playing with cryptography...

[Robert J. Hansen]

On Fri, 2008-05-23 at 12:54 +0530, Hardeep Singh wrote:
> Well, that may be true, but there are currently no options that are
> significantly better. The WOT model used by GPG is better? Maybe, but
> not significantly.

WoT gives you more options about how to determine trust levels.  This,
to me, is significant.

> There is nothing that can prove who you say you are. State provided ID
> cards only prove that you were able to convince the system that you
> have a specific name.

This is Philosophy 101 sort of stuff.  There is nothing that can prove
_anything_ in the world.  After all, the cosmos may have been created
just last Thursday, and all of our memories are just what we were
created with, etc., etc.

It's not about proof, either.  It's about probabilities.  We're not
looking for a 100% assurance the person involved really is who they say.
A confidence in the high nineties is practically just as good, and can
be achieved fairly easily by asking to see a few different forms of
government ID.