DST Productivity Drain
I am one of those contrarian freaks who despises Daylight Saving Time, regarding it as a stupid fix to a problem that does not exist. You want more daylight after work? Set your alarm clock ahead and get to work an hour earlier! Don’t go changing my schedule . . .
Most people . . . normal people, don’t feel strongly about DST. “Yeah, it is annoying to change all my clocks twice a year, but then I don’t mind the extra daylight in the evening.” Of course, my last job I was working 7AM-3PM anyway, so I was saving Daylight . . .
So, enough bitching. Let’s talk turkey. Let’s talk Operational Qualification for third party auditing of a production database system. When I run the regression tests for PostgreSQL-7.4.8, the “horology” regression test fails. The README for this version reports that this may happen if you run on the day of DST switchover . . . or if your Operating System naively applies current DST rules retroactively. But, in my case? Huh, just don’t get it:
*** ./expected/horology.out Wed Sep 24 23:58:06 2003 --- ./results/horology.out Mon Jan 23 15:12:40 2006 *************** *** 1755,1765 **** | Tue Dec 31 17:32:01 1996 PST | @ 34 years | Tue Dec 31 17:32:01 2030 PST | Fri Dec 31 17:32:01 1999 PST | @ 34 years | Sat Dec 31 17:32:01 2033 PST | Sat Jan 01 17:32:01 2000 PST | @ 34 years | Sun Jan 01 17:32:01 2034 PST | Wed Mar 15 02:14:05 2000 PST | @ 34 years | Wed Mar 15 02:14:05 2034 PST | Wed Mar 15 03:14:04 2000 PST | @ 34 years | Wed Mar 15 03:14:04 2034 PST | Wed Mar 15 08:14:01 2000 PST | @ 34 years | Wed Mar 15 08:14:01 2034 PST | Wed Mar 15 12:14:03 2000 PST | @ 34 years | Wed Mar 15 12:14:03 2034 PST | Wed Mar 15 13:14:02 2000 PST | @ 34 years | Wed Mar 15 13:14:02 2034 PST | Sun Dec 31 17:32:01 2000 PST | @ 34 years | Sun Dec 31 17:32:01 2034 PST | Mon Jan 01 17:32:01 2001 PST | @ 34 years | Mon Jan 01 17:32:01 2035 PST | Sat Sep 22 18:19:20 2001 PDT | @ 34 years | Sat Sep 22 18:19:20 2035 PDT --- 1755,1765 ---- | Tue Dec 31 17:32:01 1996 PST | @ 34 years | Tue Dec 31 17:32:01 2030 PST | Fri Dec 31 17:32:01 1999 PST | @ 34 years | Sat Dec 31 17:32:01 2033 PST | Sat Jan 01 17:32:01 2000 PST | @ 34 years | Sun Jan 01 17:32:01 2034 PST | Wed Mar 15 02:14:05 2000 PST | @ 34 years | Wed Mar 15 02:14:05 2034 PDT | Wed Mar 15 03:14:04 2000 PST | @ 34 years | Wed Mar 15 03:14:04 2034 PDT | Wed Mar 15 08:14:01 2000 PST | @ 34 years | Wed Mar 15 08:14:01 2034 PDT | Wed Mar 15 12:14:03 2000 PST | @ 34 years | Wed Mar 15 12:14:03 2034 PDT | Wed Mar 15 13:14:02 2000 PST | @ 34 years | Wed Mar 15 13:14:02 2034 PDT | Sun Dec 31 17:32:01 2000 PST | @ 34 years | Sun Dec 31 17:32:01 2034 PST | Mon Jan 01 17:32:01 2001 PST | @ 34 years | Mon Jan 01 17:32:01 2035 PST | Sat Sep 22 18:19:20 2001 PDT | @ 34 years | Sat Sep 22 18:19:20 2035 PDT ======================================================================
Figure it out? The horology test is failing because this version of the software predates a recent act of Congress, which has redefined what time it will be in Berkeley, CA on Wednesday, Mar 15, 2034.
But, think of the energy savings! Think of the extra golfing! Think of . . . it is all a farce! We’re burning untold hours of technical development each year to reprogram the computers each time Congress gets it in their minds that fiddling with the clocks is a good idea.
No more! It’s your time, dammit! Live your life by your own rules! And stop imposing this silliness on me! I’ve got work to do, and redefining time-keeping every year is blowing my productivity.
Grrr!
