Then you are no doubt aware that this problem is very heavily overstated
because very little of it IS still in use today. And most of that which is
in use will not crash at 23:59:59 12/31/99, but at or after 23:59:59
02/28/2000.
The Operating Systems (OS's) in computers, especially personal computers,
do not differentiate between 1900 and 2000. Where a two digit system is
used to define the year, one of two methods is used. The year of first
design (1980 in personal computers) is stored either as a Zero or as to
value 80. In either case, that value is incremented each year. In systems
where that value started as a zero and is stored in a four bit byte, it
will roll back over to zero again once it passes the maximum number that
can be defined in four bits; 23 or at the end of 2003 if 1980 was zero.
Some older OS's had a built in prohibition to entering the year 00 to avoid
confusion. Those will crash.
Old 8 bit Atari's are good well into the 21st century. PC's running MS
DOS 4.0 and earlier may have serious problems. DOS 5.0 can stores the year
in a 7 bit byte and is also good into the next century. The BIOS chips on
those older machines will reset the date to 1980 if a date later than 31
Dec 99 has been entered into the OS. So, you reset the date each day when
you turn on the machine. DOS 6.0 and later accepts dates in two digit
format through 99 and then they must be entered in four digit format (2000,
2001, etc.) Again, an old BIOS chip can do you in unless you remember to
reset the date each time the computer is booted.
Newer systems have Flash ROM BIOS that can be updated by downloading a
new file. Newer systems yet, are fully Y2K compliant.
There are upgrade files available on the Internet for MS Windows 3.1,
Windows For Working Groups (WFWG) and for the initial release of WIN95.
WIN95B and WIN98 are Y2K compliant.
Most mini computer programs that use dates take the date from the OS. A
few derive their own date internally. Those need updating.
Older main frames and those pesky, specialized, imbedded process control
chips (two to five percent of them) will have problems.
But not all will fail. Not even most. A small percentage of systems
will fail. But my point is the problem has been over stated and has caused
way too much worry. A lot of this over stating of the problem is the
result of greedy merchants attempting to sell products and software that
for the most part is not needed.
Instead of panicking, each of us should research the problem for
ourselves and take whatever action is needed to fix our own systems. Yes,
talk it up. But do not spread panic. Yes, prepare, but do not go overboard.
My own computers have been upgraded, or in the case of one new machine,
came Y2K compliant (yes it has been tested).
I am prepared for disasters. I keep three months supply of canned goods
on hand and at least 60 days worth of dry provisions. We have a freezer
that is kept well stocked. I have a small generator (5KW) that I intend to
replace with a larger unit; not because of Y2K, but because we want to be
able to run more electrical equipment in the absence of commercial power.
And we have 14,000 gallons of water in a storage tank that is drawn from
all the time and continuously replenished. A minimum three month supply of
LP gas is on hand always. Right now I keep 48 hours fuel for the
generator. That will increase to 30 days when the new generator is installed.
So we have food, water, and power. We do this to be prepared for
hurricanes and other possible natural (and maybe manmade) disasters. And
we have had plenty of experience surviving extended periods without power,
water, and phones. I lie to think we would do this in other locations. We
did it when we lived in Southern California and when we lived in Florida.
It sort of comes natural after one has survived one disaster.
So the Y2K bug is not out primary reason for being prepared. Being
prepared for interruptions in services is something every one should strive
for just in case of hurricane, tornado, flood, earthquake, civil unrest,
war, and Y2K.
--Dan in Sunny Puerto Rico--
dan.worley@mindless.com
To Unsubscribe: Email majordomo@ces.ncsu.edu with the command
"unsubscribe sanet-mg".
To Subscribe to Digest: Email majordomo@ces.ncsu.edu with the command
"subscribe sanet-mg-digest".
All messages to sanet-mg are archived at:
http://www.sare.org/htdocs/hypermail