[LEAPSECS] Ghosts of Leap-seconds past and future
seaman at noao.edu
Sun Dec 26 01:22:06 EST 2010
>> Which is why the international civil timekeeping standard should be tied to physical reality.
...but Poul-Henning Kamp said:
> 1. There is no "international civil timekeeping", civil timekeeping is a national legislative matter.
and later appeared to be arguing the exact opposite point:
> The entire point of the Meter Convention and of eliminating the leap-second hack from UTC, is that we don't need to deal with each government one by one.
Rather, the entire point of this attempt to eliminate leap-seconds is that UTC indeed provides a de facto - whether or not de jure - civil timescale internationally.
Civil timekeeping on this or any other planet is going to rely on a primary cadence stationary with respect to the synodic day ("mean solar time"). The only reason the insipid ITU proposal is even vaguely conceivable is because the cheat to be introduced by the ITU is currently a few milliseconds per day. The size of the kludge doesn't change the fact that it would introduce a secular tilt, however.
Only a lawyer would simultaneously argue that "law is everything" and "law is nothing". Law is something, all right - but as surprising as it may seem in the 21st century, reality does still trump the lawyers.
The lawyers have placed a proposal on the table. Their proposal is beyond insufficient. It's clear they want to have their cake and eat it, too - to orchestrate a complete redefinition of this fundamental international standard, but without investing the commensurate resources in writing the detailed system engineering plan that such an effort should entail. Due diligence is absurdly absent.
More information about the LEAPSECS