In a recent note, "Thompson, Steve (SCI TW)" said:
> Date: Tue, 26 Sep 2006 16:59:59 -0400
>
> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
> Behalf Of Paul Gilmartin
>
> Alas, UNIX doesn't do leap seconds; forbidden by POSIX; always uses
> UTC.
>
> If it doesn't do leap seconds, then it must be GMT, right?
> Isn't UTC = GMT + LEAP_Seconds?
>
> I ask this because a few months ago I made changes to most of the "MVS"
> based code in a product to have it use IBM services and NOT "STCK and
> then do arithmetic". We were finding that our time stamps were off by
> about 22 seconds (compared to what the system was doing with WTOs) where
> an ETR tied to GPS (they use UTC) was in use.
>
Actually, UTC = TAI - | LEAP_Seconds |
(at present, UTC < TAI; I don't know the formal sign convention.)
(TAI = International Atomic Time-- see recent "acronyms" thread.)
GMT is an older system which relied on ad-hoc daily corrections
before leap seconds were invented.
See:
Linkname: NIST: Time and Frequency Division - Division 847
URL: http://tf.nist.gov/
Linkname: Leap Seconds
URL: http://tycho.usno.navy.mil/leapsec.html
-- gil
--
StorageTek
INFORMATION made POWERFUL
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
