[LEAPSECS] FW: comments on DRR TF.460-6
phk at phk.freebsd.dk
Tue Sep 28 08:32:48 EDT 2010
In message <91772DEC8A29C048A9BFBE32C49CEB72E86FD4 at echo.usno.navy.mil>, "Matsak
is, Demetrios" writes:
>I asked an unquestioned expert on celestial navigation about its UT
>requirements. This was his response:
>If you want
>accurate longitude, you can't get around the need to know UT1 at no
>worse than the several-second level.
That's kind of an interesting answer.
So four seconds is the requirement, we add a factor 2 safetymargin and
get a DUT window of 2 seconds.
At the current rate of leap seconds, that means that a DUT estimate is
good for more than five years.
That allows us to merge DUT dissemination into the same temporal/pratical
domain as the whereabouts of the magnetic northpole.
In other words: After we get rid of leap seconds, you will get the
DUT and magnetic deviation from a small note in the corner of you
your charts, be they electronic or antique.
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
More information about the LEAPSECS