- We've got a problem. It may be the biggest problem that
the modern world has ever faced. At 12 midnight on January 1, 2000 (a Saturday
morning), most of the world's mainframe computers will either shut down
or begin spewing out bad data. Most of the world's desktop computers will
also start spewing out bad data. Tens of millions -- possibly hundreds
of millions -- of pre-programmed computer chips will begin to shut down
the systems they automatically control. This will create a nightmare for
every area of life, in every region of the industrialized world. It's called
the year 2000 problem. It's also called the millennium bug, y2k, and (misspelled),
the millenium time bomb. Millennium or millenium: it doesn't matter how
we spell it; this bomb isn't going away.
-
- Think of what happens if the following areas go down
and stay down for months or even years: banks, railroads, public utilities,
telephone lines, military communications, and financial markets. What about
Social Security and Medicare? If Social Security and Medicare go down,
it will affect millions of people. Yet both programs are at risk.
-
- Is this possible? It's far more than merely possible.
If you doubt me, read the report by one of America's best-known mainframe
computer programmers, Ed Yourdon, who has written two dozen books on programming.
I have posted his report under the category "Domino Effect,"
with the title, "A Programmer Describes the Dominoes: Great Depression."
This report spells out what can happen in clear, restrained, almost apologetic
language.
-
- I have also added a link to his new book, which he is
publishing chapter by chapter on the Web: "Time Bomb 2000." Access
it under the category, "Domino Effect." It's the document called,
"Senior American Programmer Writes Y2K Survival Book." You may
not believe my scenario. You had better take Yourdon's scenario very seriously.
Tens of thousands of programmers will. What he is telling them is simple:
"It may soon be time to quit your big city jobs and head for safer
places." If they do, there will be no solution for y2k. Will they
quit? I'm betting my life on it. The exodus of programmers will begin no
later than 1999.
-
- Months before January 1, 2000, the world's stock markets
will have crashed. Who is going to leave his money in his bank if he thinks
his bank's computer is not reliable? A worldwide run on the banks will
create havoc in the investment markets. People who have placed their retirement
hopes in stocks and mutual funds will see their dreams vanish. How reliable
will stocks and mutual funds be if the banking system has closed down?
How will you even get paid? How will your employer get paid? How will governments
get paid?
-
- By the way, no government tax collection agency above
the county level is Year 2000-compliant today. People will know in 2000
that the government cannot trace them. Will they continue to pay, especially
if the huge government welfare programs for the elderly have shut down?
-
- But if governments don't get paid, what happens to government
debt markets? How high will interest rates go in 1999 if investors think
that governments will default (or print mountains of paper money) in 2000?
What will high rates do to the world's economy?
-
- If someone tells you, "Our organization is Year
2000-compliant today," ask him to put this statement on letterhead
stationery and mail it to you. You will find out how confident he is regarding
his assertion. Until you have it in writing on letterhead stationery and
personally signed, don't take the claim seriously. Regard it as just another
happy-face y2k rumor. There are lots of them.
-
- Everything is tied together by computers. If the computers
go down or can no longer be trusted, everything falls apart. And it matters
not a whit to the computers whether we accept this fact or not. They do
what they've been programmed to do. They've been programmed to recognize
2000 as 1900. (Uncorrected PC architecture DOS and Windows-based desktop
computers will revert back either to 1980 or 1984. They can be corrected
briefly, but as soon as a PC is turned off, the correction dies. It will
reboot to 1980 or 1984. Meanwhile, PC programs must be redesigned.)
-
- Our first response when we hear this news is denial.
Most people will stay in denial, including the business managers whose
companies are totally vulnerable to a computer failure. This is why the
problem will not be fixed. Everyone in authority will deny that time has
run out to get this fixed, right up until December 31, 1999. They are paid
to deny this. I'm saying that it's over. Right now. It cannot be fixed.
Whatever it does, the Millennium Bug will bite us. How hard? There the
debate begins.
-
- I don't expect you to believe me . . . yet. That is why
I have created this site. On this site you will find links to other Web
sites that have posted documents related to the Year 2000 Problem. Included
are such things as military sites, government hearings, news releases,
and much more. I also include comments with each document, so that you
can understand why I think it's important.
-
- The goal of this site is not to bury you in information.
Rather, it is to give you a sense of the magnitude of the problem. The
domino effect of a computer-driven breakdown in supply delivery systems,
including the means of payment (banks), will be huge. This site will help
you to evaluate your own personal vulnerability.
