Or is it simply a lack of problems (that is, Y2K fixes are going better than expected and are on track)?

I have been following the problem intently for over a year now. Last summer I wrote a Y2K "road map" as a quarter by quarter series of expectations I had about Y2K and it's effects. I did it mostly for myself as an exercise to help me plan, prepare, and be watchfull. It has gone pretty much as I wrote, except I was too pessemistic about the stock market. I felt the end would begin by last fall.

My thoughts about missed deadlines (12/31/98 and 03/31/99) were right on. My ideas for a coming shortage of preperation products also seem to be occurring pretty much right on time. The other exception to my timeline was that I thought we might see more disruptions occurring by now as companies were putting their remediated code and embedded systems fixes in place. There have been a few scattered stories, but an accelleration of these problems doesn't seem to be happening. At least not yet. This leads me to try to figure out which of the two reasons I lead with is correct.

If companies and the government were making as good progress as they say in the press releases it is my contention that the level of glitches we would be seeing would be on the rise. I would consider this GOOD news, not bad. It wpould mean that they were at the stage of implementing their repairs. The lack of this however seems to me to be BAD news. It means that the repairs are taking longer and they are not yet at the stage of implementation. It also means as implementation begins in ernest that the inevetable disruptions will be compressed into a shorter time frame. Also BAD news.

I had hoped that we would see a gradual buildup of problems this year with the resulting publicity getting more and more people aware and prepared. It looks like instead the disruptions will be later in the year, come faster and more frequently this fall. This will cause longer times for repair as many bugs will occur in a short time frame. A kind of "I wonder which process of the 100 we implemented overnight caused THIS problem?"

Another way to look at this is to consider the famous Bell Curve principal that has been discussed recently as a way to judge how repairs are progressing. We should also see a bell curve of problems with the fixes as they are implemented. A lack of a large number of problems now means we are on the leading edge of the curve, with the peak in future. If the peak of the problems with the fixes happens to coincide with the actual roll over date of 1/1/2000, I fear we are in real trouble.

-- Kevin Lemke (klemke@corpcomm.net), April 04, 1999

Answers

I imagine that the bugs that have surfaced have not overwhelmed the IT staff yet. I'm not sure if this is good or bad.

-- d (d@dgi.old), April 04, 1999.

Kevin --- interesting post and one of the few speculative ones in the past couple of months that hits a bit of new ground, at least for me. If I understand your drift, the lack of reported problems either mean better progress than originally anticipated or significantly worse than anticipated. Considering that deadlines are being publicly missed, which suggests that remediated code will indeed be later coming online by and large, Y2K may indeed be worse than originally ANTICIPATED later this year and early next.

-- BigDog (BigDog@duffer.com), April 04, 1999.

Peter de Jager wrote an essay a few months concerning how computer failures are evidence of progress. You may want to go to his site (www.year2000.com) to check it out.

-- Incredulous (ytt000@aol.com), April 04, 1999.

Kevin,

Excellent observation. The companies 10Q's indicate they are still substantially in remediation vs implementation. I suspect some code is being put into production, fails, old code reinstalled in very short time frames. Thus, publicly, it looks either normal or "minor" glitches. They will have to start serious implementation by July to have enough time to attempt repair prior to rollover. I think it will be quiet until the end of July, then a ramp up of problems.

-- RD. ->H (drherr@erols.com), April 04, 1999.

As time flows by, spike dates pass uneventfully, and positive spin rolls in, we need more and more imagination to keep the faith. Could it be that remediation has cured the worst offenders? Could it be that the actual bug impacts were exaggerated? Could we have underestimated our ability to cope with failures? Is it possible that compliance isn't being announced because testing goes on forever and it's unwise to make guarantees in the fact of the unknown? Do you suppose we can actually get along without some of our computerization? Most terrifying of all, could their be a grain of truth in what de Jager and Koskinen are saying? This is the stuff of nightmares!

OK, false alarm, not to worry. The spike dates were actually a hoax, and we're so far behind we haven't even put our fixes into production yet. *That's* what's going on.

-- Flint (flintc@mindspring.com), April 04, 1999.

Maybe I'm a dolt, but I think the lack of problems is evidence that we are still in the year 1999.

-- Uncle Deedah (oncebitten@twiceshy.com), April 04, 1999.

Just about every business (small) we're called into fits into one of two groups. Group one has done nothing and doesn't really think it could be a problem for them. Group two has been working on it for a while and reports they are within 10% of being finished. Of course, what worries me is that these same people were reporting they were within 10% of being finished about 6 months ago. It seems that in any software project we've ever been involved in; that last 10% is the hardest. Everybody likes to work on the project when it's new and fresh. But when you get down to getting the last glitches out and enhancements finshed, it bogs down. So from what I'm seeing, I'd have to vote for the case of not as much has been put back into production. Hey, if you're running the old stuff (which has been running for the last X years), you wouldn't expect many problems.

-- Greg Sugg (gregsugg@bbnp.com), April 04, 1999.

C'mon Flint, give it a rest. That wasn't the spirit of my reply or Kevin's post.

-- BigDog (BigDog@duffer.com), April 04, 1999.

Kevin, Good Question !!

Personally, I think the "Bell curve" is going to look more like a division symbol... /~~~~ .... A sharp spike and rolling failures. Your assumptions are correct i.e. the longer we go without "testing gliches" the sharper and more damaging the spike will be.

God bless the spin meisters...they're gonna need it.

-- WebRNot (webrnot@ncap13k.com), April 04, 1999.

The "FIXED" systems can not go online till they are finished. We all know there are no errors in programing. Since there are few failures, one can only deduce the code was fixed without error or, THEY AIN'T DONE!!!!!!! This isn't brain surgery. Lies maybe, but not brain surgery!

