Open 375gnu opened 5 years ago
This link provides several useful links: http://mm.icann.org/pipermail/tz/2019-July/028377.html
Better to read the whole thread: http://mm.icann.org/pipermail/tz/2019-July/028270.html
I think it makes sense to have a custom rotation model for the Earth but I don't think it makes sense to use that to predict future leap seconds: it doesn't represent the actual definition of how UTC and TAI are linked.
We can predict leap seconds algorithmically. Old algorithm can be found in Meeus; Astronomical Algorithms. More correct one is described in https://eclipse.gsfc.nasa.gov/SEhelp/deltatpoly2004.html