[time-nuts] The future of UTC

[Sanjeev Gupta]

On Fri, Jul 15, 2011 at 23:09, Steve Rooke <sar10538 at gmail.com> wrote:

> On 16 July 2011 03:01, Poul-Henning Kamp <phk at phk.freebsd.dk> wrote:
> > In message <
> CACTjVNy8h2EtHR_M6DquxhABhJb9NfgyauhJcN1bf-UMH+KhDQ at mail.gmail.com>
> > , Steve Rooke writes:
> >
> >>Ah! I get you. Not 10 leap seconds at 20 year intervals, just an
> >>almanac to indicate when they will be for up to 20 years in advance. I
> >>guess that means they could take a bye for any scheduled event that is
> >>not required, as in the 7 year period without one.
> >
> > Nope, once they have scheduled a leap-second, it happens.
>
> And if it's not needed?
>

Then they are exiled from Gallifrey, and fed to the Daleks.

Seriously, if we are announcing 20 years in advance, we accept that DUT may
be as large as 4 or 5 secs.  In which case, having an extra one (or not
having one when required) will not materially change the _long-term_
tracking.  Within a few years, the effect should lessen.

Although I would rather that leap secs stay, and DUT is kept small, if we
are not changing the definition of UTC, but loosening the strictness of the
tracking in the "short"-term, this may be a good compromise.

PHK, in your proposal, the long term stability of "low, bounded DUT" would
be guaranteed?

-- 
Sanjeev Gupta
+65 98551208     http://www.linkedin.com/in/ghane