-- SCOTTY (BLehman202@aol.com), April 04, 1999.

Is it really that important if everything crunches at the roll over or not? If the rest of the world is down, we will follow. I look at the Kosovo situation with 750K refugees and then I think about Ko-sic-em's statement about the possibility of 30 million being without water in our own country. What do you think Dee Cee and NYC will look like?? Our last resort bug out plan is Costa Rica where we own land and know people. Beans, rice, tortillas and green banana soup don't require computers. Nor do springs and outhouses!!

Gotta Passport? Gotta boat?

-- Taz (Tassie@aol.com), April 04, 1999.

Look, regarding this month as a "spike date" due to fiscal year 2000 rollovers: Until we can see that folks in New York State, Canada, etc., got their checks on time and with the correct amounts, I would not be too sure of anything. There is a lot of stuff that can be glossed over, but that sure isn't one of them. (Of course, if in order to get those checks out, they are having to play games like extending Fiscal Year 1999 so that it extends out to 12/1999, and so becomes something like a 22 month year, that is sure worth noting too!)

-- Jack (jsprat@eld.net), April 04, 1999.

Taz,

"and then I think about Ko-sic-em's statement about the possibility of 30 million being without water in our own country."

Where and when did he say that?

-- Tom (waylink@sontos.net), April 04, 1999.

Interesting, but it's no wonder it's tough to get people to believe there's a problem. When you say, "No problem indicates there's are going to be severe problems," people can understand that (after a little explaingin). But when you follow it up with, "And if there ARE problems before the date rollover, it means there will be severe problems," it sounds like you're saying there will be severe problems, and everything proves it.

You say that your predictions about preparation supplies suppliers running low on inventory were accurate, and deadlines have been missed. But the stock market didn't crash. (And wasn't Asia supposed to have sucked the global economy down the tubes by the beginning of the new year? As I recall, I wasn't supposed to be able to even buy new boots by spring of 1999.)

I remember reading in a column on Westergarrd last December or so that the Joanne Effect was going to do some big things that wouldn't be able to be covered up. Joel Willemmsen himself (of the GAO - whom I have tremendous respect for), was quoted as saying the unemplyment system was going to do mess up and that people's checks would be missing, late, wrong, something. Nothing happened. (That I heard of, anyway. Nothing scientific in the least, but I had a couple of waitress friends who deal with a lot of construction guys watching that one closely. Nothing. And they hear just about everything...)

And speaking of the Joanne Effect, before it was called that (I think), I remember a lot of people saying last summer, "Just be patient. We'll be able to get a pretty good idea of how bad y2k will be in January of 1999." (And a lot of those people were technicians.)

And last fall... At least one of HUD's systems was supposed to go haywire and mess things up for at least some landlords with supsidized housing (not to mention the zillion other things HUD's into - loans.)

I watched Jim Lord draw a chart on a video he made about a year ago. A jagged, upward moving red line that peaked in July of 1999 when he predicted the real mass of y2k problems would errupt. Seems we should have experienced a couple of those jagged spikes by now. July's not that far away...

And now you're saying those things not happening is most likely proof that the there are substantial problems that won't be resolved.

I understand what you're saying, but it does beg the question of what would you consider evidence that problems would be less substantial than most are concerned about (meltdown, or slightly less than meltdown)?

-- listening (----@---.net), April 04, 1999.

Flint, you're being belligerent again. I bet your wife isn't home much anymore.

Listening> We bill the state for the services that our home health agency provides. They rely on their computer to screen for double billings, unauthorized services, etc. Last week when our billing came in, they had included every billing for the last three years on three of our clients--came to about 60 pages of garbage--no extra money, just aggrevation. When I called Columbia, the answer given was "New computer stuff didn't work at all." Then wouldn't explain WHAT new computer stuff. God help us when SC new fiscal year starts in July. Also, one more item. SC State Employment sends us a list once a month of all the individuals sent to apply. Major goof--list on 4/2 was over 11 pages long. Haven't called yet--still too busy laughing.

-- Lobo (atthelair@yahoo.com), April 05, 1999.

Not necessarily lack of progress - more likely a lack of adequate testing, AND a lack of adequate progress. The nuclear-grade failures - while very irritating to me as an indication they didn't foresee all problems correctly, and didn't prevent all of them adequately - indicates that those few utilities that did get to the point of advancing dates (some 50 now I've been told) are the ONLY ones able to get to a point of advancing dates.

The remaining 3950-some-odd are ????? The rest of the government agencies are ???? The rest of the country's utilities are ????? The rest of the country's processes are ?????

The "bell curve should be at 1/4 complete by now. Instead, we are no more than 1/4 of 1 percent tested - even if the banking and stock market industries are added in ..... A Fortune 500 company (or state) or federal agency should be finishing every few hours now = testing completed, everything completed - and none are. NONE.

-- Robert A Cook, PE (Kennesaw, GA) (Cook.R@csaatl.com), April 06, 1999.

What is the phrase from the field of logic? I think it's Denial of the Antecedent. Is it: Absence of evidence is not evidence of absence? Ya, that's it.

"Doubt is unpleasant. Certainty is foolish." (Voltaire paraphrased)

-- Donna Barthuley (moment@pacbell.net), April 06, 1999.

It's also a form of the Argument from Ignorance.

"Now, here, you see, it takes all the running you can do, to keep in the same place. If you want to get somewhere else, you must run at least twice as fast as that." Lewis Carroll, Through The Looking Glass

-- Donna Barthuley (moment@pacbell.net), April 06, 1999.