Magnus Danielson magnus at rubidium.dyndns.org
Thu Dec 23 13:28:49 EST 2010

On 12/23/2010 07:10 PM, Finkleman, Dave wrote:

> Do any sources of precise, accrued time have a leap second warning bit

> as DCF 77 does? Is the philosophy of leap second warning in DCF 77 a

> good paradigm for helping implement the leap second broadly?

It helps, somewhat, but it would require that systems would listen to
leap second distribution capable systems.

Just using a single bit is not as useful as:

1) Indicating the current TAI-UTC offset
2) Indicating the time of the upcoming leap second (and direction, even
if it is always in one direction in forseeable future)

It does not address security issues (i.e. can I *trust* this source).
It also does not address the issue that the system as such needs support
for leap second handling.


More information about the LEAPSECS mailing list