[LEAPSECS] LEAPSECS Digest, Vol 51, Issue 7

[Warner Losh]

On 02/02/2011 09:03, Stephen Colebourne wrote:

> On 2 February 2011 15:33, Finkleman, Dave<dfinkleman at agi.com>  wrote:

>> I suggest that the terms second, minute, hour, day, and month stated

>> without qualification have "normative" status:  the SI second, 60 SI

>> seconds, 3600 SI seconds, 86,400 SI seconds, and Gregorian calendar

>> numbers of days expressed as 86,400 seconds.

> Funilly enough, I'll start by disagreeing with those definitions. The

> meaning of the things named "second" and "day" being key, the best

> that consensus will agree to is precise definitions, rather than any

> "land grab" of key terms.


The only consensus possible here is that 'seconds is an overloaded term 
for both the length of time for so many CS atom oscillations, for the 
amount of time the earth takes to rotate though 1/86400th of its 
rotation, for 1/86400th of the average period of time it takes for earth 
to rotate averaged over a year, and for 1/8400th of the average period 
of time it takes the earth to rotate with various seasonal effects 
included.'  this corresonds to TAI/UTC, UT, UT1 and UT2 respectively.

This stems from a fundamental disagreement about time.  Should it tick 
off a fixed set of time, or should it tick off based on observations?  
TAI is the latter, UT, UT1 and UT2 are different ways of account for the 
former.  UTC tries to split the baby and do both.  There is a divergent 
opinion bout which one people generally want to do because today the 
difference between them falls well below the threshold of observable 
effects on the human psyche.  Humans are able to adopt to fairly large 
difference between clock on the wall time and apparent solar time, but 
there's no agreement as to how big this divergence can be, and the best 
ways, if any, to keep things from drifting as well as being off today.

Until the fundamental disagreement is resolved, you aren't going to be 
able to define these basic terms.


> The terms SI-based-minute, SI-based-hour and SI-based-day will, I

> suspect be acceptable to all. At the end of the process, the

> defniition of the simple terms can then be considered.

>

> Statements:

> - an SI-based-minute is formed from exactly 60 SI-seconds

> - an SI-based-hour is formed from exactly 60 SI-based-minutes and thus

> exactly 3600 SI-seconds

> - an SI-based-day is formed from exactly 24 SI-based-hours and thus

> exactly 86400 SI-seconds


Even those statements aren't completely right.  If you are in a TAI 
frame of reference, they are right.

However, in a UTC frame of reference, they are all wrong.  Minutes are 
60 SI seconds, except when they are 61.  Hours are 3600 SI seconds, 
except when they are 3601.  Days are 86400 SI seconds, except when they 
are 86401.

Also, what is meant by mean solar X?  what is the averaging time?  what 
are the requirements for smoothing?


>> Anything else requires qualification such as: mean solar seconds.

>> Actually I can't think of other forms of hour, day, etc. that require

>> qualification, since those are neither fundamental time intervals nor

>> time scales.

> Defining things that are *not*  fundamental time intervals nor time

> scales is key to the problem.


The problem is very fundamental about what time should be.

Warner

> I'll resend in the main thread an updated list so far.

>

> Stephen

> _______________________________________________

> LEAPSECS mailing list

> LEAPSECS at leapsecond.com

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

>

>

>