[LEAPSECS] Possible outcomes ?

Warner Losh imp at bsdimp.com
Tue Aug 13 22:25:05 EDT 2013

On Aug 13, 2013, at 5:59 PM, Rob Seaman wrote:

> On Aug 13, 2013, at 3:05 PM, Poul-Henning Kamp <phk at phk.freebsd.dk> wrote:


>> There are only two realistically possible outcomes:


> Interesting that you fail to mention other outcomes such as the consensus from the 2003 Torino meeting of calling a leap-less timescale something other than UTC. Among other things this would leave UTC for backwards compatibility.


>> Leap seconds go, and nothing much happens, apart from a

>> handful of astronomers needing dental repairs,


> This assertion is and always has been completely unsupported, and remains broadly untested. Where it has been tested it has been found false:


> http://www.cacr.caltech.edu/futureofutc/2011/preprints/36_AAS_11-677_Seaman.pdf


> Note also that the proceedings of the 2011 Exton meeting (http://www.univelt.com/book=3042), and soon the 2013 Charlottesville meeting, were published by the American Astronautical Society, not by the astronomy community.

You take Poul too literally. By "astronomer" he means anything that points a telescope, or other such-like device. Astronautical seems to fit that bill...

> And for another thing, leap seconds can't just "go". The long-term quadratic trend that was described in the 1999 McCarthy and Klepczynski article (http://gauss.gge.unb.ca/papers.pdf/gpsworld.november99.pdf) guarantees that we'll have to deal with the embargoed leap seconds at some point.

One could punt to the politicians, and then it becomes a once a century change to the timezones, which is in the noise as far as time zone changes go...

>> or leap second stays until they create

>> enough havoc to be removed anyway (max 2 centuries)


> It is significant that even you are saying we have 200 years to sort this out. Simple prudence suggests testing assertions from the several parties *before* making changes.

While that sounds good in theory, leap seconds were rushed into production originally. We've already debated how to fix leap seconds for much longer than the original debate to establish them...


More information about the LEAPSECS mailing list