On 2024-12-31 13:17, Doug Ewell via tz wrote:
Arthur Olson wrote:
With a New Year looming, an NPR Y2K review.
https://www.npr.org/2024/12/28/nx-s1-5116271/y2k-year-2000-preparations
The myth that the Y2K problem was overblown or a fantasy, or somehow just
“resolved itself,” has only gained traction in the past quarter century. It
really did take significant developer effort to prevent major problems in a
wide variety of systems. CNN had an excellent article about this some years
ago, but the lesson is quickly being forgotten.
The only aspect of Y2K that was overblown (other than scare stories, like
airplanes falling out of the sky) was the surge in Y2K certifications and
liability statements, even for software that had nothing to do with dates—a
novel opportunity for attorneys to get rich.
Brian Inglis wrote:
Could we please all make a New Year's resolution to stop people from
reviving two digit year formats that are starting to reappear, making
it impossible to figure out some intended dates
Yes. This.
Unfortunately, because humans can usually figure out most dates with two-digit
years, and because many of those humans no longer believe Y2K was a real
problem, persuading the public to give up two-digit years in computer usage
will be a challenge.
Here in Canada, CA(UK)/FR/US/MX(ES) date usage appears to be random, depending
on wherever they picked up their software, whether it *can* be localized, and
whether it *was* localized.
[In the best BOFH tradition, I customize %c/%x/%X and remove the %P/%p strings
as well as those above from my en-CA localizations, to check software.]
Many US multinationals, their application programmers, and software providers
appear to be clueless about such matters, and the expectations of the majority
of the world, the other ~7.7B.
[Dealing with such considerations often appears to be given to some subsidiary
in CA/UK/Nordic where English is common.]
Since the actual Y2K software problem, being related to two-digit years, was
really about the change in century and not the millennium, we (or our progeny)
are well on our way to repeating it 75 years from now.
Or possibly in 13 years and 18 days, or sooner if they deal with multi-year
durations, if they have not updated functions, libraries, or applications to
handle later times or longer periods.
*They* can always save some space and time by doing it better and faster
*themselves*!
Look around for GPS receiver issues - week rollover bug reports - about every
256 weeks, or firmware updates just prior to those dates, to delay the next
failure - about 4 years 4 weeks from now in GPS week 2560 cycle 2 week%1K 512
DoY 28 MJD 62164 ;^>
--
Take care. Thanks, Brian Inglis Calgary, Alberta, Canada
La perfection est atteinte Perfection is achieved
non pas lorsqu'il n'y a plus rien à ajouter not when there is no more to add
mais lorsqu'il n'y a plus rien à retirer but when there is no more to cut
-- Antoine de Saint-Exupéry
