At 7:56 PM +0000 9/18/11, Poul-Henning Kamp wrote:
In message <p06240805ca9bf8bc07d3@[192.168.1.101]>, Joe Gwinn writes:
I doubt that the push to drop leap seconds has anything to do with
POSIX - there are far larger forces at play.
Yes, (Armed) forces which run a lot of POSIX systems...
We can discuss who to blame, but POSIX would have been a damn good
time to fix leap-seconds.
Not really. The problem with leap seconds is that they are too rare
to allow for comprehensive testing of systems, and so such systems
tend to fail when a leap second comes along. This is true regardless
of the chosen OS, and the mission software can also screw up.
The issue came up in an ATC-derived system I worked on. While the
system ran on Sun boxes, the C-coded application software (which has
ancestry reaching back to the days before UNIX and well before GPS)
does its own timekeeping. The worry was that the software would fail
if fed a leap second. So, a full-scale test was performed, where a
one-second time step was injected into the built-in simulation. A
positive leap second produced no visible disturbance to the IFF
tracks, while a negative leap caused a visible hiccup. Given that
this was not an ATC system, despite the origin of the code, the
hiccup was not a problem. What saved the software is that it is fed
only rotating-antenna data, with about 12 seconds between updates, so
plus/minus one second wasn't that big a fractional error.
It didn't get fixed in POSIX because UNIX commercial interests at
the time were scared shitless by the amount of breakage it would
cause.
Actually, it never got that far. See next.
And nobody has bothered to update UNIX/POSIX/OpenGroup texts since
then, rendering them increasingly archaic and out-dated for modern
needs, so leap-seconds are still not fixed.
Ahh, well, the committee did try, only to be torn apart by warring
time zealots. The whole sad story is in the archives.
The initial proposal to drop Leap Seconds came out of Pentagon
and was fast-tracked through the US Government.
We have never been told the where, who, how and why of that.
I bet it is known, even if we two don't know it. And the stated
reason would be interesting.
Joe Gwinn
_______________________________________________
LEAPSECS mailing list
[email protected]
http://six.pairlist.net/mailman/listinfo/leapsecs
