* * * * * Question for ED YOURDON * * * * * : Will FEBRUARY 29 require the same JANUARY 1 world-wide production SHUTDOWN ??? Why not??? Remember New Zealand smelter disaster leap-year 1996??? (If needed, these two procedures combined would cost more than 1% of world GDP!). .... Question also for TOM ATLEE (Co-Intelligence Institute) and DALE WAY (I.e.e.e.) plus SHERI (Hi Sheri!) and BIG DOG (arrff) and dear old FLINT (Hi Flint old fox! (( It's a compliment)).... * * * * * * *

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

Dear Ed Yourdon (plus Dale Way, Tom Atlee, Sheri, Flint):

One possible explanation for the "Y2K-is-a-bust" theory is that rollover triggering of Y2K errors in embedded systems was smartly avoided (so far) by shutting down production equipment, factories, etc., (some/many still in "manual"?) plus physical control systems infrastructure, on-line transactions, etc, thus avoiding the much-feared embedded systems failures world-wide during the short-time (seconds, hours) windows of ambiguous double century status.

Taking into account downtime, start-up costs, tons of out-of-spec production, etc., this course of action could have cost pretty close to 0.5% (or more, depending upon accounting methodology) of world GDP, a cool amount of money by any standard.

Now February 29 does not fall on a week-end. Actually it's smack in the middle of the week (Tuesday). So if the same world-wide shutdown procedure were needed (question: is it needed? why not?) then costs would soar well beyond 1% of world-wide GDP. Has any thought been given to this ? What other I.T.consequences would this second shut-down have ? Consider the New Zealand leap-year smelting plant disaster in 1996...

Please input your thoughts.

Others also welcome...

Best regards

-- George (jvilches@sminter.com.ar), January 11, 2000

Answers

George:

Been wondering how you have been doing.

Any "inside" news or good rumors from Argentina? Anything besides "normal" problems?

-- mushroom (mushroom_bs_too_long@yahoo.com), January 11, 2000.


One thing to keep in mind is that the New Zealand smelter damage didn't occur until December 31, the 366th day of that year.

-- Some (food@for.thought), January 11, 2000.

I'm not Ed, but it is curious to me that you are going to that source for an expert opinion. Mr Y has made it clear he was only quoting others...Quote me on this if you like:

Nothing of consequence (that is, nothing more extraordinary than any other day that week) will happen on Feb 29th. Won't matter whether or not production is shut down. Certainly there was not 100% compliance with shutting down the millions of systems containing embedded chips at the last rollover. Serious problems from computer glitches have ALWAYS happened and WILL happen for the forseeable future. Y2K is a small percentage of overall glitch causes.

-- I'mSo (happy@prepped.com), January 11, 2000.


Mushroom, I'm doing just fine, thank you. How've you been ?

To answer your question, the answer is "NO, no way Sir" . Everything is just Y2K PERFECT around here. No glitches, nothing, zilch, nada de nada. Not in Brazil, not in Argentina, not in Chile, not in Bolivia, not in ECUADOR (for Heaven's sake), not even in PARAGUAY , which self- declared itself as Y2K-defeated.

So this one TRILLION dollar mystery goes on. De Jager says that there were less century boundary dates involved than what originally expected. To my way of thinking that is plain B.S. More, less, you still have to find them, remediate them, and fully test them. In Italy, in Russia, in Nigeria, etc. The world-wide banking system is Charlotte's web which involves 186,000 financial institutions. I could go on and on. Peter de Jager's explanation is, at least, childish. That aspect of Y2K was studied to death with frightening results.

I don't have too much time to post or lurk. I have to work ! But still I'd like some input, top-of-mind ideas about my February 29 question.

Got any Mushroom, anybody?

Stick your neck out, you've got nothing to lose gals & guys ? Y2K is over, right? So...? Explain it please. Flint old buddy, please show up. You are, by far, my favorite polly. Come on guy, let's make friends!

Best regards. Take care.

-- George (jvilches@sminter.com.ar), January 11, 2000.


I have to agree with I'mso. There is a little more chance of problems than on the 9-9-99 but nothing noticable will occur.

-- watch&wait (notoveryet@sinking.ship), January 11, 2000.


