On Tue, 13 Mar 2012 16:12:07 -0500, John Gilmore wrote:
>
>The sequence
>
> 2012 June 30, 23h 59m 59s
> 2012 June 30, 23h 59m 60s
> 2012 July 1, 0h 0m 0s
>
>will certainly appear in the transmitted sequence, but its middling
>term was chosen to call attention to itself precisely because it is
>NOT a valid date-time value. It appearance does not legitimate the
>notion that 60 is a valid term in a zero-origin cycle modulo 60. The
>mathematics of these things has been settled since Gauss.
>
Your second sentence is quite true. But what is your authority for
asserting that the seconds (as they may have been respecified by
UTC) must be a zero-origin cycle modulo 60? The excerpt above
from IERS Bulletin C:
http://hpiers.obspm.fr/iers/bul/bulc/bulletinc.dat
... and the Wikipedia article cited by the author of WP102081:
http://en.wikipedia.org/wiki/Leap_second
... both appear to imply otherwise.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
