On Thu 2014-01-09T21:57:49 -0800, Hal Murray hath writ: > The fundamental problem with POSIX timekeeping is that it pretends that leap > seconds don't exist.
While insisting that time_t match UTC, and I think not because of the name UTC per se, and not because POSIX loves mean solar days, but rather because that happens to be the name currently associated with radio brodcast time signals. My impression is that there are too many POSIX systems which have based their time_t on the radio broadcast time signals, therefore the only option for POSIX is to recommend that time_t match that time scale, whatever it happens to mean or be called. > What would you like for an API if you were starting over and wanted > to support leap seconds? I have asserted that POSIX does not want to know astronomy, does not want to track geophysics, that POSIX really wants to count atomic days rather than mean solar days, and that they have already specified enough other pieces of the OS that they could hide the difference in the zoneinfo/tzdata/tzcode. > What would you have to change in the OS and libraries? Pretty much nothing, it's all implemented and distributed already. It has been tested and anyone who wants can perform more tests without need for special equipment. http://www.ucolick.org/~sla/leapsecs/right+gps.html That's my compromise. Every second is of equal SI duration to make the realtime systems happy, and every civil day remains just the way things are now -- based on actually observing the earth rotating. -- Steve Allen <s...@ucolick.org> WGS-84 (GPS) UCO/Lick Observatory--ISB Natural Sciences II, Room 165 Lat +36.99855 1156 High Street Voice: +1 831 459 3046 Lng -122.06015 Santa Cruz, CA 95064 http://www.ucolick.org/~sla/ Hgt +250 m _______________________________________________ LEAPSECS mailing list LEAPSECS@leapsecond.com http://six.pairlist.net/mailman/listinfo/leapsecs