Y2K-RELATED FAILURES only take a few minutes to FIX

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

*Some bugs turning up now, but expected to level off soon Most people assume that Y2K problems will just happen on January 1, and then go away or be fixed. What they don't realize is that some bugs have already started to show themselves. On a bright note, the Gartner Group's Lou Marcoccio notes in the article that 97% of all Y2K-related failures take only a few minutes to fix.

-- thinkIcan (thinkIcan@make.it), October 28, 1999


The sourse of this article is the Salt Lake Tribune, but I found the article on the CNNfn site

-- thinkIcan (thinkIcan@make.it), October 28, 1999.

Here is the full article -------------------------------------------

Y2K Is Months Away, But Bugs Turning Up Now Source: The Salt Lake Tribune

A lot of people are gearing up for a Y2K computer bug that will strike on New Year's Day, but it probably isn't going to happen that way.

According to the Gartner Group, only a small percentage of actual failures will occur with the change of the millennium. In fact, the pace of failures has picked up since Oct. 1, and will probably continue to rise into early next year.

"At this point, most companies can handle the increased failures, but just barely," said Lou Marcoccio, a research director for the Stamford, Conn.-based firm.

"The problem we have is that on Nov. 1, it will go higher and in December, dramatically higher," as computer programs come across more 2000 dates, he said.

The failure rate will probably stabilize in January, then drop off later in the year.

Companies that have had their systems remediated are still likely to have some failures, though far fewer than those organizations that have done nothing, Marcoccio said.

He recommended that organizations expand their Y2K contingency planning beyond the hours before and after New Year's Day.

What might people expect to result from failures? As an example, Marcoccio said, an airport testing its emergency and fire alarm systems found a bug that caused all of its security doors to be unlocked.

"The point is, if they didn't find it, what would've happened?" he said.

The vast majority of failures are fixed quickly, Marcoccio said. About 97 percent of failures take only minutes to repair. But of the remaining 3 percent, some failures take three days or longer to fix, causing potential business disruptions.

Gartner was one of the early firms to identify the magnitude of the Y2K problem, in which computers using two-digit shorthand for a year confuse 2000 with 1900. The firm has forecast that it will cost between $300 billion and $600 billion to prepare the world's computer systems for the glitch.

Gartner said system failures related to the computer glitch will probably cause only isolated problems for people in developed countries including the United States and in most of Western Europe.

Publication date: Oct 23, 1999 ) 1999, NewsReal, Inc.

-- thinkIcan (thinkIcan@make.it), October 28, 1999.

That could very well be correct. Fixing a date/time related failure in a thermostat or elevator control might take nothing more than turning it off and on again. Fixing a date/time related failure in a computerized valve might take the same thing.

But if the valve is the cooling water flow in a nuke plant, the power down and power up might be real risky. You might not KNOW of the failure until the stuck-open valve had already damaged something else.

It's that other 3 percent that can mess us up. That will be your big databases that run social security, or the IRS, or medicare. That will be the scheduling system that runs the trains, or the stock ordering system that makes the groceries move from the farms to the stores. That will be one of the FAA systems that they claimed was compliant and now turns out not to be, and will be down for six months or a year.

There are a million thermostats for every General Motors. We could fix 99.99% of the failures and still be in a world of hurt. It's not the ratio of easy fixes that matters, its the impact of the tough fixes.

-- bw (home@puget.sound), October 28, 1999.

Sounds wonderful on fantasy island. I never repaired y2k bugs in just a few minutes and I've been in the IT business for nearly 20 years. No experienced programmer would update a system this quick without fully understanding the scope of the alterations. We've been punished too often for making quick changes. This is a bunch of bunk.

I heard this fact on that stupid watered down PBS program lately. I had to laugh. I would say, only 3% of the time you can fix something quick and 97% of the time it will take awhile (who knows??). He's got his numbers switched.

I wouldn't have bothered making any y2k repairs at all if it just takes a few minutes to fix these problems. If I listened to this crap my systems would be down right now as we speak.

-- Larry (cobol.programmer@usa.net), October 28, 1999.

Larry, I've been doing COBOL for 25+ years. Nothing on a mainframe gets fixed in 3 minutes or 30. Nothing that you care about gets fixed in less than a day, and if you did it that fast you darn well better spend the next week checking results.

The really big problems are going to take weeks or months. The IRS has been trying to clean up systems for 20 years, right?

All they're doing here is taking the simplest, fastest fixes, which in the case of a thermostat, etc, is a quick fix. And then they can say with a straight face that 97% of the time it can be fixed fast. Well, that's true, it's just not honest.

-- bw (home@puget.sound), October 28, 1999.

bw -- I agree, but the article (and what I heard on TV) was that the
"The vast majority of failures are fixed quickly, Marcoccio said. About 97 percent of failures take only minutes to repair. But of the remaining 3 percent, some failures take three days or longer to fix, causing potential business disruptions."

This is totally false. Nothing is stated about just "easy fixes." That's not what he's saying.

-- Larry (cobol.programmer@usa.net), October 28, 1999.

I have to agree with you two. Not in the mainframe arena but in the client/server arena (Windows NT, 95, 98, etc). Have yet to see a problem that could be figured out within a few minutes.

Not unless you have seen the same problem 10-20 times before.

-- STFrancis (STFrancis@heaven.com), October 28, 1999.

