failures for andy ray : LUSENET : TimeBomb 2000 (Y2000) : One Thread


The Reality of Y2K Failures By Michael S. Hyatt July 7, 1999

Amazingly, some people still think Y2K is a hoax. These people are obviously in deep denial. The number of reported failures are continuing to mount, and these are likely only a fraction of those that have occurred.

Every now and then I encounter someone who thinks that Y2K is a hoax. These people assert that there have been no documented Y2K-related failures and that the whole issue has been concocted by computer consult-ants, preparedness products suppliers, and authors trying to sell books (yours truly). Unfortunately, because the masses are so eager to believe anyone who will assure them that their present prosperity will continue without interruption, these self-appointed debunkers often go unchallenged. As a result, many naove souls are being lulled back to sleep, believing that all is well and there is no reason to make preparations.

Nothing could be further from the truth. This Y2K-is-a-hoax perspective is, at best, a result of extremely shallow research and, at worst, a product of deliberate disinformation. Either way, it is untrue, unhealthy, and danger-ous. Those who believe it and do nothing to prepare are putting themselves and their loved ones at tremen-dous risk.

To counter this viewpoint, I'd like to offer a brief catalog of some of the more public failures. Make no mistake: they are continuing to mount. Some are being reported; most are not.

Generally speaking, governments and corporations do not call a press conference when they experience a Y2K-related failure and announce their blunder to the world. There is absolutely no incentive for doing so. It puts their most important relationships-those with customers, vendors, shareholders, and employees-at risk. It would also hand to their competitors an important weapon that can and will be used against them. As a re-sult, most organizations remain silent, working against the clock, hoping that they can fix the problems without detection from the outside world. (This also explains, in part, why there were so few Y2K failure reports when we passed the April 1 and July 1 "millennium milestones.")