-
- I have many critics who believe that my scenario is too
apocalyptic. You must decide for yourself. This Web site is designed to
provide you with relevant evidence to help you make an informed opinion,
and then a principled series of decisions.
-
- If you have practical questions -- where to go, what
to buy, etc. -- ask them on one or more of the discussion forums. That
is why I have created them.
-
- When you hear good news about some organization that
is y2k-compliant, recall Ronald Reagan's statement with respect to disarmament
treaties: "Trust, but verify." Get a signed letter on letterhead
stationery that the organization is 100% compliant. Until you receive this
form of written assurance, which the outfit's lawyers have cleared, assume
the worst. Don't take seriously any promise that the outfit will be compliant
RSN: Real Soon Now. If you are told that the organization will be compliant
in December, 1998, and ready for testing in January, 1999, you have a form
letter in your hands. Just about every firm promises this, since they admit
that they need at least six months for testing.
-
- The answer to this standard form letter is a letter back:
"Have you signed an agreement to lease mainframe computer time for
testing your software, beginning on January 2, 1999? If so, with what leasing
company?" If every outfit that promises to be ready for testing by
January 2, 1999, meets its deadline -- they all won't -- then there will
be no excess mainframe capacity to run the mandatory tests. On the other
hand, if companies can still buy 1999 rental time today -- and they can
(rented by non-compliant companies) -- then ignore all assurances about
a December, 1998, deadline. The only valid proof of the seriousness of
the assurance of a late 1998 deadline is the outfit's signed contract that
leases at least six months of mainframe computer time in early 1999.
-
- Note: If my critics want to create their own Web sites
filled with "it's not going to be all that bad" evidence, they
may do so. I am unaware of any such site on the Web today. Let me know
if you find one.
-
- I am also unaware of any y2k programmer who says, "Even
if programmers don't get this fixed, there will not be big problems."
The debate is over two questions: (1) "Can the programmers get this
fixed in time?" and (2) "How big will our problems be if they
don't?" My answers: "no" and "catastrophic." You'll
have to decide for yourself, either now or later.
-
- One last warning: the governments' strategy, all over
the world, is: (1) talk this problem to death, (2) form committees, and
(3) send out PR sheets that they will make it -- without evidence. But
this problem cannot be talked to death or solved by committees. It cannot
be avoided. There is an absolutely fixed deadline. Bureaucrats are not
used to absolutely fixed deadlines. Neither are computer programmers.
-
- I'm not a programmer. My Ph.D. is in history. I take
the historian's view: things are interconnected in ways we can barely understand.
If you want to know what I think lies ahead, get copies of the three books
that I mention in my free e-mail report, "Blind Man's Bluff in the
Year 2000."
-
- "Blind Man's Bluff in the Year 2000" is ideal
for introducing the problem to wives, in-laws, and other skeptics. You
can receive a copy in a few minutes.
- Why It Will Be So Hard To
Correct
By Allen Comstock
10-25-97
I am not a programmer and so can not comment
first hand about the coding complexity. However, I am told that date fields
occur fairly frequently in many programs of most businesses and apparently
there is no way to easily and comprehensively identify just the date code
or to determine where else in the program there is other code which references
each instance and sometimes multiple instances. Further, all the different
programs an organization uses have to interact and all the stored data
has to be compatible with any changes made to the program code.
-
- The largest installations can have hundreds
of millions of lines of code. Then you have to consider that main frames
are often networked and share data and all the networked machines have
to be fully compatible and any non-compatible data transfers will corrupt
otherwise clean systems that access it. What seems like an easy fix becomes
incredibly complex.
-
- Virtually the entire computer infrastructure
of the entire world, all the computer installations of the last forty plus
years, needs to be intensely scruitinized and a large portion of the software
needs to be rewritten or replaced. Further, it's very difficult to do anything
with embedded circuitry--dedicated chips in all kinds of devices, many
of which can be date sensitive without external indicators.
-
- Many organizations are just getting started
on Y2K projects that require more time and manpower than is available.
Reportedly many others haven't even begun. Computers have no tolerance
for mistakes or wrong decisions--perfection is the primary criteria. And
remember, there is on-going complicating pressure and demands from increasingly
intense reality testing all building toward an unavoidable simultaneous
real world test of virtually all the computers in the world, time zone
by time zone, all on the same day.
|
Homepage