On 12/19/2010 18:03, Gerard Ashton wrote:
The idea that consumer grade WWVB receivers will become obsolete
supposes that legal time will be a fixed (except for daylight savings
time) offset from UTC, and UTC continues to include leap seconds. If
WWVB were to broadcast the proposed TI instead of UTC, the old
receivers would display TI-based time, which would gradually depart
from the legal UTC.
Having both would have all the problems of the current system, but one
more source of confusion. :)
I saw something on my phone while I was on vacation that implied that
the legal time would still exist, still have leap seconds and still be
knowable, even if the UTC time scale eliminated leap seconds. I have
one point to make about that, but couldn't find the original email.
If the ITU just eliminated leap seconds from UTC, and people think that
legal time would continue to have leap second, I'd maintain that legal
time would be de-facto the new ITU thing. The reason is that there'd be
no authority to promulgate when a leap second happens. If DUT1 is
evolving at 400ms/year and the offset is currently at 0ms, then you'll
have 4 changes to have a leap second and still meet DUT1 < 0.9s. Which
one is the right one? in 6 months when DUT1 is 200ms? In a year when
it is 400ms? 18 months when it is 600ms? or 2 years when it is 800ms?
There's no documented procedure today to now which one that IERS will
pick (but I'd think everybody would agree it is likely one of the middle
two). Also, this does over-simplify things a bit, since other
considerations play into this decision as well: making sure there won't
be a negative leap second, making sure it doesn't coincide with
something else that's bad (no leap second at y2k was kinda planned), a
preference for December over June, etc. Unless there's an authority to
promulgate this that everybody recognizes, the new time scale would
become the defacto legal time, most likely through a series of legal
tests as DUT1 widens.
_______________________________________________
LEAPSECS mailing list
[email protected]
http://six.pairlist.net/mailman/listinfo/leapsecs