The bottom line is that the only failures bubbling to the surface are those that cannot be contained. Based on e-mail reports I receive routinely from Y2K project managers, I estimate-and this is a guess-that for every pub-licly reported failure, there are at least ten others that go unreported. (Yes, I know, some of my colleagues will accuse me of being too conservative.) Regardless, here is a list of failures for your consideration. You may click on the links to get the full story. July 1, 1999: British passport crisis blamed on Y2K glitch. The British government's Passport Agency has been struggling to issue thousands of passports during this peak vacation season, and Taskforce 2000, an independent Y2K watchdog group, claims that Millennium Bug failures are the reason. A new Y2K-compliant computer system was installed by Siemens in some passport offices, but the system is not completely operational and has yet to be fully tested. Ian Hugo, the assistant director of watchdog group Taskforce 2000, said, "[The Passport Agency] said in March that they had halted introducing the new system but would be ready by June. In June, the situation was no better. Something quite clearly is wrong." June 22, 1999: Pueblo 911 system fails Y2K test. City Council members were informed by the Police De-partment that the 911 emergency telephone system failed to function properly when the internal clock was rolled over in a routine Y2K test. Had this system not been tested, it would have put at risk citizens in need of emergency assistance. Computer staff members tried to modify the AT&T system but were un-successful. The estimated cost of replacing the city's 911 system is $160,000 and will be covered by a federal grant. Other cities have also reported similar experiences. June 6, 1999: New county computer issues $50 million checks. Fred Rice wondered whether Monroe County, Indiana, had joined the lottery industry when he received a check for $49 million and one cent in mid-June. The county prosecutor's office recently installed new computer software to fix the Y2K problem and no one noticed that Rice's $49.01 check turned out to be a bit larger than it should've been. "It's almost enough to retire on, isn't it?" Rice said with a grin. Monroe County official Carl Salzmann said, "The computer guy is coming to fix it." He asked any other citizens who may have received eight-figure checks to return them and new checks for the correct amounts will be issued. June 17, 1999: Y2K test causes massive sewage spill. Over 4 million gallons of raw sewage spilled into the streets of Van Nuys, California on June 17 during a Y2K test of computer equipment at the Tillman Water Reclamation plant. Unbeknownst to plant officials, a computer closed a key gate during the test which resulted in the blockage of a major sewer line serving the western San Fernando Valley. The sew-age then backed up, began flowing out of a utility hole, and poured down Woodley Avenue into Lake Balboa Park. Plant Manager Bob Birk explained, "The computer didn't tell us it closed the gate." Crews worked through the night with vacuum trucks to clean up the mess. May 31, 1999: Washington Post bitten by the bug. The Washington Post doesn't want its subscriber's money at least not for a few more weeks. Washington, D.C.'s largest newspaper is unable to accept 52-week subscription renewals because its accounting Department is not yet Y2K-compliant. "We are getting all new software and a brand new computer to handle this problem over the next three weeks," said spokesperson Linda Erdos. She predicts the paper will be able to resume offering 52-week subscriptions in July. One Post reader, expressing frustration, was quoted as saying, "Now I have to call them back in two weeks just to pay my bill." May 6, 1999: Software bugs stacking up O'Hare airport (Link is dead). The Federal Aviation Administration (FAA) in-stalled new Y2K-ready software in its Chicago area air tracking systems on May 5. However, the software was so filled with glitches and bugs, the locations of planes on air-traffic controllers' screens were repeatedly misidentified. As a result, a cargo jet came within 300 feet of a Southwest Airlines plane which had just taken off, and 100 United Airline flights out of O'Hare International Airport had to be cancelled. The FAA was forced to reinstall the old software, which works well but is not Y2K-compliant. No schedule was given for when the new software would be debugged and brought back on-line. The air traffic controllers union president, Kurt Granger, said, "It looks like this is going to be a long summer." March 24, 1999: Y2K glitch makes premature welfare payments. Almost 200,000 welfare recipients in New Jersey received a windfall March 21 when computer tests at the state Department of Human Services accidentally paid out an estimated $58 million in food assistance funds. The benefits were electronically transferred to special welfare recipient accounts. Officials said the mistake occurred when Y2K com-puter repairs were being tested, although they could not say exactly how the April 1 benefits were re-leased ten days early. Grocery stores in Newark and other New Jersey cities were crowded with people eager to take advantage of the snafu. State officials have yet to decide whether to penalize those who cashed in and used the benefits. January 28, 1999: Y2K repair work snarls insurance claims. Two health insurance firms, Compcare HMO and Blue Cross & Blue Shield United of Wisconsin, have been unable to pay millions of dollars in claims to hospitals, physicians, and policy holders because of computer system problems. The snafus in the computer systems have been caused by Y2K remediation work. The delays began last fall and spokes-persons for the insurance companies say the problems should be cleared up by the end of March. Waukesha Memorial Hospital is owed $3 million; the Aurora Healthcare system is owed "multi-millions"; and thousands of enrollees, who must pay up front out-of-pocket for prescription drugs, have yet to be reimbursed. January 17, 1999: New mainframe computer paralyzes University of Alberta. The University of Alberta in-stalled a new mainframe computer to avoid Y2K problems come Jan. 2000, but it did not expect school operations to come to a screeching halt so soon. Glitches in the new system, which was supposed to handle class registration by phone, have produced constant busy signals, long lines of students, and mounting frustration. Thousands of students were forced to register in person, some waiting in line for four hours. The new mainframe turned out to be incompatible with the university's current phone system. No estimates have been offered as to when the problems will be fixed. January 5, 1999: Locked out of own building, executive has "new attitude" about Y2K (Link is dead). David Sterling, the head of New York-based Sterling and Sterling Insurance, says he used to be skeptical about the wide-spread predictions of Y2K-related breakdowns. On January 4 a date-related flaw in the brand new card-access security system in Mr. Sterling's office building locked him and his employees out. Apparently, when the date changed from 1998 to 1999, an undetected glitch in the software caused the system to lock up. Mr. Sterling now says he has a "new attitude" about the fragile relationship between dates, mi-crochips, and software code. The above examples are only a representative sampling. (If you are still not convinced, a more comprehensive list can be found at Year 2000 Bug Bytes. It includes failures of oil distribution software, utility billing systems, food stamp software, PC networks, pharmacy administration systems, tax collection software, military command and control systems, nuclear utility plant safety systems, and embedded systems in water treatment plants. Yet another list is at Year 2000 Problem Sightings. It's worth bookmarking both sites.)

Admittedly, some of my examples are trivial. They did not have significant consequences and could only be classified as "inconveniences." However, some of these were-or could have been-catastrophic, particularly in a Y2K environment. In today's context, we have a fully functioning foundation (FFF). In other words, prob-lems are generally local and isolated. Everything else is working, so problems can be addressed quickly and efficiently. There are plenty of resources available to resolve these matters with a minimum of disruption.

