[ILUG] Linux's 2038 Bug

[Brian Foster]

 *sigh*

  this is a very well known problem.

  IIRC, the rationale to the original 1988 POSIX.1 standard
 mentioned this, and postulated that the “solution” was the
 world would be 64+ bits by then (or that at least «time_t»
 would be ... I'll try to remember to check my hardcopy next
 week (it's at work)).   at any rate, I know _I_ had adb(1)
 tell me the End of Time for me in 1987 or earlier; and I
 cannot have been the first.  check Risks?

  whether or not that postulate is “correct” is questionable,
 as it presumes  (i) most of the world's *ix-based software is
 updated before then (with a fix, 64-bit «time_t» or other);
 and  (ii) it's a serious problem.   as with the 2Kyr and DST
 defects (assuming USAian Federal Summer Time rules for c.1980
 apply to the entire World forever — AFAICR, this one caused a
 lump of molten steel to pour onto a factory floor in Germany
 a few years ago!) and so on, it doesn't _always_ matter ....

  FUD?  yes.  a problem?  yes.  serious?  maybe (but can
 be very annoying!).  unknown?  NO!

  a useful reminder?  perhaps, especially since this shows
 “well known” ≠ “universially known”.

  plans?  I only known of the vague “64-bit” thinking,
 but have never seen any migration strategy/planning
 et.al. written down ....  (but I'e also not searched
 in years!) .... ?

cheers!
	-blf-
-- 
Experienced (20+ yrs) kernel/software Eng: | Brian Foster   Montpellier,
 • Unix, embedded, &tc;  • Linux;  • doc;  | blf at utvinternet.ie   FRANCE
 • IDL, automated testing, process, &tc.   |  Stop E$$o (ExxonMobile)!
Résumé (CV) http://www.blf.utvinternet.ie  |     http://www.stopesso.com