I agree nothing will occur. Feb 29 is not a sorting problem. Its only an extra day problem. The only serious consequence is on scheduling systems, that more than likely have already found out they were missing such a day. If a company finds that their software doesn't handle Feb 29 then all they need to do is skip that day and input data dated before or after that date. Anyother computer type problems can be solved by rebooting the system and resetting the dates do the current date (the next day of course).

No this is an easy one. I don't forsee any problems. This is my 27th year in this industry and I'm on the money on this one. I didn't think y2k would bring the world down. So now I'm feeling very strong about my convictions.

Justthink com

-- justthinkin com (justthinkin@y2k.com), January 11, 2000.


Some, thanks for the info. So it seems that December 31, 2000 could possibly turn out to be a pretty "eventful" day, maybe.

I wonder then if XII-31-2000 is in everybody's radar screen...

So would Feb. 29 be as safe as some have it ? What would happen to those programs and data bases that did not take into account this 400- year exception? Nothing? De Jager's explanation again ? Could this be true? We are talking about a trillion dollar waste here...

Take care.

-- George (jvilches@sminter.com.ar), January 11, 2000.


So let's assume for a minute that leap year 2000 (Feb.29) would "only" be an extra-day problem. Not more, not less.

So when would it show up, on December 31, 2000? Wouldn't it be funny/terrible if Y2K shows it's ugly face a full year later? Is everybody convinced that XII-31-2000 is the only possible event- date?

The New Zealand 1996 smelter experience would indicate it could get to be pretty bad, right? How did the smelter meltdown happen? Which equipment/embedded systems would be affected? No software, data bases affected? Would the world have to shut down again on December 31, 2000? At what cost ? What would happen if everyone believes that Y2K is already 'over' ? How hard would it be to achieve awareness?

Take care

-- George (jvilches@sminter.com.ar), January 11, 2000.


I didn't do a good job of my first reply.

Sorry.

Read the same reply and substitute the day of your choice for Feb 29.

Get it?

-- I'mSo (happy@prepped.com), January 11, 2000.


Just for what it's worth, I agree that 12/31/00 is likely to be the biggest "spike" day remaining for Y2K failures--but that it's going to be a much smaller spike than was 1/1/00 or 1/3/00. But, of course, that has to be weighed against what will probably be far less civil and corporate contingency planning and preparation.

Net-net, my guess (and, I admit, it's a pretty wild guess) would be that the next New Year will have about the same impact as the last one. Within an order of magnitude or so, anyway.

As for the leap-year itself, I also agree that it's going to be about as big a deal as 9/9/99...that is, not much of one. Problems that do crop up are much more likely to be leap-year related than Y2K related. To get a Y2K problem on 2/29, you have to (a) write code that remembers about leap years, (b) write code that gets the century exception right, and (c) write code that gets the 400 year execption to the century exception _wrong_. I have never encountered code that meets all three conditions. I rather doubt it exists.

The real 2/29 risk, as far as I can see, is from systems that don't know about leap years at all...systems that would have been vulnerable back in 1996, such as the in the famous New Zealand incident.

Just my contribution to the groupthink on this thread.

-- Craig Kenneth Bryant (Ckbryant@mindspring.com), January 11, 2000.



The real 2/29 risk, as far as I can see, is from systems that don't know about leap years at all...systems that would have been vulnerable back in 1996, such as the in the famous New Zealand incident.

Craig, I was thinking pretty much the same thing. I'm not a computer expert, but... it just seems illogical that any system which had no problems with the leap year 1996 should have problems with 2000, which IS four years later. I'd like to think that whoever put in the date info knew enough to realize that 1996+4=2000, or programmed in a formula which calculates leap years on its own by adding one day to February every four years.

The only reason, I think, (and again, I'm not an expert) that a "leap year-conscious" system would screw up on February 29th is if some incredibly stupid code jockey deliberately programmed the thing to think that 2000 was just a normal year.

-- *Rochelle. :) (rainbowdrop@usa.net), January 11, 2000.


I do know a mainframe programmer I talked to (insurance company) about Y2K wasn't worried about that, only 2/29, which she was working on.

-- Mara (MaraWayne@aol.com), January 11, 2000.

Thinking about it, this would be a big payroll accounting problem if not remediated.

-- Mara (MaraWayne@aol.com), January 11, 2000.

My dear Mr. Imso,

Sir it seems that every one (save myself) has had a turn at predicting both near, mid and in some cases far term dates, as they would relate to Y2K influenced failures.

