This is my response to Y2K Pro's comments in a post a few lines down: "Ok, if electricity, and other utilites still work after the rollover, will the Doomers defect?"

- - - - - Y2K Pro,

This agrument is getting old. I don't recall every "doomer" saying he or she was in exact lock-step with Ed Yourdon's chronology of Y2K events. Many of us who expect problems, don't necessarily believe they will begin promptly at 00:00:01 hours, Jan. 1. In fact, one could argue that many have already begun. Simply count the many code failures already occurring across the US.

The interdependence of events could create a really bizarre set of random failures after Jan 1. As Dale Way postulates: "Not all remediated code will survive interdependability (a new word?) and not all unremediated code will fail." Many subsystems of subsystems will determine the eventual outcome, based upon forward and backward derived date calculations and data. The data must be passed properly from one dependent subsystem to the next dependent subsystem.

And most important to remember: Not all buggy computer code routines will be called in the software right at 00:00:01. Some may be called much later, and some may never be called at all.

COBOL, I hope you are reading this. It's a start on your demand we give you a definition of "compliant", or "noncompliant". - - - - - -

I've written a few lines of Assembly, Fortran, C, VB, and COBOL myself over the years.

-- TruthSeeker (truthseeker@seek truth.always), November 05, 1999

Answers

Agreed. Many, if not most, of the programs that will fail will not run on holidays, weekends, etc. Some may only run at the end of the week, month, quarter, or year. They won't fail until they are run... Other programs will simply corrupt data...which will accumulate until it is noticed...after some days, or even months of processing.

-- Mad Monk (madmonk@hawaiian.net), November 05, 1999.

Guys,

The fact of the matter is that if everything is OK Jan, 1. Then Y2K is effectivly a DEAD issue.

We are worried about embedds right? We are worried about critical application systems, and monitoring systems right?

If the electricity is up Jan 1 (big if) and things look OK a few days later, then Y2K goes the way of disco. The US and other countries have BILLIONS to re-write or reproduce ANY system imaginable..

Bryce

-- Bryce (Bryce@nospam.com), November 05, 1999.

Bryce, your view is a little narrow perhaps. Jan 1, 2000 is not a be all end all date. What about software that doesn't even kick in till the end of the 1st qtr. Among other examples. Read Dale Way again. Or the import/export lists from Dept of Comm, or the CIA testimony.

-- Mitchell Barnes (spanda@inreach.com), November 05, 1999.

Perhaps it's possible that remediators have been postponing repairs to such ending programs (of months, quarters, years). But they probably haven't forgotten about them. And remember that what could have happened had repairs not been made isn't quite the same is what will happen after repairs are made. Those hundreds of billions of remediation dollars weren't *all* burned to keep warm.

I don't believe I've yet read of a single computer problem arising from the actual mishandling of a date. All of the problems I've read about have stemmed from new implementations having difficulties. And I seriously doubt anyone has decided on an "implement whole new system on failure of old one" strategy.

-- Flint (flintc@mindspring.com), November 05, 1999.

Flint,

As always you seem to be just a step off agreeing:

You say: "And remember that what could have happened had repairs not been made isn't quite the same is what will happen after repairs are made."

Yup. What you say is true, but that's not the point. The point was that just because it doesn't happen at 00:00:01 doesn't mean it won't happen. Once again, Flint, pay attention to the Way vs. Yourdon discourse!!!!! "All remediated code will not succeed and all unremediated code will not fail." It has to do with the subsystems interdependency and timing of internal code calls. Stay with the original post, please! Don't modify the post for your purpose!

-- TruthSeeker (truthseeker@ seek truth.always), November 05, 1999.

Truthseeker:

I firmly believe (but of course can't prove it) that date bugs will continue to bedevil us long after I'm dead. You mention problems most of which will show up pretty quickly, but some will be long delayed. I agree with you on that. And I expect 2-digit years to be used by new software for the next century as well, running afoul of windowing schemes for decades to come. This long-latency effect mostly affects IT departments, most of which will (I imagine) employ an extra sustaining programmer or two for a long time. Just one more thing to add to the list of "things that go wrong a lot".

-- Flint (flintc@mindspring.com), November 05, 1999.

This is the second attempt to answer Flint. Get this forum straightened up, please. The first response was there for a few minutes, but has now disappeared. Oh, well. One more attempt.

Flint,

Once more you are almost agreeing with the subject, but you missed the point. Yup. I agree with your comment that things could have been worse had no remediation occurred, BUT THAT'S NOT THE POINT OF THE POST.

