This is the latest in the continuing saga of the sleep(5) -- a new
modification that actually doesn't do a sleep(5) all the time - first we
check to see if the time elapsed during the INSERTs and UPDATEs equates
to 5 seconds, THEN we only add enough sleep() seconds to add up to 5
Every second saved is precious.
(Gregorio, you asked me about sleep(5) the other day and I actually
forgot we made this change recently. I remembered and thought I would
By the way, speaking of this new modification, the April SF run is
complete (has been for a week at least) but I am releasing the raw data
tonight, and hopefully some summary reports.
This is datasource_id=4
Also the migration of the Elon data (datasources 1, 2, 3) is complete.
We'll be releasing a java code library shortly.