[time-nuts] nubie querie

[David Forbes]

Hal Murray wrote:
> richard at karlquist.com said:
>> I vaguely remember reading that pulsars have some fantastic stability
>> like 1E-20.  I don't remember how they established this. 
> 
> Do you remember how long ago you read that?  It might have been some 
> handwaving back before they had good data.
> 
> For a while, the astronomers were seriously trying to take back the official 
> clock from the physicists.  They didn't make it.  I think a lot of the 
> interest started when somebody discovered a pulsar ticking at close to 1 KHz. 
>  That was in 1982.
> 
> They have collected enough data to see things like star-quakes which are 
> glitches in the rotation rate due to geologic type shifts similar to the 
> recent discussion of 1.26 microseconds per day from the Chile quake.
> 
> They decay by gravitational radiation and speed up when accretion adds more 
> momentum and energy.

That's what I remember from reading the literature on pulsars as well. They act 
a lot like quartz crystals, in fact, with frequency drift and frequency jumps.

In short, there is no macro-scale clock that will ever work as well as an atomic 
clock, since an atomic clock depends on the behavior of subatomic particles with 
a handful of state variables, whereas a macro-scale clock depends on the 
behavior of large and changing lumps of disparate matter with gazillions of 
state variables.

I realized as a teenager that the difficulty of predicting the future is 
geometrically (or so) proportional to the number of state variables. So the 
cesium clock is one of the few gadgets whose future behavior can be predicted to 
any reasonable extent.

--David Forbes