But what happens if the problems are more widespread? What if they are global and simultaneous? Last fall, the GartnerGroup estimated that 40 percent of all organizations worldwide will experience at least one mission-critical system failure due to Millennium Bug problems. In addition, by their own estimates 22 percent of the Fortune 500-the companies with the most resources to throw at the problem-do not expect to finish work on all of their mission-critical systems by January 1.

So far the problems encountered have been minimal. Generally, they have been look-ahead failures that occur when unremediated software attempts to process dates into the next century or roll-foreword failures that occur when programmers set the clock ahead and conduct Y2K tests. But these failures are the tip of the iceberg, unless a great deal of work gets done before now and the end of the year. Just imagine a scenario in which there are so many failures occurring at once that we simply do not have the available resources to address all the problems quickly. Add to that the "domino effect" and the cumulative effect of failing systems impairing compliant ones.

I am not advocating panic, of course. But unless you are hopelessly ensconced in denial, the failures I've out-lined above should provide all the motivation you need to either start making Y2K preparations or continue making them. To act otherwise, is to miss the handwriting on the wall. (end snip) --------------------------------------------------------------- sorry if that's a repeat

-- zoobie (, July 09, 1999


Hello Zoobie, Please take this in a friendly spirit, but I respectfully disagree.

I've prepared for Y2K, but as I've said in other posts several of the Y2K predictor dates have gone by with no problems that I've observed. My state started FY 2000 on Jul 1, why don't I see anything?

I don't mean to be inflammatory, but all I see are a bunch of scooped up articles from any computer system around the world that has a glitch. I've seen everything on this board from NRC boobs, to locust swarms to some failure that happens 3000 miles away that was seen by someones cousin. There are 270+ Million Americans and TENS OF THOUSANDS of systems, do you intend to rattle off every glitch seen in the US (or across the world)!

For one thing, zoobie, you don't cite many of your sources, so I can't follow up on much of what your saying.

If I may, can I give you some FIRST HAND experience..

I work for a large software vendor that makes IBM emulation products, custom systems, etc.. Here is some of what our Y2K test team has done to test for Y2K in the last two years to show 'due dilligance' so we don't get sued if something did happen during Y2K.

1. Run suites of Y2K tests against all two thousand of our central office PCs. None of our Gateway or Dell PCs to date have failed ANY of our tests (no, we generally don't use older PCs).

2. Try different OS versions like Win95, Win98, NT and beta 2000. Again, no show stopping problems.

3. Roll one of our IBM mainframes ahead past the year 2000 and leave it there for months.. Again, no significant problems on our host or software systems. It's still running at this very time in the year 2000.

4. Verify FEP processors, Gateways and Routers.. Still nothing.

5. Check our home suite of products (what we write). We found FIVE bugs in fourteen different product lines to date. All five were fixed over a year ago. NONE of them would have caused a significant problem with a customers application.

Testing continues..

As a vendor, we know that if our stuff breaks, we get sued, our customers lose faith in us and they switch to one of the other twenty vendors that make competing products. ALL vendors have been aware of this for at least two years.. If they haven't fixed it by now, they deserve to get sued and go broke. We didn't get this big by being stupid.

Our customers have taken this very seriously, because they run gigantic data centers (30K+ seats is typical). We have taken it seriously as well.. so has everyone else unless they are stupid.

I actually spent over 2K getting ready for this, but as the Y2K predictor dates slide by, and I really look at the stories on these boards, it looks to me like you guys are constantly trying to frighten yourselves.. Get prepared, then move on with life.

I reject your assertion that just because our read on this is different from yours, that we want to dis-inform, or that we are 'pollyannas', etc. It's easy to label people with a different viewpoint. What I see by looking around the Northwest and contacting other friends in the industry is that there was NO major system failure after the FY2000 date change of July 1. How do you explain that (I'm really asking a question)?

It seems to me that many of the examples you state, prove my point.. The airline kept running after finding a work around, under or over payments are fixed pretty darn quickly, air traffic control obviously works around my home state and across the US (I'm a pilot as well).

