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

[Warner Losh]

On Dec 30, 2011, at 4:33 PM, Michael Sokolov wrote:


> Poul-Henning Kamp <phk at phk.freebsd.dk> wrote:

> 

>> If and when they feel the need, I'm sure they will.

>> In 2600 AD, if I remember the prognosis ?

> 

> For those countries/micronations who choose to maintain their legal

> time within 0.1 s of some good-faith form of Mean Solar Time, it will

> be much much sooner.

> 

> Of course, the legal time in those countries probably isn't of concern

> for PHK as he wouldn't be advised to set foot on their soil: countries

> that feel like I do would most likely issue arrest warrants for him.


Which countries define their legal time to 0.1s of MST?  When you start tossing in terms like "good faith" into the definition, then you can no longer say 0.1s, since the two definitions are opposed.  Either it is within 0.1s of some specifically defined time, or they use weasel words like the old US law "Mean Solar Time ... as determined by the Secretary of Commerce, or his designated deputy."

Even in the UK, where time is supposed to be GMT cannot possibly realize GMT anymore since the prime meridian no longer passes through any recognized observatory to realize GMT.  This creates a legal ambiguity where UTC is published by the government labs, but the delta between GMT and UTC is only known retrospectively (although proactive predictions can be only an order of magnitude less accurate for short time periods).  So while the law may state one thing, the facts on the ground favor something else because the two are close.  The proposed ITU changes to UTC will only muddy the waters more, not less, as the old approximation GMT == UTC will cease to exist.

And besides, who is to say that a mean solar time, averaged over hundreds of years, with a tolerance of on hour wouldn't be one possible reading of the british law?

Warner