for the coming of the millennium: extra shifts for police, who would be stationed at 120 spots around the city and equipped with a back-up communications system; a network of "warming centers" -- in effect, mass shelters; school crossing guards at major intersections in case of traffic-light failure; and extra staffing at local emergency rooms.
Despite the apparently ominous portents of the Post story and the Van Nuys accident, neither got much attention outside their immediate locales, or even inside, perhaps. The Post story caught the tenor of public dialogue about Y2K computer troubles in paragraph five: "Officials are confident that most of these plans -- even those that will be put into effect regardless of any system failure -- will not be needed, and that even in the District, Y2K will be one of the century's most hyped nonevents." This disclaimer has become as obligatory a bit of boilerplate in Y2K coverage as the explanation that the problem began with two-digit date fields back in the days when computer memory was scarce. One can't write a "serious" Y2K story these days, it seems, without pausing to point out that it's really not much of a story at all.
For a month or two last winter, it looked like "the millennium bug" might be one of the major news beats of the year. The story first reached public consciousness in the summer of '98, with a handful of tongue-in-cheek, major-media color pieces about veteran and neophyte survivalists -- the latter including a striking number of computer programmers -- who were busy girding themselves against the coming digital apocalypse. It was all in good fun, the sort of story that serves subtly to reinforce the notion that anyone who doesn't think like you and me (like Time and Newsweek, in other words) is just plain nuts. But from there, public concern about Y2K began mounting gradually (excuse me, but did they say computer programmers?), and crested with the February release of a distressing U.S. Senate report.
In a cover letter distributed with draft copies of the Senate paper, Robert Bennett (R-Utah), chair of the Y2K special committee, wrote that public and private efforts to fix the problem "began late and remain insufficient, and consequently some incalculable level of economic disruption is inevitable. ... This problem will affect us all individually and collectively in very profound ways. ... [Y2K is] one of the most serious and potentially devastating events this nation has ever encountered. It deserves our top priority." If one bothered to read the report, the outlook became only more troubling: The state of preparedness across the many sectors of business and government was a mystery, the list of troublesome contingencies positively numbing.
The story then fell off the front pages even more suddenly than it had come, mainly a casualty of NATO's incursion into Kosovo and the blandishments of John Koskinen, the man selected by Bill Clinton to chair the President's Council on Year 2000 Conversion. There is nothing to fear but fear itself, the current administration line goes; avoid panic and hoarding, and everything will probably work itself out by year's end. Perhaps I exaggerate the degree of official complacency, but not by much. The Clinton PR strategy -- for now, anyway -- is to warn of the fabled "bump in the road," the "three-day winter storm," and to hope thereby to buy time for ... what? Could the prospect of millions of defective microchips and billions of lines of corrupted software code be so trivial a problem as it's made out to be -- something to be finessed away in a matter of months? And if it is not assuredly so, then what exactly has become of the Y2K story?
A couple of yarns may help to suggest where the answers to these two questions lie. (I'll take for granted, by the way, that if you care enough to read this, you already possess a little background about Y2K: the two-digit business, the prospect that afflicted computers may malfunction or shut down altogether, and so on.) Regarding the media, perhaps the most revealing moment came at a March confab of reporters and government officials in New York. "We are drowning in a sea of conflicting information," wailed Jeff Gralnick, the executive VP of CNN's Financial News division. In the words of the Reuters wire story on the conference, "[Gralnick] lamented the lack of a unified government stance, saying, 'We need a common reporting language and a common reporting document, a common understandable report card that we can measure you against.'" It hardly needs saying that Y2K is a complicated story, vast and technical, but Gralnick's complaint was of a different character. His was less the voice of a watchdog than of some poor hound put in the embarrassing position of criticizing its master's efforts at training. The government has since gotten its act together (if one ignores the troubling reports that continue to issue from the General Accounting Office), and so has the press. Investigative reporting about Y2K is nowhere to be found. Skeptical analysis of the official optimism is largely confined to the Internet, where about 90 percent of the Y2K discussion is taking place.
As to the "contained" portrait of Y2K problems that Koskinen et al have sketched, numerous cautionary tales suggest otherwise, such as the one embedded in an obscure working paper of a 24-country consortium of oil and gas producers called the International Energy Association. It concerns an unnamed oil company that in late 1997 tested the computer relay terminals controlling the flow of oil through their refineries. Engineers discovered thousands of Y2K-defective chips that needed replacing on thousands of valve controllers. But those exact chips weren't being made anymore, and as it turned out, the new ones wouldn't fit on the existing master circuit boards. So the circuit boards had to be replaced as well. It was next discovered that the new circuit boards didn't fit the old valves, so all the old valves had to be chucked, and new ones installed.
