Tab Julius wrote:
>
> No, it's not so. I don't know who started it.
>
> What we USED to do was, typically in databases, use 99/99/99 as an
> end-of-records marker.
Hmmm... unfortunately 99/99/99 isn't a valid date, so many
programs will not accept it. Especially those programs from the
"old times" when using *SIX* rows of core to hold a date
would have been grounds for dismissal.
Specifically, the 9/9/99 isn't just a shutdown code, it's basically
an exception code that was oft-abused to hold whatever exception
was likely to the application.
It stems from having only a fixed field that only holds
a date, but having the need to store a code in there that
means something other than a date, such as "I don't know" or
"no date" and so on.
A typical example might be account expiries. My account might
expire on 1/1/81, and yours on 4/1/83. But the system administrator's
account should never expire, right? So let's put 9999 in there
so it never expires. Guess what, 9/9/99 actually happens if
the system is left on long enough. ;>
Moss Motors fell victim to a similar problem with their
last years catalogs. They seemed to be using a code price
for no longer available items, I think it was $123.01 or
at least came out that way.
So when you called asking for a NOS piston set for
an MGA twincam, or a NOS TD radiator, their computer would
say "not available".
Except, they printed $123.01 in the price catalog. Imagine
how many calls they got looking for long out of stock items
that were suddenly available for a mere $123.01?
--
Trevor Boicey, P. Eng.
Ottawa, Canada, tboicey@brit.ca
ICQ #17432933 http://www.brit.ca/~tboicey/
http://www.bargainbritish.com
|