[ntp:questions] Google and leap seconds

Rob van der Putten rob at sput.nl
Thu Sep 22 10:02:31 UTC 2011

Hi there

John Hasler wrote:

> Well, we have machines on and around Mars that need to keep time and
> that we need to synchronize with.  In any case, the point is that
> planetary days and years are intervals to be measured, not clocks.  I
> understand and share the desire to localize the representation of time,
> but that should be done explicitly.  It made sense to synchronize our
> clocks with the Earth when we knew that it was a more stable oscillator
> than any we could build, but that is no longer true.  Leap seconds don't
> correct an error: they jigger stable clocks so that they track a known
> drifty one.  They should be treated as localization, not as error
> correction.  Put them in tzdata.


Wim T. Schippers for president

More information about the questions mailing list