Like I said, I bought into this thing hook line and sinker at first. I respect Ed Yourdon's work, but he laid out several scenarios, from mild to severe. This board is full of people on the severe side of things. Anyone on the mild side is instantly labeled a 'pollyanna'.

Here are a few more 'what-ifs'

1. What if, we have tested this (as we have) and found nothing. 2. What if, we are career developers that have worked on dozens of systems and we are actually competent. 3. What if, we routinely fix worse bugs than date related bugs, and we enjoy our work. 4. What if, we have a staff of one hundred engineers that will drop everything and fix whatever Y2K bug percieved or observed that a customer asks us to look at. 5. What if, a good deal of Y2K has been blown up by people writing books and selling survival supplies to lower income Americans that don't actually understand computer technology very well in the first place.


-- Bryce (, July 09, 1999.

What if, a good deal of Y2K has been blown up by people writing books and selling survival supplies to lower income Americans that don't actually understand computer technology very well in the first place.

From what I see, it isnt the lower income Americans spending the money, its the people at the top (especially computer people). We will face a big problem when the rest of the non computer people start buying.

Also you mention systems that are running fine so far, that is to be expected, what about all of them that are not? How many have to fail before the city's burn?

-- BiGG (, July 09, 1999.


Why did you think that a fiscal year beginning 7-1-99 would cause major failures that would become known to the news media? You mentioned the payroll system in one of your posts, but why do you think the payroll system would fail and employees would not get their checks? I certainly don't claim to be an expert on these matters, but I would think the payroll system would feed information to the accounting system, not the other way around. Since the checks will have July 1999 dates, why did you believe there would be a problem?

-- Dave (, July 09, 1999.

When I worked for the state, (years ago), I observed that there were hundreds of programs (i.e. AR320, AR455, etc..) These programs were the 'lifeblood', if you will, of my state's information systems (at least at the time). It's not just payroll, but state purchasing, state inventory tracking (what I worked on), and several other functions. I reason that if the July 1 FY2000 were to start, and this stuff didn't work, the problems really would be widespread and enormous. We have an active press around the capital, it it would be broadcast all over the place within days.

There were 48 states that started FY2000.. I still don't hear of any widespread problems. Lots of individual stuff (but isn't that always true!), but nothing that leads to a pattern, in my opinion.


-- Bryce (, July 09, 1999.

There seems to be a lot of wishful thinking of the form: "Well, gee, since so many places have gone to FISCAL YEAR 2000, and nothing blew up, no systems croaked, the power stayed on, the water continued to flow, etc., I guess this shows that Y2K is all hype."

In the first place, the only potential Y2K-like problems that would show up due to a FISCAL YEAR rollover are ones that deal with FINANCIAL APPLICATIONS. (Makes sense, right?) These have nothing to do with embedded chips, power, water, sewage, etc. These have nothing to do with computer operating systems, hence no croaking. Workarounds, such as "fudging" a fiscal year to extend the date have already been publicized.

Bottom line: Fiscal year 2000 problems, or the lack of them, are completely independent of the problems that may -- or may not -- occur come January 1, 2000. Probably the biggest worry about FISCAL YEAR 2000 rollover problems was that if they DID occur and were publicized, this would wake a lot of people up to what is coming, and kill off any chance of preparing, because everyone would be trying to.

So the good news is that the world is still sleeping peacefully about Y2K. The ability to prepare still exists. Do it. Please.

-- King of Spain (, July 09, 1999.


Sure, FY2000 is only one of the tests.. Last year when I was reading all of this stuff Ed Yourdon, and several others wrote about Fiscal rollover problems. This was one of the two or three legs under the Y2K stool, if you will. This whole part of their argument has been proven totally wrong. 48 states did the rollover with thousands of systems, and I'll bet (from the lack of news) that there is a failure rate of less than 1%.

So now the argument by Y2K proponents has changed.. Now it's the embedded chips that are the WHOLE problem and that we can't know about that till it blows up in our face. Come on.


-- Bryce (, July 09, 1999.

