What I Don't Know (Yet)greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread
I'll admit that it is *possible* that the anecdotal evidence out there basically reports only the worst cases, and not typical cases, and that the worst cases are very, very rare. It is possible that the evidence of power plant failures will turn out to be negligibly important (because, for example, they tested only the very worst cases, or they tested before any remediation at all was done, etc.). And it is possible, therefore, that the 100 million lines of code to remediate at the IRS (when by comparison six million, of 30 million total, were remediated 1991-6 at Social Security) will in 2000 (or rather, when the 2000 IRS fiscal year begins, sometime in 1999), the glitches will be relatively few. Maybe they can work around the glitches. Ditto for the banks -- their efforts will be complete and tested enough to avoid major problems, and the other problems will be of the variety that can be worked around in various ways. And so long as this remains a significant possibility in my mind, I will remain personally unconvinced that we have any very large chance of experiencing catastrophe.
On the other hand, I consider this often: Perhaps there is a combination of secretiveness (due to liability, bad PR, etc.) and apathy (due to managers who are even less convinced than most that Y2K can pose any serious threat that can't be fixed in a few months) at most businesses and governmental agencies. Perhaps their consultants are required not to disclose system failures in tests; perhaps there are other reasons why we are not learning of such failures. But in fact (I'm asking you to consider this possibility), in a frighteningly large percentage of legacy systems, and even relatively recent systems and software, Y2K simulations have caused systems to freeze up, not to operate when certain essential operations are attempted, etc. In fact (consider the possibility) those power plants that were tested were utterly typical and in fact further along in their remediation than others. Etc.
Suppose those are the facts (about tests and about what those leading the remediation business know). Then in the early days of 2000 (and after other relevant deadlines before then), while many small businesses and particularly conscientious banks, etc., would work perfectly fine in ordinary circumstances -- the power goes out everywhere. Many essential functions of many large banks, and even more smaller banks, cannot be performed (not even with paper or desktops run by generators). There is no dial tone in many places. The computers that run train schedules and equipment are hopelessly screwed up. The FAA's computer screens go blank (as they did in one simulation). And so forth, you get the picture, you've no doubt thought of all this before.
Honestly, I don't think that will be the full picture, though -- because if indeed simulations and testing reveals that MANY essential, "critical" systems in many industries and government will just not function, at *some* point this information is going to come out. In 1999 or later this year. At some point someone like Ed Yourdon may say, "Look -- I have not actually said everything I know about the situation, because otherwise firms doing remediation would not agree to work with me. But in fact we have done many simulations and tests and we can say, fairly confidently, that X% (shockingly high number) of systems, on average (higher in some industries, lower in others), are going to be useless on or before 2000. Only a few, out of hundreds, of power plants that have been tested were operational. And I'm telling you now that there is no way that trade and commerce will be able to continue: too many suppliers of too many essential corporations, and the corporations themselves, will be paralyzed and unable to carry out their functions. We know this because we have tested their systems and typical systems at their average level of remediation simply have failed, or been virtually useless (in one way or another). As a result, while trade may be able to continue for a short time while the power is out, traffic signals are dead, etc., it will soon grind to a halt and there will be nothing *anyone* can do about it. It will be as though some God, or some Satan, threw sand into the gears of commerce. The gears are still there and the key people know just how they operate. But they cannot be made to operate. And so we will be reduced to scrambling for survival, reorganizing, and rebuilding in many different ways. I am saying this so that you can prepare."
So I have laid out two different scenarios, each based on what MIGHT now be known (or will be known in the next year or so, I guess) by people who have done testing and simulations. Honestly, both scenarios seem extreme and unlikely to me. But what is scary is that we do not know, and it *might* well *also* be the case that (for complex sociological reasons) we have no easy way of knowing, which is correct.
But here is a reason to think that the results of simulations have been very bad indeed: viz., if they were going *well*, those facts would be loudly trumpeted by businesses. In fact, we hear very few details about how remediation efforts are going; or, when we do hear details sometimes, they are almost never good. *That* more than anything is what makes me think that the potential for catastrophic damage to essential systems is enormous.
This is why I am asking (begging!) for any sort of hard evidence of the different kinds and degrees of potential damage due to poor Y2K remediation efforts, in the "Evidence of potential damage" thread I started. Until I have such evidence (I'm not saying it doesn't exist -- I'm just saying I don't have it), I'm going to continue to suspend judgement on the question of the severity of the damage, and I am going to fear the worst and hope for the best.
That's my position at present. Please enlighten me.
-- Larry Sanger (Sanger.email@example.com), May 15, 1998
Larry: You've put in a nutshell the feelings of many people on this and other boards. I first started following the y2k issue in early 97, just before the Newsweek cover story. I've gone up and down on it ever since, as have several other people I know who are also interested/concerned/involved with issue. I was feeling rather confident about it -- the old "a catastrophe named is a catastrophe avoided" syndrome -- until a week or so ago, when reports of congressional testimony and then Robert Samuelson's prominent mea culpa in the Washington Post started racheting up my discomfort level. Then a U.S. Navy acquaintance of mine got extremely nervous and evasive when I asked him about y2k in his field. I spent this weekend pricing new and used generators.
-- J.D. Clark (firstname.lastname@example.org), May 17, 1998.