Some questions

[David SMITH]

On Fri, Aug 08, 2008 at 03:55:13PM +0200, zulag wrote:
> OK, but what exactly are the risks, what to expect to protect from ?
> Is it because of the possible password "cracking" and the password
> becoming unmodifiable (because people have a copy of the key encrypted
> with the password "P" you had when you exported your key, so whatever
> stronger password "Q" you set later, they can attack and compromise
> your key by finding "P") ?

In a nutshell, "yes".

Once they've got a copy of your secret keyring, there's no point in
changing the passphrase on your own copy of the keyring.  They can
brute-force the passphrase encryption to get your keys.  Changing
your passphrase doesn't change the underlying keys; you have to revoke
and regenerate for that.

The encryption algorithm of the "real" keys is intended to be
unbreakable[1] because the keys are truly random, and the key space that
needs to be checked is too large to be brute-forced.

The encryption used to protect the secret keys with the passphrase is
potentially weaker, though, as

 1. The key space will tend to be smaller - people are unlikely to choose
    enormously long passphrases

 2. People are stupid, and tend to use simple passphrases which are
    vulnerable to dictionary attack

 3. The encryption algorithm itself might be weaker, although I've no
    knowledge or evidence to back this up.  Frankly, I don't even know
    what algorithm is used, so my comment is purely speculation.

[1] subject to the usual discussions of key length, algorithm strength and
speed of development of computing hardware

-- 
David Smith        | Tel: +44 (0)1454 462380    Home: +44 (0)1454 616963
STMicroelectronics | Fax: +44 (0)1454 462305  Mobile: +44 (0)7932 642724
1000 Aztec West    | TINA: 065 2380          GPG Key: 0xF13192F2
Almondsbury        | Work Email: Dave.Smith at st.com
BRISTOL, BS32 4SQ  | Home Email: David.Smith at ds-electronics.co.uk