To say that the embedded system problems are finished is like wishing in one hand and taking a dump (vulgar expression; but apt for this circumstances) in the other. And seeing which hand stinks first.

With two Saturn plant, one ford plant off line, (and at least the Saturn people saying that they have production problems).

Two chemical spills ( one of them licquid nitrogen). I believe at least four oil refineries down or significantly reduced it production. The telecommuications in trouble it Austrilia. And of course, the rare (but now regular occurance) of voltage surges/spikes in our power systems.

There are way too many other occurances this past two days for me to record here. But I think that you get my drift.

Now kind sirs and gentle ladies. My humble attempt at making predictions. You may consider these predictions to become previlant (most likely to occur) at, or around these date lines. The 15th to 17 th of January. The 1st and 2nd of march; and for about a week on the March, February time line. the week of April the 1st will be interesting, as will the 15th of April.

There will be a continueing failure of parts of our oil refineries, some of those refineries will loose total production for days. There will also be some rather strange and significant causes for the shut downs of several conventional power geneation complexs (nuclear powered ones are their own special problem; and they will continue to have interuptions for the remainder of 2000).

As we get closer (and pass) the leap year dates. There will be a sudden upswing in main frame lock ups, failures of the batch systems, and total loss of "some" information in the main frames.

I believe we shall see several more strange train wrecks, close to digital controlled crossing. We will, by mid February or the 1st of March, see the first visible evidence of JIT failures.

By the fourth week of January, the horned haired powers that be, will know that we, globally have some rough water to negoiate. And spin control will once again be the agenda of the day!

I won't even try and make a guess as to the date of the stock market crash! For it has gone on far longer; and to further heights than any one could have imagined. I see also that the mid south and mid west grain belt will suffer "Dust Bowl" conditions again this year.Affecting our wheat and corn production. I see more turmoil ( this time bloody) in the South American countries. There will be several more banks hit with software (2yk) problems...And we'll hear more "I don' know what caused it! But it sure wheren't Y2K boys"!

On the global scene and not Y2K related directly: I see that Russia will use at least one tactical nuclear device in, around or on Grozny. In order to bring to a successful (if it can be called that) conclusion to their "internal" conflict.

Towards the end of this year, I see the beginnings of infomagic's nightmarish Charlott's Web senerio. Remember! Y2K is but a "trigger" for all else that is to come.

~~~~~~~~~~~~~~~~~~~~~~Shakey~~~~~~~~~~~~~~~~~~

-- Shakey (in_a_bunker@forty.feet), January 11, 2000.


Shakey,

any specific thoughts concerning February 29/December 31 ??

-- George (jvilches@sminter.com.ar), January 11, 2000.



Shaky:

Thank you.

-- ImSo (happy@prepped.com), January 11, 2000.


The Real Time Clock chips I'm familiar with (quite a few) are very simpleminded about leap years. They think every year divisible by 4 is a leap year forever. Unless this changes, the year 2100 might cause some problems.

366-day year problems are at a much higher level than the hardware clock chip level. Any problem not recognizing February 29, 2000 is in software, since the RTC chips know all about it (except in 2100, of course).

-- Flint (flintc@mindspring.com), January 11, 2000.


Shakey, it's complete assholes like you that spread unnecessary fears about the y2k hoax in the 1st place. Why don't you do everyone a favor and just disappear of the face of the earth.

Mr. Sane

-- Mr. Sane (hhh@home.com), January 12, 2000.


I insist in that the February 29 issue is not that benign, let alone that "clear" as some people think.

Take care

-- George (jvilches@sminter.com.ar), January 18, 2000.


In thread below

"UK tests show that date February 29 would cause more disruptions than Jan 1 "

Wondering postulates that going back to 1972 would not solve all the problems as many computers can only go back to 1980, etc.

Any comments? Ed Yourdon ?

Take care

-- George (jvilches@sminter.com.ar), January 19, 2000.


Only one expert has responded so far (good 'ol Flint) and his answer is pretty wishy-washy (sorry Flint).

Dale Way, Tom Atlee, Dr. Gordon, and Ed Yourdon are conspicuously absent from this thread. Is it that the Feb.29 should be completely dismissed? At the very least does it not compound on top of the Jan.1 rollover problems?

(Sigh)

-- George (jvilches@sminter.com.ar), January 19, 2000.


Moderation questions? read the FAQ