It is easy to make one prediction about the Y 2K problem: the hype will grow throughout the year. Even though Lake Superior University placed Y 2K on the top of its annual list of overused words and phrases for 1998, the beat goes on. The headlines on the left are a small sample of the clippings that overflow my desk. A book search on Amazon.com showed 60 titles at the beginning of 1999. A web search yielded more than 65,000 matches. Even Garrison Keillor featured Y 2K in his first Prairie Home Companion radio show for 1999.
If this issue were simply hype, like a pet rock or beanie babies, it could be ignored.
If it were truly the disaster scenario that some are playing it to be, it would require full disaster preparedness efforts. The difficulty with the Y 2K issue is that it is technical enough to leave room for supposed experts to sell their story. This also causes evenhanded people to oversimplify the explanation for the problem. It is serious enough to cause companies and government agencies worldwide to spend perhaps two trillion dollars getting prepared. Since computing and microprocessors are so ingrained in our everyday lives it will affect us all in some form. So what should we do?
If you run a business and have systems that you have been ignoring, it may already be too late. You had better get to work on the most critical items or you may find yourself unable to carry on business as usual.
If you are concerned about the impact on your own lifestyle, the precautions you take depend in part on your level of risk tolerance. But they should be based on a good understanding of the risks. I talked with an executive in a Wall Street consulting firm who told me she would not have her personal dollars in the market at the end of
1999. “Of course,” she added “I never have my own money in the market. I don’t have the personal tolerance for that kind of risk.”
In this column, I will describe an alternate way of looking at the origin of the problem, sketch its seriousness, give some reasons for optimism for a substantial (but not total) solution to this problem by early next year, and describe some of the secondary effects which may be more difficult than the primary ones. I will also give some guidelines for evaluating things that you read, because the serious reader will want to look at this issue more broadly and from many other sources.
The Origin of the Problem
A typical explanation of the problem is like the one in the December 31, 1998,
Seattle Times article which says it
… stems from the decades-old practice of using two digits instead of four to represent calendar years in computer programs. The abbreviation originally was considered a smart move to cut what then were expensive computer memory costs. The fear is that the old shortcut could cause computers to think [my emphasis] “00” means” 1900” rather than 2000 when performing a variety of date calculations. The problem also afflicts embedded chips the microprocessors found in everything from alarm clocks to power plants.
This is close, but it greatly oversimplifies the issue. You and I use a two-digit date much of the time, and it is not because we are trying to save ink in our ballpoint pens. Human beings in normal dialogue understand context. Computers don’t think, and not knowing context is one of the manifestations of that.
This lack of contextual understanding is much of the reason there are so many people employed in the computer industry. Fractions of these are creating new tools, while many others are “maintaining” the hardware and software systems that we use every day.
I once had a boss who asked why there were so many people maintaining software.
“Does it have to be oiled?” he asked in frustration. Maintenance involves a great number of things, but a large portion is dealing with implicit assumptions people had about the systems they were building which are no longer true.
The calculation of the passage of time by subtracting the earlier date from the later date, for example, works fine for two digit dates not crossing millennium boundaries. Equally, though, a payroll program assuming no more than 49 employees will often fail when the number grows to 50, for technical reasons not important here. These issues, both complex and simple, are dealt with millions of times every day by software maintainers.
The Y 2K problem is a specific version of this. Some say that it is different in that the problem comes up at a single point in time for systems all over the world. This is not completely true either. Most of us have insurance policies, credit cards, or other contracts that expire beyond the year 2000. There were stories of small banks in Europe where credit cards with expiration dates beyond 2000 could not be used, but most of these problems were passed without notice.
How Serious is This?
In a January 10, 1999, street survey, the Eastside Journal asked, “Are you worried about the Y 2K problem?” One respondent replied, “It’s just for people with Computers!” Unfortunately, it is not this simple. Computers are not simply an “add-on” to our lives any longer, and the Y 2K problem is not confined to programs that run on those boxes in the office or at home. One estimate said there are ten times as many microchips outside of computers (in things like the car, toaster, control systems at the power company, and the like) as in them.
Thus, your car, the ATM machine, the airline reservation system, the automated controls of the electric power system, and the check-out counter at the grocery store are all controlled by computers or microprocessors. We have come to depend on these systems to work correctly all of the time.
A second thing, which points to the seriousness of the problem, is the interconnectedness that networks and computing have enabled. The good side is that your local bank is connected to resources all over the world. The downside is that these connections may create vulnerability in your system that has been fixed, if one of the connected systems has not.
With this level of seriousness, why not line up with the doomsday prophets?
Some Reasons for Optimism
While the reality of the problem is daunting indeed, some companies and agencies have been hard at work providing solutions. Gaining an audited “Y2K compliance” has been an ongoing process that many companies largely achieved by the end of 1998. Social Security, for example, announced its systems were ready to go at the beginning of this year.
