Time (was: Character set wars
Dan Jenkins
dan at rastech.com
Tue Jan 31 17:23:00 EST 2006
Ben Scott wrote:
> On 1/31/06, Fred <puissante at lrc.puissante.com> wrote:
>
> > As far as the Y2K fiasco, that was kind of a non-event.
>
> That depends on which "Y2K" you're talking about. I was referring to
> the work done finding and fixing actual problems, not the
> media-induced "the world is going to end" crap. :) There was a huge
> amount of code that needs to be examined and various problems fixed.
> That work happened over years and even decades. The biggest myth
> about Y2K is that the problems were going to happen on 1 Jan 2000.
> My personal Y2K experience was in 1998, when a program I happened to
> be supporting tripped over a bug in an expiration date calculation
> routine. What about railroads that schedule things out years in
> adance? They started hitting Y2K in the early 1990s. What about
> banks that manage 30 years mortages? Their Y2K issues started in the
> 1970s!
>
> Getting everything working for Y2K was a long, slow, process that
> involved finding and fixing a lot of obscure, irritating bugs over
> time, and replacing a lot of ancient code that couldn't be updated.
My first Y2K bug was in 1982. My last remediation was July 2005. The
latter was a bug which
corrupted accounting data every month's closing. (Source code was
available. Funding wasn't.
I just fixed the damaged data every month until they migrated to
something newer.) Most Y2K
bugs I encountered were annoyances. Only a couple, like the one above,
broke things badly.
The media hype was simply that. The problem was most definitely real and
required decades of
work to fix. I fully expect to fix several more as time goes on. While I
hope to be around in 2038,
I also hope I won't be still fixing those sort of problems by then. ;-)
--
Dan Jenkins (dan at rastech.com)
Rastech Inc., Bedford, NH, USA --- 1-603-206-9951
*** Technical Support Excellence for over a quarter century
(By 2038, this would be over a half century. Half a century of doing
tech support...enough said.)
More information about the gnhlug-discuss
mailing list