From Cory Hamasaki's Most Recent...FWIW

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

Cory put this in his most recent WRP BEFORE the rollover...who ya gonna put your faith in? This guy has his stuff together.

------------

Year End Assessment 2

InfoMagic returned with his final assessment. While Info and I don't always agree, I find his insight interesting and in some ways optimistic. (Info's assessment has been distributed to members via the member's listserv.)

Info expected the sheeple to realize that there is a real risk as Y2K and the zeros hit the large production systems that have been chugging away for years, decades in some cases.

Well, Info was wrong. Systems failed but no one reacted to the various signs. Even when companies like Whirlpool, Hershey Foods, Bang and Olufsen, Samsonite, Waste Management, and others had trouble installing software prior to Y2K, no one sounded the alarm, told people, that this is one of the seven lean cattle.

When the Wall Street Journal reported that Chase Manhattan misplaced 40 billion dollars in their BondMaster system, no one pointed out that the fact that Chase couldn't figure out where the system was going funny; why that shows how hard it is to debug some systems.

Here in DeeCee, the city management is SCREAMING that they have everything under control, but they've shut down the city STAT-LAB due to computer problems, moved critical financial processing out of the area, and are advising longer than 3 days of preps.

The TeeVee news has been running clips of the generators (yes, these are locomotive sized) outside of various government buildings. They showed the tank trucks parked near the generators. They are getting ready for much more than a bump in the road. In fact, they're scaring me.

So here's what's going to happen. There will be multiple, persisting, enterprise systems failures. The big old systems will get very strange.

These will not be fixed in a few hours, days, weeks. Some will take longer than a few months to repair. In some cases, the systems will be in chaos for a year or longer. This was the case with the DoD's DSS.

I will be freezing my rates next year. Even though I have received requests from three clients for assistance and I could increase my rates or go with the highest bidder, I will divide my hours between the clients. Two are long time loyal clients who paid top dollar during the early 1990's when there were lots of programmers and few jobs.

If we can keep the major systems running, we have a chance to keep everything from tipping over. I'm especially worried about banking (the three incidents at Chase, the Wells Fargo problem, and the problems at DeutchBank don't help), insurance (problems with underwriting, renewals), and warehousing, order entry, bills of lading, all problems that we've seen in the SAP flubs.

[PAY SPECIAL ATTENTION FOLKS]

I'm not that worried about embeddeds. I suspect that a lot of the embedded issue falls into the same category as microwave ovens and planes falling, that is to say, polly disinformation that escaped to the wild.

If there is an embedded problem, it would be in controllers in larger machines. I've seen old IBM AT's and DEC Rainbows running microbiology lab instruments. If these fail soft, produce slightly incorrect results but keep running, people will die. I saw machines doing diagnosis, identifying micro organisms, and other life critical applications.

We'll see. The embedded problem, if it is a problem, will happen fast and will be flashy. We'll know in a week or two.

It's the enterprise systems that will fail over months, corrupt databases, kill organizations, put people out of work. I'm worried about the big old systems. Some big systems run daily but most are weekly, monthly, quarterly, or annual

-- Not Cory (just@very.thankful.he's around), January 02, 2000

Answers

Excellent catch Not Cory!

"Even though 10% of the world is experiencing minor glitches randomly in the basic infrastructure now, it should be noted carefully by all IT professionals and computer experts that the other 90% of the system hasn't yet been fired up in real-time testing. As people will go back to work on Monday fire up those systems and then we will soon see what kind of interconnectivity problems surface as the whole system is then put into full operation. If we start seeing snowball effects happening over the next few weeks we wil soon know how far the system can cope without collasping on itself.

The small glitches that we think are 'minor problems' now can easily at any time start escalating through the infrastructure, so its a waiting game to see how many of these glitches will surface. If there is enough of those little tiny glitches or outages happening all at once this can clog up or collaspe the infrastructure and grind it to a halt, but only time will tell if the glitches start increasing in their accelerations or remain at a stable level that the system and the programmers are able to cope with, for this is the big hidden danger of the y2k issue now.

I am sure there will be an increase in the amount of glitching when the other 90% of the code is put into operation when people go back to business as usual and with those systems that couldn't get their embeds upgraded because not enough time, in the meantime last minute fixing is probably being done during the holidays to try to turn back the glitch bomb before people go to work. We must also take into account that this is where the "non-compliant or incompleted systems" around the world start to play the part in provoking a interconnecting glitch snowball effect for nearly half the world's computer systems did not make it to the deadline of 31st Dec 1999 and so technicians hope to try to fix things in 2000 after the glitching starts for the danger is far from over, if the glitches snowball, it may outdo the number of technicians and programmers to fix it, it's just the beginning, wait and see.

Now it all depends if there is enough glitching to outdo the number of programmers patching the systems up, this will be quie interesting. The bomb has gone off, but it hasn't quite detonated fully yet, its a Year 2000 Timebomb, not a Day 2000 Timebomb. So many think that because everything in the basic infrastructure is still operational so far with only minor glitches that this y2k bomb is only a fizzler.

We will know if the infrastructure stands once it starts going into full real-time testing, we can't know if the system is going to pass the test until the whole thing is tested. Before we celebrate thining the y2k bug has been beaten, not everything was put in full operation on New Year's celebrations, because we see the power and phones still working or basic utils, this don't mean that the whole system has jumped the hurdle over y2k.

TO SUM UP =========

The immediate effects of y2k are not really immediate but over time, its a timebomb, not a daybomb. Not until the whole system is put into full real-time operation will be begin to know of the ramifications of y2k over the next few weeks. y2k is not a day event, its a year event, possibly even longer. Pollies can gloat all they like about the rollover event, for it dosen't matter to the billions of glitches and unfixed systems around the world, they don't take day events. If there are problems with embeds, they might take a while to show up. The non comp ones though would probably crash as soon as they are fired up. But the y2k in the software code plus the billions of glitches will prob slowly detonate over time.

Brent N"

-- Andy (2000EOD@prodigy.net), January 02, 2000.


Hang on to my preps?

-- Gary (nada@rollover.gov), January 02, 2000.

Yup. For a while.

-- Truk (truk@loa.moc), January 02, 2000.

If there are problems with embeds, they might take a while to show up.

And this statement is based on what? Experience? Knowledge of how they work?

-- Cherri (sams@brigadoon.com), January 02, 2000.


Moderation questions? read the FAQ