GW170817
========

Everybody's favourite wave.

1.7 second delay between merger and light, of the order of the
distance between the earth and the moon.


It is long overdue for me to fit a model to this observation, based on
the arriving galaxy model.

144 million light years away, or 1% of the Hubble distance.

A 1.7 second delay over this distance gives a speed difference of less
than one part in 3.74e-16.

Now in the arriving galaxy model,  the distance is 100 times greater
and we get 3.7e-18 as the potential difference in the speed of light
and the speed of gravitational waves.


CMB oddity
----------

This is commensurate with the size of the Sciama effect expected for
the CMB: all the photons of the CMB induce a rotation on the
surrounding space-time.

A tiny ripple.  It is not that light travels slower, rather, it
travels further.

Gravity waves of a given frequency all travel at the speed of light,
but light itself follows all the tiny wrinkles.

If this is what is happening the 1.7 seconds should turn out to be
quite consistent across observations.

In the arriving galaxy model, we see a speeded up version of the
galaxy's history, so the light is averaged across lots of time.  As
such we should expect the peak to be very close to the galactic
centre.


::

   >>> 144e6 * u.lightyear / sp.cosmo.hubble_distance
   <Quantity 0.00996433>
   >>> 1.7 * u.s / u.year 
   <Quantity 1.7 s / yr>
   >>> _.decompose()
   <Quantity 5.38697493e-08>
   >>> 1.7 * u.s / 144e6 * u.year 
   <Quantity 1.18055556e-08 s yr>
   >>> 1.7 * u.s / (144e6 * u.year )
   <Quantity 1.18055556e-08 s / yr>
   >>> _.decompose()
   <Quantity 3.74095481e-16>
   >>> _ * (144e6 * u.lightyear / sp.cosmo.hubble_distance)
   <Quantity 3.72760915e-18>