The testing and fixing that has been going on has brought interesting results in addition to the large amount of money being spent. One power company (more likely many) moved its computer clocks forward and saw the whole system fail shortly after “midnight, 2000.” This is a good thing, not a bad thing. Doing it in simulation both proved that the problem is real and validated the correction process, all before it caused a real disruption in service.
Another reason for optimism comes from the description of the origin of the problem. If this were an unusual kind of problem that we had never seen before, that would be one thing. But as a computing system maintenance problem, similar to what millions of programmers and systems analysts work every day, even those problems that carry over to the year 2000 are likely to be fixed quickly.
Since many of the problems, though certainly not all, are in the older mainframe computer systems, this actually simplifies the solution. Mainframe systems have all of the code in one place, usually as an integrated whole. If we were further along in the implementation of distributed systems, this would be a tougher problem. This much deeper but rarely addressed problem will be the subject of a future column.
The Information Technology Association of America (ITAA) publishes its Year 2000 Outlook every Friday. In its December 31, 1998 issue (Volume 3, No. 47- note that this has been going on for some time! ) are the following comments:
With just a year to go, an air of cautious optimism seems to be in the air. Those enjoying this more hopeful prospect include Peter de Jager, perhaps the first of the Year 2000 gurus. De Jager predicts “hiccups” along the way to the century rollover, but not an economy choked by its digital appetites. “Will there be problems? Oh yes. But not of the doomsayers’ [variety]”’ de Jagers says. He claims society has awakened to the threat posed by the Year 2000 and it has responded. “We got our act together,’ he says, ‘Reality started.”
That view of a brighter tomorrow seems to hold constant among a variety of Y2K observers and practitioners. David Baker, managing director of Schwab Capital Markets and Trading Group, says he is feeling more optimistic although the good feeling ends at the water’s edge. “I’m more optimistic than I was six months ago,” he says. But Baker also believes there is “potential for regions of the world to become isolated” on the wrong side of the digital divide.
Embedded systems consultant David Hall says the Y2K situation is “head and shoulders above where we were last year. “Still, not all observers enter the new year with new found confidence.
The bottom line is that no one knows for certain what will happen in year 2000. By looking hard enough, you can find someone who agrees with you no matter what your view. So you must evaluate various things being said against your own tolerance for risk in deciding what to do.
I recently attended a Saturday morning Y2K seminar because I wanted to hear first hand what some of the views are. The two speakers were very sincere, but both were Y2K consultants and their biases quickly showed through. One opened his presentation with a 15-minute infomercial from his Y2K consulting firm, describing all of the consequences of not acting.
This experience, combined with other reading, suggested the following questions to ask when evaluating what you are reading on this subject. You may want to add a few more.
What are the biases of the speaker or writer? If they are optimistic, do they try to be fair in describing the reality of the problem? If they are doomsday prophets, do they point to the reality of the progress that has been made by many?
What is the expertise of the speaker or writer? Do they understand the technical nature of this problem, or have they simply read about it? Are their quotes primarily based on motivational comments to get people working on the problem? Those who don’t understand may not be able to separate critical and nuisance problems, motivation or progress, or to define the true nature of the problem.
What does the speaker or writer have to gain from one outlook or another? There are those who are more interested in selling their services than in presenting a balanced view.
What is the speaker or writer’s tolerance for risk? A person afraid to fly or drive
(unrelated with Y2K) could give a convincing argument of the risks. What you do with that argument depends not on what they say but what it means to you.
Is the speaker or writer sensational in what is said? Catastrophes seem to sell more books or draw larger audiences than analytical approaches. This may account for the large number of sensational titles and speakers out there, but it does not make this a reliable point of view.
In addition to the real problems with Y2K, secondary responses to some of the rhetoric may cause more difficulty than the Y2K problem itself
People fearing the loss of availability of cash from the banks could put a run on currency (and more dollars are being printed in anticipation of this). Remember the gas lines of the ’70s (excuse me, 1970s)? Withdrawals from the market based on similar fear could cause a dip independent of any possible real problem with the market.
There is also a concern over new types of fraud. We can’t begin to imagine all that people might do, but here are two examples. A bank has a temporary loss of access of information, and a customer uses desktop publishing to recreate a printed “copy” of his or her records with different values. An increase in burglaries may result from all of the cash people have withdrawn from the banks.
The problem is real, and if you run a business and have been ignoring it, get help quickly. While this may not be the international problem some fear, it could spell disaster for individual businesses that depend on computing to carry out their work, even if they are not fully aware of how dependent they are.
As an individual, no one knows for sure what will work and what won’t come January 1, 2000. Like a winter storm that could take out power for a few days, we could be prepared with food, a full tank of gas, and some cash. The Y2K difficulties may not happen, but the winter storm may. My wife will probably want to do this; I probably wouldn’t on my own.
Al Erisman is executive editor of Ethix, which he co-founded in 1998.
He spent 32 years at The Boeing Company, the last 11 as director of technology.
He was selected as a senior technical fellow of The Boeing Company in 1990,
and received his Ph.D. in applied mathematics from Iowa State University.