On 18 Feb 2005 12:19:40 -0800, jim rozen wrote:
In article , Ed Huntress says...

Software managers who were commenting on the Y2K problem before the turn of
the century said, for the most part, they had to squeeze the fields down to
the smallest possible size to keep record sizes down and that they never
anticipated that the software would still be used a quarter-century later. I
don't know what the whole story is but it sounds very reasonable, having
worked with those guys every day back in the '70s.

My understanding from talking to the folks who did a lot of this
is that you are correct. They were just trying to save two digits
in the data, multiplied by a lot of data files.

Yes. I specifically remember being taught to use 9999 as a flag
in the date field as "done with job". The last entry in a table would
be 9999 which was the exit condition for whatever routine was operating
on it.

I guarantee that if anyone was still using several programs I wrote
when Y2K came around, they had problems on September 9th of 1999.

Remember, when you bought a PC then, it had no hard drive.
Those were aftermarket. 20 megs of hard drive space was
huge.

Heh...I didn't get a hard drive until about 6 years into it. No
floppies for the first 3 or so, it was all on magnetic tape. At
150 baud.