[Cryptography] Best internet crypto clock: hmmmmm...

John Gilmore gnu at toad.com
Wed Oct 22 10:12:37 EDT 2014

>> IIRC a lot has been done to verify video and audio as having come 
>> from a certain moment in time or general location based on recovering
>> the precise 'drift' of the omnipresent 60-cycle (or 50-cycle if 
>> you're Australian) hum of the surrounding electrical system.
> http://www.bbc.co.uk/news/science-environment-20629671

That's fascinating!  Turning an annoyance of audio engineers and home
stereos into a forensic tool.

But once the technique is known, it can be forged, by pasting a
recording of the "hum" from one time or place, into a recording made
or edited at another time or place.  It would be amusing to file a
FOIA request for the FBI's recordings of US power networks' hum, 
and watch them squirm trying to find a reason why you couldn't have it.

Also, in theory, even if nobody was recording the hum continuously,
the hum could be extracted from two or more existing recordings and
compared to determine whether they happened at the same time (or
copied to another recording).  For example, two concerts that were
recorded at the same time should show the same hum if they were done
within the same power grid.

Isn't there also some research showing that over time, you can tell
what time zone a remote computer is in, by pinging it for timestamps,
and noticing when its oscillators run minutely faster during the heat
of the day, and slower during the cool of the night?


More information about the cryptography mailing list