The point is: What will happen at 00:00:01 hour? If you read the Way vs. Yourdon discourse, you would understand it has to do with subsystem's ability to communicate across software and networks. All buggy subsystem routines will not be called immediately. Ultimately, certain accounting system modules, time/date sensitive modules, etc. will have to gather dates and data. Whether this happens at the 'magic hour', or later, is irrelevant.

Please don't change the subject for your own purpose!

-- TruthSeeker (truthseeker@ seek truth.always), November 05, 1999.

Hi Mitchel,

I agree that there certainly will be problems. Systems that my wife works on for the EPA are DEFINITELY NOT READY.

These systems won't affect my family's (or anyone else's) safety.. That's the issue I'm really concerned with.. Food distribution, basic infrastructure.

-- Bryce (bryce@seanet.com), November 05, 1999.

I believe that we ought to allow one fiscal quarter to pass before making any big noises about how severe Y2K was. Come January, PLENTY of systems and applications may keep on ticking, it's just that the data that is produced is wrong, wrong, wrong. Like the recent delivery fiascos with Hershey.

I think if we can get to April and say, "Hey, this is not too bad, it's under control", that will be reasonable.

-- King of Spain (madrid@aol.cum), November 05, 1999.

Truthseeker--

I don't know if this is really for me, but here's my opinion either way. If nothing happens 1/1/2000, then great. I never claimed to know much about embedded systems. I can only read what everybody else reads and draw my own conclusions.

As I've always stated, I can only speak about business applications. I agree with some on this forum. It won't be just a January 1st event (even if no infrastructure problems arise). Programs will run at various times and data can be corrupted (unaware to programmers). We have may "automatic run routines" where I work. Jobs run on a daily schedule without any intervention. So, in fact data could be corrupted on Jan 1st. Programs can also bomb, or run improperly. This will not cover every condition under the sun on January 1st, simply because every program doesn't run every day.

I will certainly feel better if no visible signs of infrastructure damage are apparent during the first week of January. However, we still aren't out of the woods. The point being, businesses must still function or we've still got a very big problem. Even though it's nice to have water, heat and electricity, what good are these things if you have no job and can't buy the necessities you need?

We don't need the infrastructure to go down to have a serious problem on our hands. Massive business failures will be bad enough if it does happen. I think we should all be able to see this.

Let's just hope this doesn't happen, but I'm expecting failures.

-- Larry (cobol.programmer@usa.net), November 05, 1999.

We should know if y2k is going to be a major problem when the majority of companies release their first quarter financial reports. If they are having y2k problems, they will have to acknowledge them at that time, or face lawsuits by shareholders. Also, the price trend of petroleum products during the first quarter will provide an important clue. If supply problems develop, we will see rising prices during the first quarter.

-- Dave F (dannco@hotmail.com), November 06, 1999.

One small point. Y2K is already a major problem. I am not talking about failures. I'm talking about the amount of money that has been diverted from increases in productivity [not that some changes won't eventually increase productivity]. The first the year is important: if embedded's [with a proper date set] are important. If all of the problems are spread over the first 6 months of the year, they can be handled [effects will depend on where you are].

Best wishes,

-- Z1X4Y7 (Z1X4Y7@aol.com), November 06, 1999.

Actually, even if the power company keeps it together on the first, that doesn't necessarily mean everything is spiffy. Remember, they are simply one of the 'obvious' and 'visible' targets. But so is the water supply. And this could take several days to show up. The problem doesn't have to be in the delivery system. It could be a failure to perform a task (disinfection), or a 'performed task *too* well' (dumped all available chlorine into the water all at once). And then there are the transportation issues. Suppose that the railroads can't run because all of the computer driven interlock systems are dead. And the switches can't be operated manually? I shudder to think of attempting to replicate the railroad carrying capacity with trucks. And if the rails stop running, then Y2K is a problem about one week later, as the utilities begin to run dry of coal.

Once again, the difficulty is in tracking down the possible effects of any given system. Agreed that not all systems are equal. Nobody cares whether *your* car starts except you and the service department at the dealer. Nobody cares whether VCRs and toasters quit working, at least not in the grand scheme of things. But there are systems which are crucial to our ability to run this civilization, systems which could make for a very dismal performance in the new year. And some of them are not so obvious as things like the water and power.

And for the reasons given above, I, for one, remain unconvinced that just because the lights don't go out, that everything is A-OK.

-- just another (another@engineer.com), November 06, 1999.

Larry,

The post was not intended for you, unless of course, you are COBOL

-- TruthSeeker (truthseeker@ seektruth.always), November 08, 1999.