[ntp:questions] A Suggestion For Abolishing the Leap Second

Quadibloc jsavard at ecn.ab.ca
Sat Jun 2 23:34:08 UTC 2007

jlevine wrote:
>    The proposal to incorporate leap seconds by changing the length of
> the
> second was tried before in the early days of atomic time. It was not a
> very
> workable idea then and it would be even more difficult to implement
> now.

Perhaps this is not what you are referring to, but I finally have
remedied my ignorance concerning the time standard in use prior to

In the period from 1961 to 1972, UTC used a second that was lengthened
from the SI second by 15 nanoseconds, and was further kept within 0.1
second of UT2 (mean solar time exclusive of the seasonal variations)
by means of occasional seconds that were lengthened, usually by 100

While this met people's needs for a while, it had its limitations, and
something more modern was required, leading to the present UTC.

I'm proposing a scheme which is somewhat different in principle, which
could be closely tied to something similar to the current UTC,
although, as you've noted, the need for discrepancies greater than 0.9
seconds between it and UT1 causes problems as well. There are never
any 100ms steps or 1s steps, only frequency offsets that take the
place of 1s steps. So this modified time scale could exist as an
alternate service, just as SAT and UTC co-existed for a brief period.
As an alternative service, it wouldn't *depend* on UTC being modified;
it could, I suppose, have a greater lag in keeping up with changes in
mean solar time than UTC.

So UTC could be left alone, but a system similar to my suggestion
could serve as "Internet Time" for the benefit of all those who can't
deal with leap seconds, and yet who need accurate timekeeping.

John Savard

More information about the questions mailing list