Many observers have CCD cameras on their scopes and will not 
want to remove them - and thus might conclude that they cannot 
usefully contribute to this exciting event. 

For all those people, I invite you to have a look at the technique 
described at: 
http://www.users.bigpond.com/reedycrk/driftscantiming.htm 

This describes a technique of recording such events by turning the 
telescope drive off at 'the right time', opening the camera shutter 
at a known time, and letting the star drift across the field - and 
ending the exposure at a known time. You can then extract the event 
times quite accurately, by measuring the transition points along the 
star trail.  This is possible with narrow fields of view because the 
time uncertainty is less than +/-20 seconds.

The technique works best if you have a mechanical shutter, where you 
can record the noise of the shutter opening/closing against a time 
base. But an electronic shutter is acceptable provided you have some 
idea of the latency. 

Remember that if this event is well observed, the profiles of the 
components will be resolved at the 1-km accuracy level, with 
relative positions being determined to within a few 100 micro-
arcsecs (yes... micro-, not milli-). So I encourage everyone near 
the predicted path locations to join in the group activity and 
monitor this event! And remember that the uncertainty in the path 
location could be a good 100km or more - so even if you are outside 
the predicted occultation paths, you should still monitor the event.