[LEAPSECS] Civil timekeeping before 1 January 1972

[Tom Van Baak]

> The tabulations of the times of emission of radio broadcasts of UTC
> were given in units of, and with an accuracy of 0.0001 s; i.e., 100
> microseconds.
> 
> The tabulations of the intercomparisons between the time scales in
> those laboratories are given with decimals to 0.1 microsecond, or 100
> nanoseconds.
> 
>> I don't believe 3 ns is significant for any time stamp from that
>> era.

Steve and Paul,

To further add perspective to 1960's timescales, read these wonderful papers:

Correlating Time from Europe to Asia with Flying Clocks
http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1965-04.pdf

World-Wide Time Synchronization, 1966
http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1966-08.pdf

'Flying Clock' Comparisons Extended to East Europe, Africa and Australia
http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1967-12.pdf

The articles should give you a better feel for what timekeeping, clocks, and timescales are. Does your C# code incorporate the notion of error bars? If your users are scientific, perhaps they need to know how uncertain timescale conversions are.

Thanks,
/tvb