Hi Tom,

>> No, we're used to leaping local civil time.  UTC as international
>> civil time remains an indicator of Earth orientation.  The two
>> are kept separate whatever local authorities choose to do.
> 
> I thought UT1 was the "indicator" of Earth orientation.

Currently UTC ~ UT1.  They both serve as simply "Universal Time".

> Are you confusing a timescale with a measurement?  DUT1 exists in order to 
> convert a timescale back to a measurement.

This conversation has been going on for a dozen years.  I'm not precisely sure 
what this talking point is about, but presumably we've discussed it before.

…but thanks for mentioning DUT1.  Whatever else the ITU proposal would do, it 
would magnify the role of DUT1 significantly.  Shouldn't significant issues be 
discussed in the proposal (or addenda)?

> You aim your precise telescopes using UT1 not UTC, don't you?

There's that word "precise" again as in the "precision timing community".  
Precision is only one issue - and not the most fundamental.  More fundamental 
is the data model.  TAI and UT1 are two different things, whether "timescales" 
or "measurements".  Currently UTC ~ UT1.  The ITU is proposing that UTC == TAI 
(and apparently that TAI also be "suppressed").  This has implications.

Telescopes are generally pointed using UT as in the general concept of 
Universal Time.  Other classes of astronomical software similarly use UT (or 
labeled GMT or MJD or other things).  There has never been a reason to specify 
either UT1 or UTC in general.  If the ITU proposal is passed there will be a 
*new* requirement to distinguish the two.

Whether such UT or MJD or whatever values are "precise" depends on the workflow 
to generate the value.  All of those workflows will have to be vetted.  This is 
a significant Y2K-like issue (DUT1 will exceed 0.9s), a significant 
infrastructure issue (deploying new infrastructure to retrieve DUT1 and/or 
UT1), and a algorithmic issue since UTC will no longer approximate UT1.

Rob

_______________________________________________
LEAPSECS mailing list
[email protected]
http://six.pairlist.net/mailman/listinfo/leapsecs

Reply via email to