On Sun, 2003-11-30 at 09:02, Mary Gardiner wrote: 
> which is why you shouldn't be trusting of many of the keys in your key
> ring.

Of course, this is the whole reason for key signing events.

If you show me a copy of your key [fingerprint], and a copy of some
photo identification, and assert that they key and ID are yours, then I
have a reasonable grounds to go back to my keyring and say "yes, I trust
that this key really is the digital public key of that person".

Doesn't mean you trust the person's *character*, just that you have been
reasonably convinced that the key matching that fingerprint really does
belong to that person, and so can trust that a digital signature made
with that key came from that person.

--

The process can stop there, but there is one more step. It can get a
touch complicated from here, but this is how a "web of trust" grows:

If we want, we can sign the other person's key indicating that we trust
it, and then send a copy to the other person. Each person can, if they
care to, submit these signatures to the public key servers; future
downloads of that public key will result in a key which carries along
with it these signatures from other people. 

So, ultimately, even if I don't know *you*, but your key is signed by
someone I *do* trust, then I have a reasonable assurance that you are
who you say you are.

--

I've always wondered at what point this algorithm would break down under
it's own weight - assuming mass adoption of the OpenPGP key system world
wide, when would either keyring file size, complexity in the trustdb, or
length of time needed to validate signatures, cause the whole thing to
grind to a halt?

AfC


-- 
Andrew Frederick Cowie
Operational Dynamics Consulting Pty Ltd

Australia: +61 2 9977 6866

http://www.operationaldynamics.com/
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Reply via email to