I'm really getting tired of hearing the "Nobody knows what will happen in Y2K" mantra. This has become an oxymoron. Of course, nobody know exactly what will happen as a result of Y2K glitches, but can'nt we come up with some educated guesses?

Sysman recently did a survey and came up with the fact that there is more than 700 manyears of programming experience of the regulars that visit this site. This being the case, I think some individuals can make some pretty good estimates of what is likely to happen when Y2K glitches occur.

What I'm suggesting is this. I'm assuming that their are a good number of programmers here that have experience in many different industries. I would like to hear some estimates of what they think may happen. I don'nt have a programming background,my field is project management. But I can tell you what the programmers told me would happen in the insurance industry if Y2K wasn't fixed. Clients could be quoted the wrong rates because actuaial (sp) tables were screwed-up. They could be issued policies that were extremely favorable to the customber ( a million dollar policy for peanuts). The company's cash flow could be screwed-up etc.

I have read Andy's posts with interest re how corrupt data from one bank could spread to another bank. But Andy did not give any examples (not that I recall) of how banks would be impacted by these tranactions.

Anyway, I think it would be extremely helpful to have real life examples of what could (probably will) happen in Y2K. Thank You

-- Watcher (anon@anon.com), April 15, 1999

Answers

Watcher, I too am tired of the "nobody knows" mantra. I also believe that there are a lot of programmers, engineers and technicians who know exactly what will happen in their specific area of expertise (some good, a lot not so good). I think the problem is that the average journalist doesn't have the background to adequately report the story. When our local newspaper published a column that categorized people concerned about Y2K as "talk radio conspiracy nuts," I wrote the following reply (If some of my comments seem off- topic, e.g. talk radio or politics, remember that they refer to statements in the editorial):

"Dear The Peoples Forum,

"Re: "Radios unidentified frying object: the Y2K meltdown" (3-6-1999)

"While Mr. Saunders characterization of Art Bell is not altogether inaccurate, Bell is no more responsible for Y2K "conspiracy-nut radio listeners" than Rush Limbaugh is responsible for the Clinton presidency. If people are anxious about the year 2000 computer problem, it is brought about in large part by the confused and conflicting messages coming from our leaders, ersatz experts and to a large extent the press, about what the Y2K problem will do to our technological infrastructure. I suppose there could be a conspiracy, but its probably just complete incompetence.

"There is not much that generates more feelings of fear and insecurity than uncertainty about ones future. Yet almost every story on the millennium bug includes a quote from a governmental official or IT professional reinforcing that uncertainty. In essence they all say that nobody knows what will happen. The stories tell us that various devices may fail, but not which ones will fail. While public reaction may be unpredictable, the technology is not. If problems are not fixed yet or cannot be fixed in time, tell me what will and will not work.

"It is true that no single individual knows what will happen with every computerized system. But the programmers and engineers who designed and built the equipment should know. Those who monitor, maintain and repair the equipment on a daily basis should know. The technology at question is not something handed down from an ancient civilization and no longer understandable by modern man. Most of the original programmers are still alive, and most of the hardware manufacturers are still in business. There was even a stack of electronic component data books at Readers Corner a while back.

"Plenty of people know what will happen with their technology next January. If The News and Observer and other members of the media would report that information, folks lying awake listening to Art Bell might finally get a good nights sleep."

-- Robert Neely (robert_neely@ncsu.edu), April 15, 1999.

Hi, Robert, we cancelled our subscription to the Raleigh N&O last year. Way too lefty for me and I'm pretty tolerant. Mind you, the more conservative Durham Herald-Sun is nearly as bad.

Have you been in touch with Critt Jarvis at middleground@critt.com, who's in the process of creating a web site and forum for the Carolinas?

-- Old Git (anon@spamproblems.com), April 15, 1999.

Watcher,

As a programmer, I can construct an example of two programs with the same Y2k error which have only two small nonerroneous differences between them. In one case after the Y2k error there is only a one-to-eight-cent error in bank account balances, while in the other case the aftermath of the Y2k error is almost any result, such as data base destruction (within the range that the bank program can produce), that one cares to contemplate.

A programmer could predict the result in the first case with precision in a matter of seconds, but in the second case it would be more cost-effective for the programmer to say "I can't tell what would happen" than for s/he to track down the thousands of possible combinations of consequences.

-- No Spam Please (No_Spam_Please@anon_ymous.com), April 16, 1999.

robert,

you are correct- y2k is way beyond the average journalist, because it is not only technological, it is economic/financial/business as well. plus it's international. add to that the obfuscation and spin from many (though certainly not all) quarters. it's the biggest headache anyone could imagine.

-- Drew Parkhill/CBN News (y2k@cbn.org), April 16, 1999.

Assuming the power generally stays up, most banks stay open, and the phones work most of the time:

1) Thousands of database conversions will transpire over the millennial weekend. Some of these won't go well.

2) The bulk of Y2k testing for a number of systems will begin on 1/1/00. The results will be interesting.

3) Archived data will be very difficult to restore and access because virtually everyone will have overlooked converting their archives and bakcups to the expanded format. This situation will come as a complete surprise to a number of shops.

4) Rooms will begin to fill will countless slips of paper recording transactions that cannot be entered into wayward computer systems. The management of these backlogs will eventually make or break a number of companies.

-- Nathan (nospam@all.com), April 16, 1999.

