[LEAPSECS] Leapin' on the Merry-go-round

[Rob Seaman]

Poul-Henning Kamp wrote:


> Lets see, a couple of centuries ago the Napoleonic wars where in  

> full swing,


And its our clocks and calendars that permit these to be placed in  
their proper place historically.  Tying these to the rotation of the  
Earth, rather than to some arbitrary ensemble of even interval clocks  
is...manifest destiny.

Dawn on September 7, 1812 obviously had more to do with the Battle of  
Borodino than some extension of an atomic clock that accelerates at  
1.7 ms/century relative to the natural rhythms pertaining to the battle.

I might also say that the first exercise from the graduate system  
engineering course I took last semester was an analysis of Borodino  
from the point of view of modern system engineering principles.  The  
past is with us still.


> I wonder how many of the rules and regulations laid down back then  

> we still follow to the word, without having refined, reconsidered or  

> replaced the with more modern approaches.


And your point is that the one standard that ties it all together from  
distant past to far future should drift aimlessly over the centuries?   
That the analysis of future battles should require knowledge of  
timezone usage precisely in those territories in which civil authority  
has collapsed?

Surely it is better to plan than to punt?


> Any rulemaking we may propose, that our grandchildren or later will  

> have to implement, is at best a pointless waste of time, they're  

> going to do it their own way anyway.


Again - you're making my point.  If your assertion is true, then it  
forever remains our responsibility to do the right thing right now,  
and to continue to evolve the standard under coherent design principles.

Time isn't just a frequency standard to tie computer networks together.


> All we have to do is to leave them a clear notice what the issue is,  

> then they can decide if it is a problem and how they want to fix it.



Ah yes - the Yucca Mountain Johnny gambit.

Rob