If you really want to cover all time stamps, then you must also cover the sporadic insertions of leap seconds just before midnight 1Jan or 1July. They are documented in the princ of op. The last one was 2012 July 01 and IERS has yesterday reported that there will not be any leap second insertion at 2014Dec31/2015Jan01. /Paul
On Tue, Jul 8, 2014 at 1:18 AM, John Gilmore <[email protected]> wrote: > We seem to have been here before, many, indeed too many times. > > There are arguments for displaying or printing local times; there may > be arguments for scheduling things manually using local times. In > the United States, alone in the world, there may perhaps be some > justification for providing hoi polloi with am|m|pm, twelve-hour-clock > values in such situations. > > There is no intellectually respectable argument for running an > operating system on anything but UTC. When STCKE values and their > derivatives are UTC-based these pseudo-problems, those of missing and > duplicate timestamp values, do not arise. > > I am not so great an admirer of Ludwig Wittgenstein as some > professional philosophers whom I respect, but he did make some points > memorably, and among them was that > > About things we know not, we ought to be silent. > > > John Gilmore, Ashland, MA 01721 - USA >
