Actually isn't 999999 going to be 1999-99-99, both an invalid day and
month?  I know that the 9999 problem was fairly common in COBOL shops that
decided to try to use 16 bit integers for dates and didn't at least use the
calendar intrinsic to generate it.





Stan Sieler <[log in to unmask]> on 09/08/99 06:14:32 PM

Please respond to Stan Sieler <[log in to unmask]>

To:   [log in to unmask]
cc:    (bcc: Shawn Gordon/IS/FHM/FHS)
Subject:  Re: Another Y2K date tomorrow




Larry asks:
>
> Okay Ken, call me dumb, but why would April 9 be a problem for Y2k?
Other than


With a 6-digit date (MMDDYY or YYMMDD), there are known instances where the
programmers chose "999999" as "never" or "expired" or something special...
forgetting that it's a valid date (1999-09-09).

For example, if you have an IBM labelled tape that you didn't want to
expire, you may have put "999999" as the expiration date.  (Yeah, they
could have used 991231, but that would imply immediate recognition of
the Y2K problem :)

--
Stan Sieler                                          [log in to unmask]
                                         http://www.allegro.com/sieler/