On Mon, 28 Oct 1996, Sherri Malouf wrote:
> So here is my wandering (and actually a little bit of ranting - please
> forgive...). Gary and I spoke briefly about the fact that many computers
> are not capable of going past the year 1999 and that we really don't have
> a handle on the true impact this will have. It reminded of the only
> article I have seen about the problem and it was on a consultant hired by
> the Federal government to deal with the problem. His comment was
> basically that too little was being done too late.
>
> So, I then thought -- well the government is probably focusing most of its
> efforts on security issues. Then I thought -- this is crazy! If
> computers had been invented in the 1500's I can understand the calendar
> thing. But most computers have hit the market in the last 20 years.
> While technology moves quickly didn't anyone think about this? We have
> complex relational data bases but we don't have the year 2000?
The problem isn't with the computers themselves, it's with the fact that
many organizations have enormous databases in which dates are stored with
only two digits to represent the century, and many of the programs that
use and update those databases are old ones (custom-written programs, not
off-the-shelf ones) that have been modified and patched over the years to
the point that the parts of the code that are dependent on a particular
date-storage format are scattered throughout the programs and changing
them isn't a trivial matter. Also, converting over the databases is going
to take a lot of time.
How did this happen? You have to remember that until a few years ago,
disk space was quite expensive, and that until around the mid-1970s,
computer time was more expensive than human time. Therefore, it made
economic sense at the time to use every trick in the book to reduce the
amount of storage space and computer time required to deal with those
databases. And as time went on, modifying the databases and programs to
deal with the fact that 2 digits was soon to be inadequate for
representing the century part of dates was seldom a high priority because
it would have required spending time and money that would show no
*immediate* return, and there were always more urgent things to do. It's
almost a classic example of suboptimization; what appears to be
cost-cutting really turns out to be cost-*shifting*--in this case,
shifting the costs into the future. It's the old case of management
preferring to fight fires rather than preventing them.
--Eric Bohlman <ebohlman@netcom.com>
Learning-org -- An Internet Dialog on Learning Organizations For info: <rkarash@karash.com> -or- <http://world.std.com/~lo/>