[LEAPSECS] content of the NTP-aimed leap-seconds.list files

[Rob Seaman]

On Feb 15, 2015, at 10:32 AM, Zefram <zefram at fysh.org> wrote:

> OK, have a play with the attached.  Comments welcome.  Second implementation especially welcome.

There was a missing perl JD module from the host I first tried.  Will try again later from a different machine.

Man page looks solid.  Format appears well designed.  Did not vet for technicalities with the Gregorian calendar and such.  Sidesteps the issue of the meaning of a calendar day since the difference between TAI and UTC "remains the same over the course of a calendar day of UTC”.  This only remains true so long as leap seconds continue to exist.  (But that’s an issue with the underlying standards, not the file format.)

Does not address DUT1.

Would require an externally authenticated secure channel.

No objection to naming this after Msgr. Lemaître, but he had little direct connection with the issue, having died 6 years before UTC year 0 ;-)

Remains an issue how best to integrate this with other activities such as the tzdist project.

> The binary format is concise enough that it could be usefully applied to encode the entire leap schedule in a DNS TXT record.

Perhaps others can comment on DNS TXT versus A versus AAAA versus some new proposed DNS record type?  It seems to me that there are several use cases to address and more than one delivery mechanism might be required to meet them all.

Rob