[ntp:hackers] Any comments on the 'secret proposal' to ITU from several US Agencies to do away with the Leap Seconds

M. Warner Losh imp at bsdimp.com
Wed Sep 10 17:35:30 UTC 2008


In message: <007f01c91367$b4365a70$6401a8c0 at tsg1>
            "TS Glassey" <tglassey at earthlink.net> writes:
: Any comments from this group on the need to do away with the addition of 
: leap seconds in UTC?

I would imagine that ntp would be the right way to distribute an
approximate DUT1, since that information would likely need to be
published if the approximation floats bigger than +/-0.9s.  I would
imagine we'd also need to for a Stratum 1 server to say "My time
device doesn't know, so I'm not publishing that."  I'd also imagine
that this could be done with extensions of some kind, as defined in
the protocol, but I'm not entirely sure.  I'm also not entirely sure
what the right model would be to propagate quality metrics about the
DUT1 estimate to downstream servers, since I'm not sure how quickly
this information changes, or if different sources would have different
levels of precision in their answer.

Warner



More information about the hackers mailing list