Y2K:a mathmatical certainty

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

TEOTWAWKI reconcidered


If you have a problem with that link go below for a hot link


-- Maggie (aaa@aaa.com), October 07, 1999


correct link


-- Maggie (aaa@aaa.com), October 07, 1999.

The weakest part of this argument is that it treats all failures as permanent. If you just multiply up the odds of individual system failures and assume they all last forever, of course you can get high odds of overall failure. Dressing this up with some probability tables makes it look more substantial than it is.

In the short term, these odds may be more or less right. Many of the failures will occur simultaneously, and many companies will be out of operation for at least a while. But the real affect of the economy depends on how long they stay that way. There are workarounds, and some amount of repair work will get done.

Of course, there will be a shortage of programmers. And very limited access to help lines and web sites due to congestion. And infrastructure problems can paralyze a whole region, making repair impossible until utilities are restored. But this article does not consider any of that.

Take a real world example -- when a hurricane hits, it knocks many business out completely. Not just with system failures, but with physical destruction. This does not unravel the whole world economy. The affected areas do eventually recover. Even in parts of the world where very little outside help arrives, there is recovery.

It would take a lot to produce a real end-of-the-world scenario. This article is absolutely no help in predicting how much it would take or how severe Y2K will be.

-- You Know... (notme@nothere.junk), October 07, 1999.

"...expect the old system to be ditched almost entirely and a new system to emerge."

-- (@ .), October 07, 1999.

Coment from Gary North about this subject:

Subject:The bigger they are,the harder we fall:chapter 2000 bankruptcy

Comment: The author of this web page begins with the premise: a 1% complete failure rate that shuts down a critical system is not farfetched.He takes this idea through a series of mathmatical exercises to conclude that large,complex systems are vulnerable to complete shut downs.The social implications of this are frightening.He understands the division of labor.

This analysis should be concidered in light of the Beach/Oleson pain index of system failure.

Link previded in post above.

-- Maggie (aaa@aaa.com), October 07, 1999.

"But the real affect of the economy depends on how long they stay that way. There are workarounds, and some amount of repair work will get done."

Very true. I agree that the analysis is simplistic and flawed. The real debate is over the number, extent, and duration of the critical system failures that occur: Can they be fixed or worked around before the company goes belly up or is severely financially damagedor before an agency is unable to fulfill it's mission?

Failures don't need to be permanent in order for a business to fail, with resulting unemployment -- they only have to last a relatively short time. Will all the failures that cause loss of a critical system be fixed in time? Can work-arounds be implemented in time? Are they efficient enough?

This is the key issue that separates optimistic observers from the pessimists.

When the rollover occurs, every company or government will be operating in Fix On Failure mode. It's a "come as you are" war. Some failures probably will be easy to correct, others may take weeks and even months. It's this last category that will sink companies and the economy.

-- de (delewis@XOUTinetone.net), October 07, 1999.

* * * 19991007 Thursday

The possibilities and complexity of FOF strategies FAIL to consider-- as "unthinkable"; I know, I've heard, many times, management dismiss this scenario in Y2K planning meetings--the lack of one or more critical infrastructure ( e.g., power grid, water, waste treatment, telecommunications, financial ) segments/components.

Power brown-outs and black-outs are effectively differences without distinctions. Operating electrical motors or electronics is impossible in both scenarios. Electrical components will fry.

There aren't enough deployable generators or logistics to provide long-term fuel production/supplies for back up power.

The USAF bases have been running and evaluating multiple exercises for just such contingencies since March of this year (1999)!

This is the scenario that will nigh make POST-Y2K remediation/replacement of electronic components and computer code a veritable nightmare of 5-15 years in duration.

Management and politicians don't want to hear or consider such scenarios: "_They_ (utilities) wouldn't possibly let it (Y2K) come to that!"

There are unprecedented events (i.e., Y2K) that can and will trump even the assumed almighty motivation that utilities are business that must make profits.

With evidence of Y2K failures piling on, day by day, due to management negligence and ineptitude, the probabilities are increasing exponentially that there will be major, global show- stopper disruptions in the critical infrastructure. These disruptions will put a real crimp in the best laid plans of "sheep" and men.

Get it?!?

Regards, Bob Mangus

* * *

-- Robert Mangus (rmangus1@yahoo.com), October 07, 1999.

The strongest point made by Bob Mangus is that management is inept. If you don't get it, open your eyes and LOOK around you... think of all the pin-headed decisions you were forced to work your way through to the point where mgt. finally had to admit a harebrained idea and give it up. I'm no programmer, and I don't even know that much about system stuff - never wanted to learn. That's the job of MIS and I tell them so. (Hey, bozo; I don't ask you to learn how to illustrate so you can fix my stuff - I've got enough to contend with - if I learn your job too, what do we need YOU for?) But the Peter principle has been at work for many years and I've known many who were promoted not only to their level of incompetence, but on to the extreme.

-- Patrick (pmchenry@gradall.com), October 08, 1999.

Yep. Virtually certain that the year 2000 will come, on schedule...

-- Mad Monk (madmonk@hawaiian.net), October 09, 1999.

Moderation questions? read the FAQ