-=> Maurice Kinal wrote to Tony Langdon <=-

 MK> -={ Monday, 02 May 2016, 00:35:37.070065305 +1000 }=-

 MK> Hey Tony!

 TL> I just find resolution exceeding accuracy to be "not right". My
 TL> engineering studies showing up. ;)

 MK> Normally I'd agree but in this case it is more a function of what the
 MK> inner landscaper finds.  If we decided to drop the %N part it should be
 MK> accurate to the second given modern syncing routines but then we lose
 MK> in the uniqueness department.

 MK> How long does your inner engineer calculate before nanosecond accuracy
 MK> becomes a reality and we're considered visionary?

Hmm, good question.  Maintaining nanosecond synchronisation with latency and
jitter of intervening links orders of magnitude greater is tricky!  1 nS is
equivalent to about a foot of travel at the speed of light! :)  The jitter is
the more problematic issue.  Maybe when atomic clocks are built into every
timekeeping device. ;)

I was thinking something like millisecoond accuracy.  But where do we have to
be unique?  On a single system?  Across all systems?

Nanosecond or greater precision also creates issues with synchrinisation -
where is the absolute time reference?  What about clock drift due to altitude
or speed differences? (general/special relativity)?  At the nanosecond
precision, these are all measureable and significant! :-)  GPS satellite clocks
have to run at a slightly different frequency to compensate for relativistic
effects, because time passes at a different rate up there than down here,
mostly due to Earth's gravity, which is much stronger down here at the surface,
causing time to pass more slowly here.


... Truth has nothing to fear from examination
--- MultiMail/Win32 v0.49
 * Origin: Freeway BBS - freeway.apana.org.au (3:633/410)