Robert J. Hansen
rjh at sixdemonbag.org
Mon Mar 1 04:31:40 CET 2010
> Understood, and I agree it makes no such statement. However, it does make a reasonably good statement that you were physically located near that person at a certain point in time, roughly what that time was, and roughly where (geographically) it happened.
This is assuming the signature is known to not be someone attempting a credibility attack, or that the signer didn't sign it by accident while intending to sign a different key, etc., etc. I agree that once those assumptions are made you can learn an awful lot, and I agree that these assumptions are usually correct. Not too many people sign keys by accident, or do credibility attacks, etc.
Maybe it's an artifact of my upbringing. I see the world as broken up into things you can prove, things you suspect, and things that might be. Signature analysis lets you know a lot of might-bes, and might be a good basis for suspicions, but without those preconditions I think it's pretty hard to prove things.
I imagine we're in agreement here. I still look forward to seeing your results. :)
More information about the Gnupg-users