What is going on here? I cannot believe my eyes. People are fighting over the unknown. All sides need to step back and look at the real picture. The REAL picture is that it does not matter if this is going to happen or not. What does matter is that if you feel there is a problem, or not you have to do what your heart, soul, and mine tells you to do. What this whole thing boils down to is as simple as all that. No matter how many times you butt heads with each other, or other people it will make no difference in the long run. Everyone is entitled to his or her own idea. If you had an old car, and it was running fine, but you thought it might break down, you might go out and get a new one. Your neighbor will either think..Hey nice car, your old one needed replacing, or what was wrong with the old car? What really matters is what you think is best for you and yours.

This whole thing about Y2K is a no win situation. For the y2k'ers if this happens, who is going to be able to rub it in. Do you laugh at Flood, Earthquake or other bad things when they happen, and say you should not live there, or you should have left. NO!!!!And you so called Pollies, how about you? Ha Ha you spent your last dollar trying to save your family for nothing. What is happening to people. No wonder we can't fix this Y2K problem, we are to busy trying to find a place to put the blame.

I am no saint, but there is too much hate, and dislike in this world now. If this country can't start pulling together to fix the wrongs we never will.

We let big government pick up the bills on our poor, homeless, and everything else that they will, and complain all the while. Well fellows put someone up there by voting them in. Don't just sit back and whine. We don't vote for the party anymore, just the lesser of the evils.

If we over prepare, and nothing goes wrong (and I pray to GOD that nothing will.) we could all give something to people who have nothing. To the homeless children, the hungry, or just the down on his luck Joe Doe.

Everyone had better stop wasting his or her time with he is wrong or right and do what his(her) soul tells them to do.

If everyone thought the same, there would be no Brave New World. If you tell someone they are not doing for their family what you think they should be doing, what kind of response do you think you are going to get?

I for reasons in the upper posting am not going to give my real name, but I will say that for the most part found TB2000 a useful page. For when you can learn to support your own self for whatever reason, you start learning who you really are.

I for one think that a little self support never hurt no one. But my friends and neighbors will have to learn this for themselves. I am willing to help anyone if they ask, but ask they must. If you tell someone to think for themselves, that does not mean think like you do.

Use these posts in a reasonable manner, and people will listen, use it as a pissing contest, and there will be no winners.

Just A Thought

-- Just A Thought (justathoughttoo@Yahoo.com), October 28, 1999.

Just a thought,

What the hell are you talking about, it appears to me that the posters on this thread are in agreement with each other. They don't seem to agree with the author of the Salt Lake piece but I don't see any reason to go off on them. I've seen a lot of threads where your final offering would be more appropriate. Love and kisses!

-- Grateful (nofighting@here.thread), October 28, 1999.

I am sorry, for whatever reason This thread got posted to the wrong thread. This was suppose to be posted to Stop the Presses. I am sorry for posting this on the wrong site, and it should teach me not to post to sites at all. Again to all on this site, I am truly sorry

Just A Thought

-- Just A Thought. (justathoughttoo@yahoo.com), October 28, 1999.

Nice posting, just a Thought. I admire your altruism and generosity. But what heppens if there are thousands who ask you for a hand and food... And what if the problems and food shortages go on for months? Consider. Davo in Oz, waiting for it.

-- Davo (now@yjose.com.au), October 28, 1999.

But "FIX" does not equal fix, test, integration test, and install... Yes, code changes can be made quickly, but you run a considerable risk of intruducing further errors if appropriate testing does not follow.

-- Mad Monk (madmonk@hawaiian.net), October 28, 1999.

Just have your checkbook ready, and enjoy the new millenium!!

-- Robert X. Cringely (I_left_my_brain-in_S@n_.Francisco), October 28, 1999.

Larry and bw -- Keep in mind, you are both looking from the mainframe perspective. A whole lot of code (and bw, if you been on the hot seat for 20 years you KNOW this is true) would look real natural with a nice marinara sauce over it. That being true, you don't *DARE* do a fix without REAL careful consideration of the potential consequences in some code 88 zillion lines away that nobody in his right mind would have ever thought had anything to do with this, except the moron who originally thought that this would be a handy place to stick this algorithm.

That is what a Y2K problem looks like to you. To a client/server guy, it looks a lot like a twenty line script. (sometimes ;-)) And, even then, he may not find out til 54 days later that that twenty line script was actually doing double duty, and the 'fix' he put in to correct some rinky-dink two-bit problem with an output screen has been happily cranking away trashing database records in it's REAL job.

And to an embedded system designer, it may look like a reason to change jobs. Since his program, as Flint put it, is now a piece of silicon and is 'uploaded' by means of a soldering iron.

And to an end user, the problem may look like a railroad switch, that will divert the speeding Amtrak train from the track with the ammunition special, and he is looking for the hole to put the switch bar in, but there isn't one, because they removed them for maintainance costs, and now what is he supposed to do?

Keep in mind, that this stuff sort of depends on your perspective.

Just a Thought --

Don't quit posting because you got the wrong thread. (Actually, though, given the nature of your post, are you SURE it wasn't deliberate? Cause you would have to have the devil's own luck to find a thread where the sentiment expressed didn't apply to at least a couple of the posters. ;-)

-- just another (another@engineer.com), October 28, 1999.

just another - you are absolutely correct. I'm a mainframe and PC guy, I know the implications of inernals problems, the lure of quick code fixes and the pain of finding out it was a bad idea. I'm almost clueless regarding embeddeds and C/S systems.

That's why I come here - I learn all kinds of stuff from other people, which I couldn't pick up any other way, given the free time I have. Or rather don't have.

Hang in there, y'all, 1500 hours and counting.

-- bw (home@puget.sound), October 29, 1999.

Moderation questions? read the FAQ