Not exactly sexy details, I realize, but what stories like this offer is a glimpse of the enormous practical complications that surface when it comes to the mundane realities of fixing the Y2K bug. One might reasonably ask why you or I should care if an oil company incurs a few million dollars in unanticipated repair expenses, but ponder the question a moment and the answer becomes clear. What's true of computers that open and close valves in refineries is potentially also true of computers that track bank transactions, issue payroll checks, generate electrical power and route it through the nation's massive energy grid, process satellite signals, and coordinate the mechanized procession of automobile and processed-food manufacturing lines. One begins to get the picture, and it isn't a pretty one. Now consider what the pros have to say about the typical software re-engineering project. They will tell you that, whatever the claims about being on schedule, more than 50 percent of these projects will come in monstrously late; that fixing software bugs inevitably introduces new bugs; and that the testing phase can take as long as the repairs themselves. One Internet site compared the magnitude of Y2K software troubles to confronting a Grand Canyon full of marbles; only a comparative handful need replacing, but you're on your own finding them.
How is it that so much has been left to the last minute? And with so much apparently at stake, how are so many people, from the president on down, maintaining such exquisite public denial?
The object lesson of Y2K is that the headlong plunge into the information age that has occurred over the past quarter-century has wholly outstripped human comprehension of its implications. Almost no one grasps the extent to which computer systems control the everyday economic and social functions we take for granted, from international currency transfers to ATM withdrawals to the medical-testing devices in any hospital. It follows that almost everyone is prone to drastically underestimate the possible effects of simultaneous computer failures across the land. As one friend put it to me in an e-mail posting about an essay he'd read online: "A depression? On the basis of THIS single material problem?" He was inclined to put down the essay's alarmism to the author's crypto-right-wing survivalist prejudices. And it's true: Even apart from the predictable millennialist types who abound on the Internet, a distressing number of the avid Y2K commentators seem the sort of folk who might have Ayn Rand shrines in the living room.
But the point is that widespread computer failures, if they occur, will not be a "single material problem," but thousands or possibly millions of them. Consider the following scenario: A vital piece of robotics breaks down on a car-assembly line, temporarily halting production. In the normal course of events, the problem is isolated and replacement parts secured. Now consider the same "single material problem" in the context of more generalized Y2K failures. First, is there anyone on hand to fix the problem? Are payroll systems up and running so that employees can be paid on time? If there is a problem with microchips, as is likely, can the chips be procured? Is the factory in Taiwan that produces them still running? Are shipping networks serviceable? More basically, is there electricity in the plant?
And if major automakers, to take one prominent example, do find themselves experiencing troubles of this magnitude, what of the economy in general? Common sense, of course, says breakdowns on this order can't happen -- but we might do well to remember that "common sense" is just shorthand for the lessons learned from the sum of experiences, and there has never been an event in the short history of the computer age remotely similar to this one.
Another of the comfortless dictates of common sense is the belief, often expressed in everyday conversation about Y2K, that we did all these jobs manually once, and if need be we can do them that way again. Unfortunately, this is not true. As a close reading of the Senate and General Accounting Office reports to date will indicate, only a portion of the water utilities and wastewater treatment facilities around the country still have equipment in place to make manual bypass an option; electric utilities, phone companies and the vast majority of automated manufacturing operations have long since dismantled the equipment used to do the job manually. (Which gets at an aspect of Y2K that no one has seen fit to mention: Call it the revenge of all the displaced workers who have marched dismally past evening news cameras in the last two decades.) A while back my father told me of a conversation he had with a friend, a man who used to do repair work for the local water utility. Could it be manually operated in a pinch? It could have been until about two years ago, the friend replied; all of the old physical plant was torn out then. And this was in a Midwestern town of 800.
Then, too, there is a dimension of Y2K denial that has everything to do with being American. It has been more than half a century since the last epic trauma on these shores, the Great Depression of the 1930s, and the generation that still recalls its lessons in hardship and deprivation is shrinking in size and receding from public life. Aside from Vietnam, the memory of which seems effectively expunged by the remote and surreal Nintendo wars of the Bush-Clinton years, the post-WWII era of American empire has fostered a sense that any contradiction can simply be outrun, any looming crisis wished away -- a faith nowhere better illustrated than in the huge and long-standing stock market bubble that has created so many paper fortunes in the past 15 years. There is at the same time a vast and growing class of Americans on whom the other shoe has already dropped -- the 70 to 80 percent of the populace which has seen its work week expanding and its real wages contracting for the past 25 years. But they are without voice in public life, and most of them are too busy dealing with today's problems to worry about tomorrow's. In the true America -- the one where the winners reside -- things will work out because that is what things do.
