[LEAPSECS] Java JSR-310 Instant class: suggested changes

[Steve Allen]

On 2011 Jan 29, at 14:31, Stephen Colebourne wrote:

> Thanks for the thoughts. I can see where you are driving at, but since

> UT/UT1/UT2 are not well-defined in the far past AFAIK, then the

> definition below isn;t ideal either. I agree that something needs to

> change though.


The issue of the past is that until 1972 all practical long-interval
time measurement was based on subdivisions of the calendar day.
That's what is wanted for charging rent and accumulating interest.

At such time as all radio broadcasts of precision time signals
omit leap seconds all practical long-interval time measurement
will be based on integrations of cesium counting out SI seconds.

Right now we are in a cultural/legal/procedural limbo, or maybe
even hell, where there are national and international agreements
which demand one, or the other, or which don't even make it clear
that they recognize the need for a difference.  I don't know how
to implement an API for that.
                                                                                
--
Steve Allen               <sla at ucolick.org>              WGS-84 (GPS)
UCO/Lick Observatory      Natural Sciences II, Room 165  Lat  +36.99855
University of California  Voice: +1 831 459 3046         Lng -122.06015
Santa Cruz, CA 95064      http://www.ucolick.org/~sla/   Hgt +250 m