[LEAPSECS] The ends we seek

[Mark Calabretta]

On Thu 2012/01/19 18:37:25 PDT, Warner Losh wrote
in a message to: Leap Second Discussion List <leapsecs at leapsecond.com>


>It is "discontinuous" for some definition of "discontinuous."


If UTC is discontinuous in any sense then so must the Gregorian
calendar be, with a "discontinuity" 86400 times greater on Feb/29.

Noone seems at all concerned about that "discontinuity".


>While strictly speaking time is an ordered, monotonically

>increasing full ordered set (which would make it continuous),

>it is none-the-less irregular as it has a non-uniform radix.


The problem is the unpredictability of the radix, not its
non-uniformity or even its time-variability which the calendar
already has to a greater extent.


>People that speak of it being "discontinuous" do so because

>that's easier to say and explain than non-uniform radix, even

>if it is pedantically incorrect.


Proponents of dropping leap seconds regularly describe UTC as
"discontinuous" as though it were a fundamental flaw.  Just
read the ITU press releases, with "discontinuous UTC" duly
echoed by the press for public consumption.

What's so hard about saying "the long-term unpredictability of
leap seconds creates problems".


>When arithmetically mapped using 86400s days, it becomes

>discontinuous at the leap second.


"When arithmetically mapped using 365d years, it becomes
discontinuous on Feb/29" sounds less convincing.

Here is my point.  A "discontinuous" time-scale is a timescale
that is broken beyond repair.  On the other hand, a timescale
with some element of unpredictability is something we have a
fighting chance of doing something about, whether by improving
the predictability or relaxing the tolerance or both.

I think we agree that removing some of the unpredictability,
at least in the medium term, is the way to achieve consensus.

Mark Calabretta