Where public complacency about Y2K is concerned, it doesn't help matters that corporations and government bodies routinely lie through their teeth about their preparation status. The mendacity mainly involves redefining terms and adjusting parameters: recalculating project duration so as to exclude testing time and contingency planning, assuming rather than verifying that all outsource suppliers of essential goods and services will be in perfect trim come zero hour. As one anonymous corporate contributor to a survey by the website Y2Knewswire.com put it, "We redefined [terms like] 'compliance' and ‘critical systems' after [our] first 10-Qs got us reamed by the market analysts."
As this observation implies, much of the rosy reportage about Y2K progress in the private sector is driven purely by market pressures. Whatever the real stage of this or that company's progress, businesses are compelled by considerations of stock price and real-time competitive advantage to report that everything is just swell. Even so, signs of hemorrhaging grow more evident: A May survey by the management consulting firm Cap Gemini showed the largest U.S. corporations falling further behind on Y2K, with 22 percent admitting they would not finish their critical systems on time, as compared to 16 percent last November and 12 percent in August. It is likewise telling that business lobbyists have converged on government halls across the country, demanding legislation to protect them from the liability suits that will be filed in the event of Y2K-related product failures.
At the federal level, both the House and Senate have passed bills imposing caps on corporate Y2K liability, which the president has reluctantly promised to sign. What plays as public-policy debate is in large measure a battle royal between Democrats and Republicans over the allegiance of certain key factions of the money-giving classes. Y2K poses a conflict between two donor strongholds of the Democrats -- high-tech capital, which wants the liability caps, and trial lawyers, who don't -- and so Republicans naturally want to use the issue to their advantage. Not only might they leverage more high-tech money into their coffers; they would also buy themselves a campaign issue by picturing the Democrats in bed with the shysters. Here is American politics at its best, but needless to say, there has been no serious talk of extending any Y2K legal protections to those mere citizens who face the prospect of erroneous credit reports and botched financial transactions.
The situation within the federal government's computer systems is no better than in the private sector. According to the Office of Management and Budget, most federal agencies missed their June 15 deadline for filing contingency plans. And despite the much-ballyhooed progress of the Social Security Administration, the little-noted fact is that its electronic funds transfers to recipients are conducted via the Treasury Department, which has received some of the poorest Y2K grades in the entire government. Medicare and Medicaid, like the health-care infrastructure they serve, look to be disasters.
The rest -- smaller companies, small independent utilities, state, county and municipal governments -- are anybody's guess. But even the mistiest of Y2K optimists conclude that these smaller sectors will be far worse off than their larger counterparts. And this is to say nothing of the situation internationally; it is universally agreed that nearly every other country is in worse shape than the U.S. No matter. After a decade spent flogging the public with the "global economy" and the growing pains we must endure in its name, most business and government officials now profess confidence that what happens around the world won't make much difference here after all.
Talk with any government official or self-styled Y2K expert and the conversation will eventually come round to "the Iron Triangle," which denotes the strategically critical triad of banking, telecommunications and the power grid. Whatever the medium- and long-term repercussions of Y2K for business and the broader economy, it's these three sectors that matter most for preventing widespread public panic and civil upheaval. By most accounts, banking has outperformed every other segment of the economy in gearing up for Y2K; the telecommunications industry is likewise thought to be in comparatively good shape. In both instances, however, all the usual caveats apply: Even the president's Y2K man, Koskinen, has warned that the thousands of small banks and 1,400 or so small phone companies remain a worry.
With respect to the most vital component of the Iron Triangle, the power grid, questions abound. As in every other sector, hard data on preparedness is impossible to find. Industry officials are quick to point out that isolated failures on the grid will not make the lights go out. This, they say, is because the system was designed to accommodate isolated outages; electrical power can be rerouted to cover generation or transmission failures with barely so much as a flicker. The technical term for this feature of the grid is "redundancy." But there's a sticking point, and it's simply this: At what level of system failure does redundancy break down, resulting instead in cascading power outages? It turns out that no one has an answer, even in theory. And thus expert opinion predictably falls along two lines -- those who think the grid will survive the individual outages that everyone concedes will occur, and those who think it won't. (Rick Cowles, a Y2K project manager for Digital Equipment Corporation -- probably the most visible and articulate authority on the workings of the power grid -- started out a qualified optimist, but hedged his assessment considerably after an April General Accounting Office report on utilities.) There appears to be a consensus that sustained mass power outages are unlikely, but phrases such as "rolling brownout" and "rolling blackout" have begun to pop up on industry-insider chat lists.