Robert,

If Sysman is correct about forum members having over 700 manyears of experience, I would think we could get some educated estimates from these individuals, without waiting for the news media. How about some input guys!

-- Watcher (anon@anon.com), April 16, 1999.

Sorry, Watcher, but the reason nobody knows is because nobody has the big picture with all the interdependencies mapped out. Each programmer can only speak with authority about his/her specific systems or technology. The economic/social/political consequences of Y2K can only be predcited by compiling the risk of these technical problems along with a representation of their interdependencies (and factoring in human reaction).

It can't be done. There are too many variables.

Nobody has a clue how this thing will play out. In fact, I think the uncertainty will grow as we approach the rollover, instead of it shrinking. I've stopped guessing about the consequences because (1) my guess is probably wrong and (2) it won't change my preparation plans, which makes the exercise mostly pointless.

-- Codejockey (codejockey@geek.com), April 16, 1999.

Codejockey,

Thanks for your response. I agree that nobody knows the total big picture re Y2K. There are so many interdependies that it's a planner's worst nightmare. Please let me make myself clear. I was not asking for the Macro view but rather the micro view- a view based on a programmer's individual industry experience. For instance, I'vbe read your posts with interest. What industy are you in? What programs are you working on? Will they be Y2K compliant based on your experience? What will be the impact of non-compliance based on your knowledge and experience?

-- Watcher (anon@anon.com), April 16, 1999.

Watchful, After I re-read my last post I realized that I came off sounding grumpy. My apologies. I'm not grumpy, but perhaps fruststrated that there are better answers than "nobody knows". I've been a contract programmer for the last ten years doing client/server and web development. I know that everything I've written since 1996 has had four-digit date considerations designed in. But more importantly, non of the systems I've worked on in that time frame would be considered mission-critical.

I've worked at the Federal Reserve as recently as a year ago. They run a tight ship.

On the other hand, I understand how easily a system could be non-compliant, and how much data corruption even the smallest amount non-compliant code could create. Computers are cold and unforgiving -- they won't question your instructions, but instead carry them out with lightning speed, no matter how destructive.

With a problem as pervasive as Y2K, everything needs to be tested and re-tested. Then tested again. I know we (the world) don't have time to test as much as we should. There will be a lot of testing in production, which will in many cases result in business disasters.

This I can say -- whatever doesn't get fixed in 1999, will be extremely difficult to fix in a timely manner in 2000.

-- Codejockey (codejockey@geek.com), April 16, 1999.

codejockey,

you're right on the money about the variables. i've told people for a long time- maybe a year- that y2k is statistically unforecastable, because the number of variables is literally infinite. and even with finite variables, forecasting is tough.

-- Drew Parkhill/CBN News (y2k@cbn.org), April 16, 1999.

Everyone,

I understand how the complexity of Y2k makes it impossible to see "the big picture." I certainly don't have enough brain cells left to contain an object that large. Nor is there a system powerful enough to run a simulation of that complex a model.

But I would like to second Watcher's request. If those of you with real-world experiences will contribute pieces of the puzzle so to speak, we may have a clearer, albeit still fuzzy picture of what will happen.

-- Robert Neely (robert_neely@ncsu.edu), April 16, 1999.

Nathan,

You have a very good point about the archived data. For two reasons. Many companies are doing OS upgrades or switching entire OS platforms. When this type of change occurs, you may not be able to read your archives any longer. This could be some serious problems for companies when they realize that bad data has entered their systems and they need to recover to a prior date.

The second reason is worse. Say you are a hospital. Some former patient has problems and wants to sue. You have to be able to produce the records or you are in deep crap. I think that many of the hospitals that are trying to fix their systems are overlooking this fact. If time goes by, they may forget what the format of the records are that are archived, and thus not be able to provide the actual information in any form that a court could recognize this. The health care industry needs to wake up to this fact, and the sooner the better.

-- (cannot-say@this.time), April 16, 1999.

Good evening Watcher. The total from the survey is now at 875 man-years.

I think most agree that power is the number one concern. Without power for a long period, we're in big trouble.

My current job is systems programmer and network administrator for a company that provides services to publishers. We do photo-typesetting, CD-ROM mastering and web-site develpoment on the back end, and write their data base programs, formerly using Clipper and Fox-pro on the front end. All of our new development, including conversion of one of our existing accounts, is being done with web-based technology, SQL server, Java, Visual Basic and the like. None of our work can be considered critical, as far as public safety or well being.

We have already suffered from Y2K for a couple of reasons. First, we are all very busy converting our biggest customer off of the mainframe, because of their extensive use of dates. We figured it would be easier to rewrite the system, than to attempt to patch the hundreds of Cobol and Assembly programs, many going back almost 20 years, that have been worked on by countless programmers over the years. We are also busy fixing our other big customer's programs that will remain on the mainframe. As a result, we have lost some existing accounts, because we do not have the time or resources to fix their systems. They have found other service companies that can deal with their problems. For the same reason, we have had to turn away some new business.

As far as our #1 customer, we will not be 100% finished with the conversion when their summer 2000 production is scheduled. We will have enough done to get the basic products out the door, but they will be missing some details that were in the 1999 products. They understand this, and they are willing to work with us. They also understand what would have happened, had we attempted to patch the existing mainframe system. <:)=

-- Sysman (y2kboard@yahoo.com), April 16, 1999.