Bryce,Michael Hayatt wrote the above article I posted.I am not a computer person,so I actualy rely on the y2k-failure-related- disruption-optomists(polyanna seems easier but,hey I'm from madison,and used to bending to the politicaly correct)to help me cut through the techno gibberish.Just because I'm predisposed to paranoia and I am quite pessimistic,I'd like to know if there's basis for my angst.What scares me and doesn't seem to get debunked are things like the reported progress and things like the Army saying they'll be ready by nov. with a full month for testing,the cia assessment and the nature and type of candor(or lack thereof) between department heads in your average corporation.fbi cancelling all banks,telcos,governments,or fourtune 1000 companies giving any reason for confidence.Unfortunatly,it is the intagibles that make y2k so terrifing.Kenny Decker's assessment of y2k as religion seems to to have a lot of basis.The type of person who sees technology as being able to fix any problem it creates(golden calf),then you'll see y2k as a hoax since it's inconcievable that GOD(technology)will ever abandon you.I see corporate america as a fucked up system that feeds on itself so I inherantly see it as poorly capable of dealing with this particular type of problem where every company really needs to share info in a manner contrary to competitive bussines.

-- zoobie (, July 10, 1999.

Hello zoobie,

I respect your viewpoint, many organizations are screwed up, especially the Govt. I guess I never expected them to actually take care of this in a competent way. In private industry, and as a person who works for a major vendor, I see things differently though. The Govt employees will still have jobs after their stuff dies all over the place. We vendors may not. We have taken this a lot more seriously than most people assume. As I've said, I'm sure there will be failures (probably a lot of them). But for us this is actually an opportunity. Think of it, Banks and the Govt have got off cheap by re-using old computer systems written in the 70s and 80s. If they break, they will almost certainly need to be replaced. Lots of work.

Also, I don't believe in the domino effect that is described by Yourdon, companies don't close if their billing system breaks. They will re-assign sales people, accounting people, programmers, and even useless marketing people ;-) to the task RIGHT NOW.

Private industry is EXTREMELY flexible. One day, (at a smaller company) one of our disk duplicating people was sick for a week. I took a crash course in how to use a 'hopper' to make disks. Senior developer to 'hopper' operator in about two hours.


-- Bryce (, July 10, 1999.

Well, here I go again, sticking my neck out and preparing to don my asbestos suit.

Bryce, I agree with a lot you say. Zoobie, I think perhaps the only significant thing in Hyatt's article was that Y2k is NOT hype. It is a VERY real problem, and always has been. I'm no more comfortable with the concept of fix-on-failure as anyone else on this forum, although I do believe it could work in SOME cases...mostly VERY small businesses.

On the other hand, ALL the more pessimistic folks have been using these 1999 dates to indicate milestones at which all would realize that their predictions were true. By ALL the more pessimistic folks, I refer to those who wrote books or made themselves prevalent on the internet as forces of some expertise. I'm quite sure failures have occurred. I'm quite sure that some are even still being worked out. Failures have CERTAINLY occurred in the embedded arena. That was the whole point of testing. SOME failures were Y2k-related and some were not.

In IT, Zoobie, as Bryce well knows, the entire goal is to make problems that occur transparent to the users. It's why we spend 36 hours cooped up in a little space waiting to verify results before an install, and another 24 hours baby-sitting that system once it goes into production. I may take this opportunity to answer George's question regarding how I know so many people so intimately that they would tell me their financial intentions. I've been in IT all my grown life, working on mainframes for the most. I've spent 10 years ...or is it 11 now? as a contractor, sometimes taking short-term contracts. Each production install throws us into an environment wherein while some people work on a problem others are waiting on results from something they've done. The night grows long without conversation, and as fatigue and comradery set in, folks talk to relieve the tension. There's no room for political correctness as jokes are told and EVERYONE laughs. Men lose sight that I'm female and discuss the most intimate details of their sex lives. They discuss problems with their families, their finances, their fears, etc. They see me simply as a competent programmer willing to work side-by-side with them to get the job done. We become a family in a way, and when the contract ends, I hear, "Kid...I'd be proud to work alongside you on ANY project." Don't ask me why they call me kid. I'm probably the same age they are. The next site produces the same results and the next site the same. We all continue to keep in touch to help each other locate contracts that need good programmers. We spend 2 minutes discussing future jobs and 50 minutes discussing personal matters, including how the current job is going. Each of us have met folks that moved on to other contracts and have stories to share about those projects also. The network of information grows and grows as we all move on to other projects at other sites. Many take contracts in other states when the local work dries up. The contacts made in those states may then move on to OTHER states, but the chain of contact is never dropped. In this way, we hear news from sites in all areas of the country and some in other countries. The news I've heard is basically positive. That news makes me more optimistic as to the outcome of Y2k overall than some who may have to rely more on the media or the experiences of IT folks with a more limited scope. In defense of Cory's experiences, I've worked on some contracts that had the incompetent friend of a friend. I also wasn't about to sit there working my butt off while some fool slept at his terminal. It doesn't take too many times to yell, "Jack, wake up!" before an entire office realizes that there's someone on a project that's pulling progress down. There's no room to play politics when a deadline is involved. I've never lost my job over this, and the sleeping Jacks were promptly removed.