Behind its sometimes cajoling, more often yawning public facade, the federal apparatus is quietly laying contingency plans for delivering emergency services and putting a distraught rabble back in line if things get ugly. The major media have observed a virtual blackout on this line of inquiry, but it is a subject of much discussion on the Internet. Details are sketchy, and news sources frequently suspect, but a number of more or less credible accounts have emerged. According to one story widely circulated on the Net -- and confirmed by the state official in question -- Y2K czar Koskinen is said to have told the chief information officer of Georgia's Y2K team to make sure the state's contingency plans included three weeks without electrical power. And Chuck Lanza, an emergency management specialist from Dade County and columnist for the Westergaard Y2K website, recently reported receiving a leaked document from the Federal Emergency Management Agency (FEMA) that put the lie to the feds' public posture regarding limited, mostly inconsequential fallout from Y2K.
In the introduction to the FEMA document, under the heading "The Unique Challenge," is the following: "Unlike other disasters (hurricanes, floods, etc.), the consequences of the millennium bug are more challenging to the emergency management community for several reasons. (A) The effects of the Y2K conversion may have a rapid onset and could potentially impact every State in the United States as well as every country in the world. (B) The interdependent nature of the vulnerable systems could result in an incremental or cascading series of impacts. Loss of functionality in one essential service could cause significant disruptions in other services. (C) This hazard has the potential to adversely affect all levels of government, both in day-to-day operations as well as in response to the consequences of other natural or technological disasters and/or emergencies that could occur concurrently."
Along more militarized lines, it has been evident -- and occasionally reported -- that since the first of the year, state National Guard contingents across the country (including that of Washington state, which promises to mobilize at least 2,000 armed guards) have been making preparations for a possible Y2K deployment. I asked an acquaintance of mine who serves as a reservist in Minnesota's National Guard, and he characterized their status this way: "The military appears to be trying to walk a fine line; they don't want to alarm the troops, but they do want them to prepare their families for possible power outages, food and water shortages, telecommunications breakdowns, etc. -- so that [troops will be comfortable] responding to a call-up due to Y2K-related infrastructure failures."
These preparations are not confined to state guard units. In May, the syndicated columnist Jack Anderson wrote of a "civil unrest exercise" at the Marine base in Quantico, Va., that was inadvertently captured on film by a suburban D.C. newspaper photographer. One Marine official reportedly told the photographer what was going on and asked him not to report it; publicly, the Marine Corps denied everything. But according to Anderson, Senate Y2K committee chair Bob Bennett confirmed that military contingency plans were under way. "This [Y2K] problem is everywhere and nowhere all at once," he quoted Bennett as saying. "We can only take a snapshot of portions of infrastructure and attempt to provide the most accurate information we can. But there is simply not sufficient time to understand where all the problems are going to surface, so we must be practical and prepare for the worst." And the website WorldNetDaily.com carried a copyrighted report claiming that Army Special Forces personnel had infiltrated the security staff at a Tennessee food warehouse in order "to learn how to protect food distribution centers in the event of a Y2K emergency or other disaster."
In Canada, by contrast, the staging of Y2K-related military exercises has been regularly reported in the media -- attesting to the Canadian government's lack of that lovably paranoid American style, or perhaps to its faith that the notoriously polite Canucks won't make that much fuss anyway. In any case, the Clinton administration has taken a different Y2K tack, with the essential complicity -- give or take Bob Bennett -- of Republicans equally eager not to get caught on the wrong side of the crystal ball. A few months ago, when Koskinen and company followed the alarming Senate report with soothing words about preparing for a three-day winter storm, there were those who presumed it was part of a strategy to stem public fears by proceeding gradually: first three days, then a week, then two or three. But it's getting rather late in the day to pursue that course. It's entirely possible that Clinton means to bring America through Y2K in the same way he has surmounted his own personal crises: in complete and cheery denial, buoyed by a sense of imperviousness to serious consequences. And no doubt a great many people prefer it that way. For the rest, there will be bottled water, firewood by the cord, and rice. Lots of rice.