On Fri, Jan 20, 2012 at 10:13 AM, Tony Finch <dot@xxxxxxxx> wrote: > Nick Hilliard <nick@xxxxxxx> wrote: >> >> Your arguments in favour of abolishing leap seconds are all good. But can >> you please do us all a favour and provide a similarly lucid list of reasons >> that an apologist would use to say that leap seconds should be kept. > > I agree that leap seconds are a horrible bodge that would be best got rid > of. The reasons to keep them mainly revolve around systems that require > UT1, i.e. the angle of rotation of the Earth, and the protocols that > disseminate UT1. > > These include: > Astronomical systems > Space and satellite ground stations > Time broadcast standards > All of these are computer controlled now-a-days, and have been for some time. Even in the 1980's, the computer systems used UT1, not UTC. > Many protocols and software implementations rely on the guarantee that > DUT1 (the difference between UTC and UT1) is less than 0.9s. For instance > many time broadcast formats don't have space for larger values of DUT1. > > Many instruments that point at the sky rely on the fact that DUT1 < > 0.9s for establishing an initial rough aim. Not so much the professional ones. > > For more examples see http://futureofutc.org/preprints Look at Daniel Gambis' survey Preprint-668 (~ 75% in favor of not changing anything). > > At a more philosophical level a lot of people find it difficult to accept > the idea of decoupling time from Earth rotation, to the extent that they > say it is obvious nonsense or foolishness, even though the rate error is > only one second every year or two. However UT1 is slowing down > quadratically, so the time scales will diverge increasingly rapidly. Yes. Note that for the same reason leap seconds will, in a century or so, start to come very frequently. > > We can paper over this difference by adjusting time zones every few > hundred years. In fact time zone adjustments will work further into the > future than leap seconds. But speculating about what will happen that far > into the future is foolish. > > Numbers for divergence between UTC and TAI: > http://www.ucolick.org/~sla/leapsecs/dutc.html#dutctable My actual proposal, if I were to make one, would be to keep UTC, but to make TAI "Internet time" and try and move most electronic things to TAI, keeping UTC only for civil time. Note that time is tricky and slippery and _any_ change, even this one, will have some painful consequences, for somebody, and past experience is that some of those won't be found until the change is made. Regards Marshall > > Tony. > -- > f.anthony.n.finch <dot@xxxxxxxx> http://dotat.at/ > Sole, Lundy, Fastnet, Irish Sea, Shannon: West 5 to 7, occasionally gale 8 > later in Irish Sea. Moderate or rough, occasionally very rough except in Lundy > and Irish Sea. Occasional rain or drizzle. Good, occasionally poor. > _______________________________________________ > Ietf mailing list > Ietf@xxxxxxxx > https://www.ietf.org/mailman/listinfo/ietf _______________________________________________ Ietf mailing list Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf