On Feb 15, 2016, at 12:49 PM, Steve Allen <sla@ucolick.org> wrote:
On Mon 2016-02-15T12:14:52 -0500, Tim Parenti hath writ:
I prepared the attached patch which uses the file from IERS shortly after the 2016-01-11 announcement, but was sitting on it while we waited for NIST, since the conversation at the time seemed to indicate that NIST was our preferred source. I think, though, that since IERS actually decides the leap seconds and seems to be updating their file more promptly, it might be worthwhile to consider switching to IERS as our canonical source for this file.
That makes sense. Announcing the leap seconds is part of the charter of the IERS, and staff there are on public record as having interest in disseminating the information. It is not part of the charter of NIST, and staff there are on public record as not wanting leap seconds to exist.
I'll second that. And even if it were part of the NIST charter, NIST would only be a secondary source, not the primary source. IERS is the originator of the data; why use a middleman? If we currently have anything saying that we use any source other than IERS as the supplier of leap second data, we should fix that. paul