Sorry I went so far astray.

I'll address one more thing before I address preparedness. This whole date of completion thing has me baffled. It only makes sense that look-ahead logic needs to be implemented first, as it's the first possible problem to occur. We do try to prioritize, right? However, what baffles me is to go apeshit when either self-imposed or government-imposed deadlines are not met. If the whole point is to get the project done by January 1, 2000, why the big fuss because some won't make it by June 30, 1999? The other thing that baffles me is why folks assume that if a deadline changes it means that the testing timeframe is necessarily shortened. We have absolutely NO idea how many firms were actually done with remediation at the end of 1998, nor do we know how many had 1 program left, 1000 programs left, etc. However, just because they move their dates out does NOT necessarily mean that X% of their systems haven't already been completely tested. They may very well have tested 90% of their systems completely already, but encountered problems with the last 10%. It shouldn't take much testing time to test the remaining 10% and then run a final test on the 100%. The other thing I would throw in here is the argument regarding mission-critical systems that are FED by non-mission-critical systems. I don't buy this argument at all. A mission-critical system by definition would include any system that feeds it.

While I'm creating War and Peace here, I'll include another thing. It was mentioned in a separate thread a while back, but I didn't see any feedback on it. Sites don't tend to clean up OLD programs. Initial estimates based on how many programs exist in a system may very well include as many extinct programs as active ones. I encountered this when I was asked to estimate the activity involved in remediating a subsystem. I was at first overwhelmed by what I found. Further investigation, however, found that HALF the programs still existed in the source and production libraries but were never used.

Now...about preparation. It's a common fallacy (as propogated by Mr. Hyatt in his article) that polyannas discourage preparation. I'd have to sit down with King of Spain and discuss HIS preparation (outside of the mud-wrestling videos) and determine who was more prepared, but my first guess is that I'm better equipped to handle immediate needs than he. I've said this before, but I'll say it again. Preparation plans should BEGIN with how your local area stands remediation-wise. I already know that if our electricity stays up, our water will be fine. I know this from being in constant contact with my water director for over a year. If the electricity doesn't stay up, I have water set aside. Kroger has been working on the problem for years, as has Albertson's. They've contacted their suppliers and transportation folks, yet I have plenty of supplies in case someone lied. My banks are ready, and to the dismay of many I still intend to keep my money there. I know many of my neighbors, and have discussed Y2k with some. SOME will prepare, and for those that don't, I have extras that I intend to freely offer if the need arises. I don't see the need for guns and the killer poster scares me. I understand his point, yet he obviously has a background of warfare that I lack.

Regarding these OTHER things, it may simply be that having 3 teenagers has encouraged me to save my BIG GUNS for the BIG things, but I can't grow concerned with Locusts, explosions, and all the other catastrophies that occur on a regular basis despite Y2k. I can't concern myself with Russians growing cold due to their government choosing to spend money on other things than remediation of their nuclear plants or grids. If there's nothing I can do about a problem, what sense is there in adding it to my worries? Recession? I've lived through a few. Unemployment? I experience it regularly. It's the reason why I save more than I spend. Contractors are the FIRST to go on a site, oftentimes with a 1-day warning.

Is it really TEOTWAWKI if nobody notices? It WILL be TEOTWAWKI in SOME areas of the world, and it's up to each individual to determine how Y2k will affect their lives. PHEW! End of Rant

-- Anita (, July 10, 1999.

Bryce; I guess you haven't heard of all the companies that are keying in dec. 1999 dates to give them more time to fix their systems?????? Do that, and everything continues to work fine.

Hi Zoob:)

-- FLAME AWAY (, July 11, 1999.

Moderation questions? read the FAQ