[LEAPSECS] "China move could call time on GMT"

[Warner Losh]

On Dec 30, 2011, at 10:57 AM, Rob Seaman wrote:


> Clocks worldwide have always kept time-of-day.  Astronomers are the canaries in the coal mine in this situation.  That astronomers will incur great expense and require great effort to fix what ain't broke *does not* imply that everybody else is somehow off the hook.


clocks around the world have kinda kept time-of-day.  This is an old debate.  These days, clocks are set to what the national authorities say.  The actual relation to local sun doesn't matter, so long as midnight is dark and noon is light in local time.  The "time of day" that is kept is currently just an approximation of a time at an idealized spot somewhere nearish to the location of the clock.


> It is this unprecedented proposal to move away from time-of-day that is demanding everybody - not just astronomers - go to inordinate lengths.


Dropping leap seconds will still keep the time of day approximately the same for quite some time.  This approximation may screw things up that are fine now, and fix some things that are brokenish now.  We won't know until we throw the switch, and once thrown, it may be hard to unthrow.  It won't affect most things, but it remains to be seen how the actual impact of the things it does touch...


> Also "sounds like" is not a coherent engineering plan.  No community - not even the atomic timekeepers themselves - have invested appropriate resources in investigating the implications of and mitigation required by this proposal.


I tend to agree, even though this is an outcome that would make my life easier... :)

Warner


> Rob Seaman

> National Optical Astronomy Observatory

> --

> 

> On Dec 30, 2011, at 9:30 AM, Peter Vince wrote:

> 

>> It sounds like the astronomical community have the systems in place to deal with the erratic rotation of our planet, can they not just leave the rest of us to have a simple timescale - save us all going to inordinate lengths and expense?

>> 

>> Peter Vince

> 

> _______________________________________________

> LEAPSECS mailing list

> LEAPSECS at leapsecond.com

> http://six.pairlist.net/mailman/listinfo/leapsecs

> 

>