[LEAPSECS] Cheating means more planning, not less

[Rob Seaman]

Michael Sokolov wrote:


> But civil time *is* a low-precision timing application!


Civil time is not a timing application.  It is not an application at  
all.  Whatever the past or future of civil timescales, these form  
infrastructure that applications are built upon.  Precision is one of  
many requirements incumbent on applications.  How the necessary level  
of precision is obtained may drive choice of one or another or several  
different timekeeping technologies and standards.

M. Warner Losh wrote:


>> And if you don't solve the problem for high-precision timing  

>> applications, I'm not sure that it is a viable solution.

>

> Those need to use a diffirent time scale decoupled from civil time,  

> i.e., TAI, GPS, TT, whatever.


This is often true now, and will often be true in the future.   
Requirements driving this will tend to split into clock requirements  
versus interval timing requirements.  (Search the archive for  
"chronometer" to track down our previous discussions on similar topics.)


> There need to be two different seconds, a civil second and a  

> scientific second.  The latter would be better renamed to essen.


Well, as Warner might say, "that boat has sailed".  We would certainly  
have more clarity of understanding if the SI second had been called  
something else, and especially if it had been chosen to have no  
relation to 1/86,400 of the length of the solar day.

Maybe we can just call it "Celsius time" and make a Celsius second  
equal to 9/5 of a civil second (as of 1820).

Rob